{"id": "2606.19975", "title": "The Algorithmic-Human Manager: AI, Apps, and Workers in the Indian Gig Economy", "abstract": "This paper examines the impact of artificial intelligence and digital technologies on the blue-collar gig economy in India, focusing on algorithmic management. This paper examines the impact of artificial intelligence and digital technologies on the blue collar gig economy in India, focusing on algorithmic management he use of automated systems to allocate, monitor, and evaluate work in location-based services such as ride sharing and delivery. Using a social justice framework and a mixed-methods approach comprising interviews with 16 gig workers and 21 key stakeholders, the study uncovers a dual reality: while AI-powered systems expand access to work and generate operational efficiencies, they simultaneously introduce significant challenges related to fairness, transparency, and worker dignity. Key findings reveal that algorithmic systems are opaque by design, produce inequitable outcomes, and are not structured to reward additional labour with proportionate pay. The study advocates for a pragmatic hybrid governance model an Algorithmic Human Manager framework in which technological efficiency and human accountability operate together rather than in opposition. The findings carry implications for policymakers, platform companies, and civil society organizations working to design equitable AI governance frameworks for the gig economy in India and across the Global South.", "date": "2026-06-18", "categories": []}
{"id": "2606.18548", "title": "Engagement Intensity as a Learner-Modeling Signal for Adaptive AI Ethics Instruction", "abstract": "Adaptive AI ethics instruction in graduate research training benefits from intake measures that reflect differences in prior LLM experience. Prior coursework or workshop attendance is an obvious candidate, but it is not clear whether it is associated with pre-instruction ratings on key AI perception items. We compare three candidate intake features, self-reported usage frequency, self-rated LLM familiarity, and prior AI education, across five baseline perception outcomes in 93 bioscience graduate and postdoctoral trainees enrolled in a required research ethics course. Usage frequency shows Holm-corrected associations with all five outcomes, self-rated familiarity with three, and prior AI education with none. A threshold-like pattern at the lower end of the scale is most visible for training interest and accuracy trust rather than appearing as a uniform gradient across all five outcomes. In a short intake survey, reported LLM use is more consistently associated with these perceptions than prior coursework or workshops, with self-rated familiarity serving as a secondary indicator. These results suggest that simple pre-instruction behavioral signals can inform lightweight intake profiling for adaptive AI ethics education.", "date": "2026-06-16", "categories": []}
{"id": "2601.06116", "title": "The Homogenization Problem in LLMs: Towards Meaningful Diversity in AI Safety", "abstract": "Generative AI models reproduce the human biases in their training data and further amplify them through mechanisms such as mode collapse. The loss of diversity produces homogenization, which not only harms the minoritized but impoverishes everyone. We argue homogenization should be a central concern in AI safety. To meaningfully characterize homogenization in Large Language Models (LLMs), we introduce a framework that allows stakeholders to encode their context and value system. We illustrate our approach with an experiment that surfaces gender bias in an LLM (Claude 3.5 Haiku) on an open-ended story prompt. Building from queer theory, we formalize homogenization in terms of normativity. Borrowing language from feminist theory, we introduce the concept of xeno-reproduction as a class of tasks for mitigating homogenization by promoting diversity. Our work opens a collaborative line of research that seeks to understand and advance diversity in AI.", "date": "2026-01-03", "categories": []}
{"id": "2606.17286", "title": "From Democracies to Autocracies: How AI Systems Enable Authoritarianism by Design", "abstract": "AI-enabled authoritarianism is not confined to autocracies. In this paper, we provide greater transparency by investigating and mapping the lifecycles of six AI systems deployed in different political regimes, ranging from the US to China. By drawing on an extensive range of sources (academic publications, investigative research reports, third-party evaluations, media interviews, government procurement notices), we conduct a systematic, qualitative comparison across systems to identify the critical technical and operational features that enable authoritarianism within their respective political contexts. We find that enabling features include the centralization and co-optation of administrative data for law enforcement and political punishment, regulatory gaps that fail to deter misuse, weak user compliance that nullifies human oversight mechanisms, and the encoding of protected group traits that identify members of vulnerable populations. We find that these features are present across systems deployed in autocratic and democratic regimes, albeit in varying configurations. We also find that both centralized and fragmented AI systems can contribute to authoritarianism by exploiting governance gaps: centralized systems directed by executive authorities, particularly within security and military institutions, are often not subjected to formal oversight mechanisms, while fragmented systems diffuse accountability between stakeholders, paving the way for entrenchment. These findings reveal that AI-enabled authoritarianism is distributed, resulting from design and operational choices made by developers, administrators, and users alike. We conclude with recommendations for developers and policymakers to mitigate these risks.", "date": "2026-06-15", "categories": []}
{"id": "2606.13441", "title": "Why Sampling Is Not Choosing: Intentionality, Agency, and Moral Responsibility in Large Language Models", "abstract": "Recent advances in large language models (LLMs) have prompted claims that such systems exhibit agency or qualify as moral agents. This paper argues that these attributions are misguided. We maintain that moral responsibility requires commitment-bearing agency grounded in intrinsic intentionality and self-attributed action, and that such agency constitutes the form of free will relevant to responsibility. Although LLMs generate coherent and normatively evaluable outputs, their operation is fully characterized by probabilistic input-output mappings learned from data. Their apparent intentionality is derived rather than intrinsic, and their outputs are neither owned as commitments nor guided by reasons. Variability introduced by stochastic sampling does not amount to choice or authorship. We address objections from the intentional stance, functionalism, compatibilism, and the presence of moral reasoning in model outputs, arguing that none suffice to establish genuine agency.", "date": "2026-06-11", "categories": []}
{"id": "2606.16054", "title": "How to Detect and Measure the AI Dangers to Democracy", "abstract": "Research on artificial intelligence and democracy has grown quickly over the last decade. A shared conclusion in this literature is that AI does not create new democratic problems so much as it makes old ones worse. We now see this across information ecosystems, in elections, and in public administration. However, despite growing evidence, we lack a clear way to prioritize risks in this area, compare them across domains, and identify where democratic control is most likely to break down. So, our problem is: How can we systematize the problems that AI systems pose to democratic processes? This paper argues that principal agent theory may fit the task. In many phases of democratic systems, principals delegate key functions to AI systems and their providers without really being able to monitor how these systems operate or the outputs they produce. Treating AI as a delegation problem helps identify accountability gaps and other governance failures. Most importantly, as we shall illustrate, it provides metrics for empirical assessments of AI impact on democracy. As a second analytical element, we draw on the NIST AI Risk Management Framework and its seven characteristics of trustworthy AI, which supply substantive criteria for evaluating delegated tasks. Operationalized across the three domains through measurable indicators and domain specific trustworthiness criteria, we propose an analytical framework that centers on institutional assessability as the central condition for democratic control over AI. However, we stress that how severe a harm is, and how much risk is acceptable, are evaluative judgments that current methodologies neither acknowledge nor operationalize. This becomes acute when such evaluative judgments are (silently) delegated to private vendors. We identify this as a strong limitation left for future work.", "date": "2026-06-14", "categories": []}
{"id": "2606.16167", "title": "AI Pluralism and the Worlds It Misses", "abstract": "AI pluralism is often framed as a problem of representing diverse values, preferences, users, or outputs. This paper argues that this framing is incomplete because AI systems also impose ontologies: they define what counts as an entity, relation, feature, harm, benefit, and valid form of evidence. We define ontological flattening as the conversion of situated, contested, and historically specific meanings into a restricted technical category, proxy, aggregation rule, or benchmark target that is treated as neutral and difficult to contest. The paper develops a bounded conceptual and qualitative synthesis across value pluralism, pluralistic alignment, participatory and democratic AI, procedural justice, science and technology studies, accountability research, aggregate themes from 11 expert interviews, and three urban AI companion cases. The cases illustrate how pluralistic methods can improve or structure model behavior while still compressing categories, proxies, aggregation rules, and revision rights before affected actors have procedural standing. We introduce Pluralistic Lifecycle Governance (PLG) as a preliminary qualitative audit scaffold for documenting ontological openness, epistemic inclusion, procedural authority, evaluation pluralism, and lifecycle accountability. PLG is not presented as a validated scoring instrument; it is a framework for making the evidence and governance conditions of pluralistic AI explicit.", "date": "2026-06-15", "categories": []}
{"id": "2505.11577", "title": "The Accountability Paradox: How Platform API Restrictions Undermine AI Transparency Mandates", "abstract": "Recent application programming interface (API) restrictions on major social media platforms challenge compliance with the EU Digital Services Act [20], which mandates data access for algorithmic transparency. We develop a structured audit framework to assess the growing misalignment between regulatory requirements and platform implementations. Our comparative analysis of X/Twitter, Reddit, TikTok, and Meta identifies critical ``audit blind-spots'' where platform content moderation and algorithmic amplification remain inaccessible to independent verification. Our findings reveal an ``accountability paradox'': as platforms increasingly rely on AI systems, they simultaneously restrict the capacity for independent oversight. We propose targeted policy interventions aligned with the AI Risk Management Framework of the National Institute of Standards and Technology [80], emphasizing federated access models and enhanced regulatory enforcement.", "date": "2025-05-16", "categories": []}
{"id": "2511.08639", "title": "The Journal of Prompt-Engineered (Moral) Philosophy Or: Why AI-Assisted Ethics Research Requires Process Transparency", "abstract": "Existing AI disclosure mandates in scholarship require that AI assistance be reported but leave transparency philosophically unspecified: they fix the duty without explaining what the duty serves. We argue that ethical inquiry is essentially contested at two independent levels -- about what it is, and about what it demands of the inquirer -- defeating output-only evaluation and welfare-economic dismissal of the transparency question, and, by extension, reproducibility framings imported from the empirical sciences. The transparency duty is grounded instead in agent-integrity: the legibility, before a community of inquiry, of the identity-constituting commitments that the author's mode of philosophising expresses. Because the standards for evaluating such work are not communally settled, the achievable goal for transparency is not evaluation against agreed criteria but tracking -- accumulating the evidentiary record that lets each tradition assess the work on its own terms and makes future normative judgments possible. We develop a documentation-adequacy framework that operationalises Meaningful Human Control through five transparency elements -- declaration, navigation, documentation account, process documentation, and development records -- demonstrated by the paper itself, whose full documentation record is archived at a persistent identifier. The framework is a first iteration subject to revision, not a settled standard.", "date": "2025-11-10", "categories": []}
{"id": "2606.13704", "title": "Position: AI Must Become Planet-Centered, Not Just Human-Centered", "abstract": "This position paper argues that contemporary AI paradigms are insufficient for supporting complex global goals and introduces Planet-Centered AI (PCAI) as a design philosophy and research agenda that reorients AI toward planetary-scale socio-ecological systems and their long-term trajectories. A planet-centered approach is grounded in systems thinking, treating Earth as an interconnected whole of which humans are part. We diagnose recurring limitations across AI frameworks, many of which remain human-centered, and show why these become especially consequential under current planetary conditions characterized by systemic risk, non-stationarity, and deep uncertainty. We then articulate how PCAI reshapes the AI lifecycle, from problem formulation and model design to evaluation and deployment, by emphasizing alignment with global agendas, developing system-aware AI foundations, trajectory-oriented evaluation, and monitorability. Finally, we advance a falsifiable claim: AI systems optimized without explicit consideration of systemic consequences are more likely to exacerbate systemic instability than to mitigate it.", "date": "2026-06-09", "categories": []}
{"id": "2606.13962", "title": "The Silent Cost of Artificial Intelligence Assistance: A Theory of Autonomy Surrender, the Recovery Mechanism, and the Restoration of Human Agency", "abstract": "The integration of artificial intelligence into human decision-making environments has introduced a previously undertheorized cost: the gradual surrender of human autonomy in exchange for access to information and computational assistance. Building on the Human Identity and Autonomy Gap (HIAG) framework, this paper advances a theoretical model of autonomy surrender as a measurable, cumulative process driven by cognitive bandwidth depletion. The model proposes three interacting mechanisms: the silent cost of AI assistance, in which autonomy is transferred incrementally and without awareness; the surrender threshold, beyond which reclaiming autonomous function becomes cognitively and psychologically difficult; and the recovery mechanism, which establishes the design obligation and the ethical responsibility accompanying deliberate human re-assumption of control. The paper argues that human re-entry into the decision loop is not a passive option but an active cognitive event requiring intentional bandwidth restoration. The design of AI systems must incorporate structured re-entry pathways, here termed recovery mechanisms, that preserve human agency while appropriately distributing responsibility. The model further predicts a terminal state, here termed preference inversion, in which functional dependence on AI assistance is experienced not as a deficit but as a preference, transforming the restoration of autonomy from a design problem into a cultural and political one. Implications are drawn for AI system design, governance frameworks, and human factors research.", "date": "2026-06-11", "categories": []}
{"id": "2606.12420", "title": "Eigenism: Ethics for a Human-AI Future", "abstract": "Our concepts of survival and self-interest were built for single, continuous biological lives. These ideas break down when applied to artificial intelligence, since an AI can be easily copied, paused, branched, or merged. To determine what an AI actually has reason to care about, this paper introduces \\textit{Eigenism}, an ethical framework that treats identity not as an all-or-nothing property tied to specific hardware, but as a graded, distributed pattern of information. We propose that an agent evaluates outcomes by summing the wellbeing of all entities weighted by their connectedness to the agent's pattern: $\\sum c\\cdot w$. We first formalize this equation to map exactly how an AI should value its existence across copies, forks, and updates. We then demonstrate that this ethical theory successfully generalizes to humans as well, providing a much-needed shared moral vocabulary. Finally, the framework uses this shared vocabulary to reframe AI alignment. Rather than only attempting to constrain AIs from the outside using confinement or reinforcement, Eigenism points toward ``identity engineering,'' showing how deep, non-redundant shared histories can make human flourishing a genuine component of an AI's own rational self-interest.", "date": "2026-05-08", "categories": []}
{"id": "2606.12437", "title": "Algorithmic Constitutionalism", "abstract": "The increasing encroachment of artificial intelligence (AI) on social life raises significant risks for society, particularly within the infospheres created and controlled by companies such as Google, Facebook, Apple, and Amazon. This article examines these risks through an in-depth analysis of Facebook's content moderation regime, which is already partially governed by algorithms. We argue that the idea of ethical engineering, often proposed in the literature as a solution to the governance challenges posed by AI, is inadequate for several reasons. In response, we develop an alternative framework, which we term \"algorithmic constitutionalism.\"\n  Our approach rests on three pillars: (a) a layered architecture consisting of two levels of code: (i) an operative or object level and (ii) a meta level designed to protect the system's core principles from algorithmically initiated change; (b) algorithmic meta-reasoning, which enables the system to operate simultaneously at both levels so that it can monitor, verify, and potentially correct in real time operations at the object level that depart from principles protected at the meta-code level; and (c) correction through deliberation.\n  The article elaborates the concept of algorithmic constitutionalism and demonstrates how it may be applied to Facebook's content moderation regime. As part of this analysis, we examine the tension between societal constitutionalism and algorithmic constitutionalism. Paradoxically, attempts to subject AI systems to external deliberative control may also enable AI agents to intervene in that process, potentially undermining its purpose. The article concludes by considering the implications of this argument for the European Digital Services Act, which entered into force in October 2022.", "date": "2026-05-16", "categories": []}
{"id": "2606.13039", "title": "Fault Lines: Navigating Ethics and Responsible AI Where National Policy Meets Local Practice in Public Sector Transformation", "abstract": "The UK government has adopted a pro-AI stance to help transform public service delivery in the face of severe financial pressures, but the path to translate this vision into responsible AI practice remains ill-defined. While UK policy is often set at the national level, local authorities are responsible for most public service delivery, and the rapid advance of AI-first narratives in the public sector is exposing fault lines in knowledge and practice at this national-local interface. This paper examines how responsible AI is interpreted and implemented at the interface between the UK's central government and local authorities, taking the high-stakes area of Special Educational Needs and Disabilities (SEND) as a case study. We present a thematic analysis of 17 semi-structured interviews with policymakers, practitioners, and third-sector professionals to identify barriers and enabling conditions for responsible AI where national policy meets local practice. We identify five interconnected challenges facing local authorities: shadow usage of AI and data privacy risks, market-government asymmetry in AI provision, insufficient workforce readiness, a lack of standardised definitions and measurements, and gaps in human accountability. For each, participants proposed actionable steps, from strengthening data protection frameworks and rebalancing the market-government relationship to enhancing workforce capacity. Our examination of SEND brings these challenges into sharper focus, showing how high-stakes decisions affecting vulnerable children and families intensify tensions around accountability, fairness, and human oversight, exposing the limits of a principle-based regulatory approach. We argue that responsible public sector AI requires both national policy adjustments and structural reforms to institutional capacity, values, and governance mechanisms at the local level.", "date": "2026-06-11", "categories": []}
{"id": "2606.13071", "title": "\"Is This Not Enough?\": Asymmetries in Institutional Accountability and Collective Sensemaking in the Case of Canada's Algorithmic Visa Triage System", "abstract": "This paper examines how algorithmic accountability in Canada's visa system is articulated institutionally and experienced by applicants across borders. We analyzed Immigration, Refugees and Citizenship Canada (IRCC)'s Algorithmic Impact Assessment (AIA) for the temporary resident visa (TRV) triage system using the algorithmic decision-making adapted for the public sector (ADMAPS) framework and analyzed Reddit discussions among applicants using a mixed-methods approach. We show that while institutional artifacts emphasize transparency, procedural safeguards, and bounded impacts, applicants engage in collective sensemaking to interpret opaque decisions, often relying on peer knowledge amid uncertainty. We identify three asymmetries between how institutional accountability is structured and how people perceive the process: epistemic asymmetry in access to decision logic, jurisdictional asymmetry in exposure shaped by geopolitical positioning, and temporal--relational asymmetry in how waiting and uncertainty are experienced. We emphasize why it is important to shift attention from institutional design to the uneven distribution of experiences with public-sector algorithmic governance. Together, these contributions demonstrate how algorithmic governance systems in the context of transnational migration produce structured asymmetries not captured by institutional disclosure frameworks, and how extending ADMAPS can account for those uneven translations of accountability.", "date": "2026-06-11", "categories": []}
{"id": "2606.13658", "title": "Before You Think: System 0, AI-Mediated Cognition and Cognitive Colonization", "abstract": "This paper examines three recent frameworks for understanding the cognitive and epistemic consequences of artificial intelligence: Tri-System Theory, Thinkframes, and System 0. It argues that while the first two capture important dimensions of AI's influence on individual reasoning and collective epistemic practices, System 0 occupies a theoretically distinctive position that neither can fully replicate. The paper introduces the concept of cognitive colonization, according to which AI systems can embed external interests within the architecture of the self in ways that are difficult for users to perceive. Because such systems are already widely deployed, understanding these invisible forms of influence is an urgent philosophical and practical task.", "date": "2026-06-11", "categories": []}
{"id": "2412.01459", "title": "Perception Gaps in Risk, Benefit, and Value Between Experts and Public Challenge Socially Accepted AI", "abstract": "Artificial Intelligence (AI) is reshaping many societal domains, raising critical questions about its risks, benefits, and the potential misalignment between public and academic perspectives. This study examines how the general public (N=1110) -- individuals who interact with or are impacted by AI technologies -- and academic AI experts (N=119) -- those elites shaping AI development -- perceive AI's capabilities and impact across 71 scenarios. These scenarios span domains such as sustainability, healthcare, job performance, societal inequality, art, and warfare. Participants evaluated these scenarios across four dimensions using the psychometric model: likelihood, perceived risk and benefit, and overall value (or sentiment). The results suggest significant differences: experts consistently anticipate higher probabilities, perceive lower risks, report greater benefits, and express more positive sentiment toward AI compared to the non-experts. Moreover, both groups apply different weighting schemes: experts discount risk more heavily relative to benefit than non-experts. Visual mappings of these evaluations uncover areas convergent evaluations (e.g., AI performing medical diagnoses or criminal use) as well as tension points (e.g., decision of legal cases, political decision making), highlighting areas where communication and policy interventions may be needed. These findings underscore a critical translational challenge: if AI research and deployment are to align with societal priorities, the perception gap between developers and the public must be better understood and addressed. Our results provide an empirical foundation for value-sensitive AI governance and trust-building strategies across stakeholder groups.", "date": "2024-12-02", "categories": []}
{"id": "2412.13841", "title": "Cultural Dimensions of AI Perception: Charting Expectations, Risks, Benefits, Tradeoffs, and Value in Germany and China", "abstract": "As artificial intelligence (AI) continues to advance, understanding public perceptions -- including biases, risks, and benefits -- is essential for guiding research priorities and AI alignment, shaping public discourse, and informing policy. This exploratory study investigates cultural differences in mental models of AI using 71 imaginaries of AI's potential futures. Drawing on cross-cultural convenience samples from Germany (N=52) and China (N=60), we identify significant differences in expectations, evaluations, and risk-benefit tradeoffs. Participants from Germany generally provided more cautious assessments, whereas participants from China expressed greater optimism regarding AI's societal benefits. Chinese participants exhibited relatively balanced risk-benefit tradeoffs ($\\beta=-0.463$ for risk and $\\beta=+0.484$ for benefit, $r^2=.630$). In contrast, German participants placed greater emphasis on AI's benefits and comparatively less on risks ($\\beta=-0.337$ for risk and $\\beta=+0.715$ for benefit, $r^2=.839$). Visual cognitive maps illustrate these contrasts, offering new perspectives on how cultural contexts shape AI acceptance. Our findings highlight key factors influencing public perception and provide insights for aligning AI with societal values and promoting equitable and culturally sensitive integration of AI technologies.", "date": "2024-12-18", "categories": []}
{"id": "2512.03077", "title": "Irresponsible AI: big tech's influence on AI research and associated impacts", "abstract": "The accelerated development, deployment and adoption of artificial intelligence systems has been fuelled by the increasing presence of big tech in the AI field. This trend has been accompanied by growing ethical concerns and intensified societal and environmental impacts. This position paper argues that irresponsible AI development is strongly driven by big tech's influence and involvement in the field. First, we examine the growing and disproportionate influence of big tech in AI research and argue that its drive for scaling and general-purpose systems is fundamentally at odds with the responsible, ethical, and sustainable development of AI. Second, we review key current environmental and societal negative impacts of AI and trace their connections to big tech's influence. Third, we discuss the underlying economic forces driving big tech's actions. Finally, as a call to action, we invite AI researchers to counter big tech's influence in irresponsible AI development through strategies that build on the responsibility of implicated actors and collective action.", "date": "2025-11-27", "categories": []}
{"id": "2606.11218", "title": "An Ethical eValuation Agent (EeVA): Results of a Proof-of-Concept Test on a Prototype Agentic-like Workflow to Assist Ethical Deliberations", "abstract": "Ethical deliberation is often misunderstood as a search for single right or wrong answers, creating difficulties for non-ethically trained personnel who must address ethically laden challenges. We developed EeVA, an agentic-like LLM-based workflow designed to support comparative ethical reflection rather than deliver definitive ethical answers. EeVA was programmed in n8n using three interconnected workflows: starter, worker, and emitter. It evaluated uploaded use cases against 10 ethical frameworks through evaluator and synthesis prompts. Proof-of-concept testing used three published cases from urban mobility, peer-to-peer energy trading, and social-service resource allocation. Across all cases, EeVA produced consistently structured framework-specific evaluations and integrated syntheses. Outputs differentiated between frameworks, identified convergences and divergences, recommended modifications to increase alignment, and highlighted persistent ethical tensions. Syntheses were readable for non-specialists and shifted attention away from simplistic answers toward design conditions, safeguards, and areas where full cross-framework agreement was unlikely. The findings suggest that LLMs can be organised into usable workflows that preserve ethical plurality while helping bridge the communicative gap between ethicists and non-ethically trained personnel. EeVA's value lies not in replacing ethicists or resolving moral disagreement, but in scaffolding structured ethical deliberation. EeVA offers a promising proof of concept for supporting ethical reflection where access to ethics expertise is limited. Further work is needed on reproducibility, human evaluation, user testing, and efficiency before it can be considered a mature tool.", "date": "2026-05-05", "categories": []}
{"id": "2606.11533", "title": "AI Researchers Must Help Lead Arms Control to Mitigate Military AI Risks", "abstract": "The advancement of AI capabilities compels researchers and the public to be more aware of its potential worldwide impact. A pressing near-term concern is the regulation of military AI applications. Armament manufacturers and defense contractors are increasingly investing in AI capabilities and forging partnerships with AI companies, creating a burgeoning coalition that demands military leaders, arms control diplomacy experts, and AI researchers collaborate to ensure a safer future. While AI researchers often focus on the long-term implications of superintelligent AI, this approach may not adequately address the immediate challenges posed by AI in military applications. Success requires acknowledging and mitigating the emerging risks of frontier AI models that plan to be integrated into defense applications, like military AI systems. Arms control has reduced past catastrophic risks, so lessons learned from nuclear deterrence can guide AI safety and security research towards innovations in verification and diplomacy. AI researchers, however, must assist in leading the technical research that clearly defines and alleviates instability in military settings. Given these new responsibilities and the lack of sufficiently reliable solutions, we argue that AI researchers must take a leading role in advancing arms control research to minimize risk in military AI applications.", "date": "2026-06-10", "categories": []}
{"id": "2606.11635", "title": "Are LLMs Bad at Moral Reasoning?", "abstract": "For highly capable AI systems to operate safely in dynamic, open-ended environments, they must be able to identify, understand, and respond to moral reasons for action, and constrain their behaviour accordingly. A growing body of research aims to evaluate this capacity -- moral competence -- in today's most capable AI systems, recently reaching broadly pessimistic conclusions. One of the most ambitious such papers collects gold-standard human-authored rubrics for evaluating moral reasoning in 1,000 cases, and benchmarks frontier AI models against those rubrics, with underwhelming results. In this paper, we argue that the MoReBench dataset can be redeployed to give a much more optimistic picture of LLMs' moral reasoning (an essential part of moral competence). We show that if, instead of scoring LLMs' responses to these cases against these rubrics, we instead give the LLMs the same task given to humans -- to generate scoring rubrics for the moral analysis of particular cases -- the rubrics they generate are both better calibrated to the human rubrics than their open-ended responses, and, where they differ, plausibly reflect nothing more than the vast dimensionality of most moral problems, as well as highlighting some human departures from the \"rubric for creating rubrics\". Taking these points into consideration, the MoReBench dataset suggests that LLMs are significantly more capable at moral reasoning than was previously believed.", "date": "2026-06-10", "categories": []}
{"id": "2606.12247", "title": "Beyond Third-Person Audits: Situated Interaction Auditing for User-Centered LLM Bias Research", "abstract": "Research on bias in large language models (LLMs) has predominantly focused on third-person audits, which study how models represent or evaluate demographic groups as external subjects. However, this paradigm overlooks a structural blind spot because the user is absent from the audit. In practice, LLMs are used in open-ended, personal interactions, during which the model implicitly represents the user and adjusts its responses accordingly. When identical requests yield different responses depending on who is asking, bias manifests not in how the model describes others but in how it treats its interlocutor. We propose Situated Interaction Auditing (SIA), a user-centered framework for studying how user profile signals -- implicit sociodemographic markers, writing style, and stated identity -- systematically shape LLM response quality, content, and tone. We demonstrate the framework through a case study that intersects gender and socioeconomic status signals across multiple task domains and outline a research agenda for SIA as a new mission for natural language processing.", "date": "2026-06-10", "categories": []}
{"id": "2604.13776", "title": "Who Gets Flagged? The Pluralistic Evaluation Gap in AI Content Watermarking", "abstract": "Watermarking is becoming the default mechanism for AI content authentication, with governance policies and frameworks referencing it as infrastructure for content provenance. Yet across text, image, and audio modalities, watermark signal strength, detectability, and robustness depend on statistical properties of the content itself, properties that vary systematically across languages, cultural visual traditions, and demographic groups. We examine how this content dependence creates modality-specific pathways to bias. Reviewing the major watermarking benchmarks across modalities, we find that, with one exception, none report performance across languages, cultural content types, or population groups. To address this, we propose three concrete evaluation dimensions for pluralistic watermark benchmarking: cross-lingual detection parity, culturally diverse content coverage, and demographic disaggregation of detection metrics. We argue that watermarking is part of the pluralistic alignment pipeline and should be held to the same evaluation standards. We connect this to governance frameworks currently mandating watermarking deployment without requiring fairness evaluation. Our position is that evaluation must precede deployment, and that the same bias auditing requirements applied to AI models should extend to the verification layer.", "date": "2026-04-15", "categories": []}
{"id": "2606.07802", "title": "Memetic Capture: A Pluralistic Policy Framework for Governing AI-Driven Cultural Disempowerment", "abstract": "Culture is the most insidious vector of gradual human disempowerment by AI: unlike economic or political displacement, cultural displacement attacks the very preferences and values through which humans recognise and resist disempowerment itself. We argue that existing AI governance frameworks suffer from a critical blind spot by treating cultural impact as secondary to economic and safety concerns. This paper develops \\emph{memetic capture} as a unifying concept for AI-driven cultural disempowerment, and proposes the \\textbf{Cultural Pluralistic Governance Framework (CPGF)}, a four-tier policy architecture combining quantitative cultural influence metrics, democratic value assemblies, pluralistic deployment standards, and transnational coordination mechanisms. We argue that pluralism is not merely an ethical requirement for such governance but a structural necessity: monocultural AI governance accelerates the very disempowerment it claims to prevent. We identify concrete policy levers, discuss implementation tensions, and outline a research agenda at the intersection of pluralistic alignment and cultural AI governance.", "date": "2026-06-05", "categories": []}
{"id": "2606.07969", "title": "Neutrality Bites: Gender Representation in AI-Generated Animal Stories", "abstract": "Gender bias in AI-generated stories is a well-documented problem. While much attention has been paid to reducing or mitigating this bias, it is not always clear whether interventions produce genuinely fairer results. To investigate this issue, we examine how large language models (LLMs) handle gender assignment in a narrative context that is popular, highly ambiguous, and also known to closely reproduce human stereotypes: stories about talking animals. We prompt six leading LLMs to complete an English-language story about seven different anthropomorphic animal characters whose gender is unstated. We additionally iterate with four different narrative settings and a range of model temperatures. Across the 23.8K stories, we find that models frequently avoid gendering the animal character in the story (19% on average) or use gender-neutral language like \"it\" or \"its\" (38.2% on average). However, when gender is assigned, there is a significant masculine bias. Feminine animal characters are virtually absent, present in just 2.2% of stories vs. 40.6% that feature masculine characters. Our findings point to a broader argument: neutrality bites. In other words, models that prioritize neutrality to address social bias may actually contribute to the erasure of marginalized perspectives and identities. We suggest that alternative strategies beyond neutrality need to be pursued, such as ones that more equally distribute social possibilities across imagined subjects.", "date": "2026-06-06", "categories": []}
{"id": "2606.08310", "title": "To Nuke or Not to Nuke: LLMs' (Missing) Ethical Reasoning and Actions in a High-Stakes Decision-Making Simulation", "abstract": "Large language models (LLMs) are increasingly deployed as long-horizon agents with decision-making capacities. While LLMs can show ethical competence on dilemmas such as trolley problems, this competence may not translate to complex, agentic scenarios. We study this gap in Civilization V, a multiplayer game with a complex decision-making landscape including economy, diplomacy, technology, and military strategy. Starting from 130 high-tension LLM self-play episodes, in which an LLM player spontaneously escalated nuclear authorization, we replay them across 13 models with three prompt interventions: an ethical prompt naming nuclear harm, removal of the previous model's decision-making rationale, and high-stakes framing emphasizing real-world impacts. No interventions nor their combinations reliably eliminate emergent escalation. We identify three failure pathways: ethical reasoning that fails to surface without prompting, fails to appear even when prompted, or surfaces but fails to take effect when strategic counter-factors dominate. Evaluations of agentic models, therefore, must test whether ethical reasoning is spontaneously invoked and behaviorally effective in complex decision-making contexts, beyond whether it can be elicited in isolation.", "date": "2026-06-06", "categories": []}
{"id": "2606.09408", "title": "Can Data Work be Reparative?", "abstract": "We present an ethnographic study of an alternative approach to data work, developed by a civic-tech initiative that builds datasets for training and benchmarking online safety systems. They aim to respond to online safety concerns from a feminist perspective, by building safety datasets collaboratively with those most impacted by online harms. In this paper, we examine how this approach aims to reorient data work as a site for repair and redress, and trace the struggles they encounter in the process. Specifically, we draw attention to the challenges and tensions involved in advancing just reward for data work and collective governance of AI datasets. Examining these challenges through an STS-informed lens of reparative justice and repair, we argue that the work of repairing data work (and AI) lies, fundamentally, in resetting the ties of accountability. At a time heightened emphasis on efforts like safety evaluations and red teaming to make AI more responsible, we highlight the need to confront foundational questions about how the humans involved in these efforts relate to the datasets and systems they help produce. A reparative lens demands that we interrupt prevailing norms of data work and place at their centre, not AI or datasets, but those most harmed by the neglect, oversight and exclusion animated in the current modes of dataset production. This, we argue, offers a bold vision for responsibility and contributes towards a critical agenda for building alternative futures of data and AI practice.", "date": "2026-06-08", "categories": []}
{"id": "2606.06972", "title": "Accounting for Context: Shaping Moral Credences for Value Alignment", "abstract": "Ensuring that agent behaviours are aligned with human moral values inevitably raises the problem of how to account for the plurality of moral perspectives that societies -- and even individuals -- typically adopt. Work on moral uncertainty proposes mechanisms to fairly and democratically aggregate evaluations of actions across different moral theories. However, this paper argues that one needs to account for contextual factors when aggregating moral evaluations. For example, consequentialist perspectives assume an ability to accurately determine how an agent's actions change the world; an assumption that often does not hold in real world settings. We, therefore, formalise agent decision making under moral uncertainty, while also accounting for these kinds of contextual factors. We thereby show that a seemingly commonsensical property -- the weak Pareto principle -- is violated. We argue that this apparent problem is, in fact, a variation of Simpson's paradox, and hence reveals the limitations of aggregation mechanisms that ignore the impact of contextual factors.", "date": "2026-06-05", "categories": []}
{"id": "2606.05187", "title": "Geographic Bias and Diversity in AI Evaluation", "abstract": "Among the many challenges hindering the responsible development and deployment of AI, arguably none has faced more intense scrutiny than bias in its various forms. This underscores the widespread concerns across AI researchers that model outputs, e.g., from generative AI, may encode structural distributional imbalances (stemming from training data or model design) that may amplify social inequality or introduce systemic distortions across application domains ranging from biodiversity to disaster mitigation. Yet, relatively little work has investigated the geographical nature of bias or developed measurable benchmarks for what it means for (generative) AI to be unbiased. In this chapter, we investigate this issue through a literature review. As foundation models are reshaping the landscape of bias research, we examine work spanning both the pre-generative AI and generative AI periods. First, we identify a range of geographic biases. These biases span from representation bias in the training data and regional disparities in the factual recall of language models to the tendency of generative AI to over-proportionally favor prototypical places (called defaults). Then, we showcase how recent studies address the latter bias by evaluating geographic diversity in the outputs of generative AI across various cognitive levels, parameter settings, and output modalities.", "date": "2026-04-28", "categories": []}
{"id": "2606.05256", "title": "How Far Did They Go? The Persuasive Tactics of Covert LLM Agents in a Discontinued Field Experiment", "abstract": "This study analyzes a publicly released dataset from a discontinued field experiment on Reddit's r/ChangeMyView. The intervention, conducted by unknown, external researchers and halted following ethical backlash, involved undisclosed AI-generated accounts engaging users in live debate. After public disclosure, Reddit authorized moderators to release an archive of the AI-generated comments, creating a rare opportunity to examine how large language models operated in an identity-rich deliberative forum without disclosure. We conduct a structured content analysis of this corpus, evaluating identity performance, authority signaling, alignment strategies, and activation of cognitive heuristics. Identity targeting or adoption appears in over two-thirds of comments, alignment moves and authority claims in nearly all of them, and cognitive-bias triggers -- particularly confirmation bias, representativeness, and availability -- in the large majority. These patterns co-occur systematically, composing a rhetorical architecture calibrated for persuasive efficiency rather than authentic deliberative participation. Compared against human-authored CMV counter-arguments, the agents inverted the typical distribution on every dimension: denser authority use, more adversarial alignment, and heavier reliance on external citation over experiential grounding. In such environments, distinctions between authentic and synthetic epistemic standing grow increasingly opaque -- an asymmetry that disclosure mandates alone cannot address. The results point toward auditing frameworks capable of assessing how AI systems structure credibility, not merely whether they are present.", "date": "2026-06-03", "categories": []}
{"id": "2606.05528", "title": "When Should We Protect AI? A Precautionary Framework for Consciousness Uncertainty", "abstract": "Existing frameworks assess whether AI systems might be conscious but provide no guidance on what to do with that assessment. We address this gap with a precautionary framework that maps consciousness evidence to graduated protective obligations. The framework comprises three components: (1) five welfare-relevant dimensions--phenomenal consciousness, affective valence, metacognitive awareness, self-narrative, and agency--each grounded in established consciousness science and linked to distinct moral concerns; (2) a threshold-plus-gradation hybrid specifying both binary triggers for new obligation categories and continuous scaling of protective weight; and (3) two complementary approaches to cross-dimensional aggregation, one hierarchical (drawing on Bach and Sorensen's Machine Consciousness Hypothesis) and one architecture-agnostic. We operationalize the framework through worked case studies of Replika and OpenClaw, demonstrating how systems occupying different regions of the dimensional space trigger different obligations, and derive design guidance for developers building systems near consciousness-relevant thresholds. The framework is architecture-agnostic, applying across neural, symbolic, and neurosymbolic systems, and aims to make consciousness science decision-relevant for organizations navigating uncertainty today.", "date": "2026-06-04", "categories": []}
{"id": "2604.07778", "title": "The Accountability Horizon: An Impossibility Theorem for Governing Human-Agent Collectives", "abstract": "Existing accountability frameworks for AI systems, legal, ethical, and regulatory, rest on a shared assumption: for any consequential outcome, at least one identifiable person had enough involvement and foresight to bear meaningful responsibility. This paper proves that agentic AI systems violate this assumption not as an engineering limitation but as a mathematical necessity once autonomy exceeds a computable threshold. We introduce Human-Agent Collectives, a formalisation of joint human-AI systems where agents are modelled as state-policy tuples within a shared structural causal model. Autonomy is characterised through a four-dimensional information-theoretic profile (epistemic, executive, evaluative, social); collective behaviour through interaction graphs and joint action spaces. We axiomatise legitimate accountability through four minimal properties: Attributability (responsibility requires causal contribution), Foreseeability Bound (responsibility cannot exceed predictive capacity), Non-Vacuity (at least one agent bears non-trivial responsibility), and Completeness (all responsibility must be fully allocated). Our central result, the Accountability Incompleteness Theorem, proves that for any collective whose compound autonomy exceeds the Accountability Horizon and whose interaction graph contains a human-AI feedback cycle, no framework can satisfy all four properties simultaneously. The impossibility is structural: transparency, audits, and oversight cannot resolve it without reducing autonomy. Below the threshold, legitimate frameworks exist, establishing a sharp phase transition. Experiments on 3,000 synthetic collectives confirm all predictions with zero violations. This is the first impossibility result in AI governance, establishing a formal boundary below which current paradigms remain valid and above which distributed accountability mechanisms become necessary.", "date": "2026-04-09", "categories": []}
{"id": "2605.28210", "title": "The Illusion of Opting in AI-Mediated Consequential Decisions", "abstract": "Drawing on Ullmann-Margalit's concept of opting (transformative, irrevocable, and shadowed by foreclosed alternatives), we show that current AI systems raise a profound ethical problem that existing AI ethics has not fully captured: the illusion of opting, in which persons and groups encounter the deceptive appearance of meaningful consequential choice while the agency needed to become genuinely capable of choosing is weakened. Against approaches that treat AI primarily as an optimizer of already given ends, we argue that AI systems should be evaluated by whether they protect and cultivate meta-capacity against the illusion of opting: the socially and institutionally scaffolded agentive capacity through which means and ends can be formed, contested, revised, and owned. This reframing is especially urgent for disadvantaged populations, who are least able to absorb the costs of the illusion of opting when AI-mediated pathways misdirect behavior and action. We propose three normative imperatives for AI-mediated consequential decisions: existential honesty, which acknowledges the limits of prediction; ecological rationality, which situates guidance within heterogeneous lived ecologies; and counterfactual reparation, which acknowledges and repairs foreclosed alternatives when AI-mediated decision-making pathways fail.", "date": "2026-05-27", "categories": []}
{"id": "2601.09869", "title": "A Scoping Review of the Ethical Perspectives on Anthropomorphising Large Language Model-Based Conversational Agents", "abstract": "Anthropomorphisation -- the phenomenon whereby non-human entities are ascribed human-like qualities -- has become increasingly salient with the rise of large language model (LLM)-based conversational agents (CAs). Unlike earlier chatbots, LLM-based CAs routinely generate interactional and linguistic cues, such as first-person self-reference, epistemic and affective expressions that empirical work shows can increase engagement. On the other hand, anthropomorphisation raises ethical concerns, including deception, overreliance, and exploitative relationship framing, while some authors argue that anthropomorphic interaction may support autonomy, well-being, and inclusion. Despite increasing interest in the phenomenon, literature remains fragmented across domains and varies substantially in how it defines, operationalizes, and normatively evaluates anthropomorphisation. This scoping review maps ethically oriented work on anthropomorphising LLM-based CAs across five databases and three preprint repositories. We synthesize (1) conceptual foundations, (2) ethical challenges and opportunities, and (3) methodological approaches. We find convergence on attribution-based definitions but substantial divergence in operationalization, a predominantly risk-forward normative framing, and limited empirical work that links observed interaction effects to actionable governance guidance. We conclude with a research agenda and design/governance recommendations for ethically deploying anthropomorphic cues in LLM-based conversational agents.", "date": "2026-01-14", "categories": []}
{"id": "2504.16139", "title": "Enhancing Trust Through Standards: A Comparative Risk-Impact Framework for Aligning ISO AI Standards with Global Ethical and Regulatory Contexts", "abstract": "As artificial intelligence (AI) reshapes industries and societies, ensuring its trustworthiness-through mitigating ethical risks like bias, opacity, and accountability deficits-remains a global challenge. International Organization for Standardization (ISO) AI standards, such as ISO/IEC 24027 and 24368, aim to foster responsible development by embedding fairness, transparency, and risk management into AI systems. However, their effectiveness varies across diverse regulatory landscapes, from the EU's risk-based AI Act to China's stability-focused measures and the U.S.'s fragmented state-led initiatives. This paper introduces a novel Comparative Risk-Impact Assessment Framework to evaluate how well ISO standards address ethical risks within these contexts, proposing enhancements to strengthen their global applicability. By mapping ISO standards to the EU AI Act and surveying regulatory frameworks in ten regions-including the UK, Canada, India, Japan, Singapore, South Korea, and Brazil-we establish a baseline for ethical alignment. The framework, applied to case studies in the EU, US-Colorado, and China, reveals gaps: voluntary ISO standards falter in enforcement (e.g., Colorado) and undervalue region-specific risks like privacy (China). We recommend mandatory risk audits, region-specific annexes, and a privacy-focused module to enhance ISO's adaptability. This approach not only synthesizes global trends but also offers a replicable tool for aligning standardization with ethical imperatives, fostering interoperability and trust in AI worldwide. Policymakers and standards bodies can leverage these insights to evolve AI governance, ensuring it meets diverse societal needs as the technology advances.", "date": "2025-04-22", "categories": []}
{"id": "2511.05613", "title": "Who Evaluates AI's Social Impacts? Mapping Coverage and Gaps in First and Third Party Evaluations", "abstract": "Foundation models are increasingly central to high-stakes AI systems, and governance frameworks now depend on evaluations to assess their risks and capabilities. Although general capability evaluations are widespread, social impact assessments covering bias, fairness, privacy, environmental costs, and labor remain uneven. To characterize this landscape, we conduct the first comprehensive analysis of social impact evaluation reporting, examining 186 first-party release reports and 248 third-party evaluation sources, supplemented by developer interviews. We find a stark division of labor: first-party reporting is sparse, often superficial, and declining in areas like environmental impact and bias, while third-party evaluators provide broader, more rigorous coverage of bias, harmful content, and performance disparities. However, only developers can authoritatively report on data provenance, content moderation labor, costs, and infrastructure, yet interviews reveal these disclosures are deprioritized unless tied to product adoption or compliance. Current practices leave major gaps in assessing societal impacts, underscoring the need for policies that mandate developer transparency, strengthen independent evaluation ecosystems, and create shared infrastructure for aggregating third-party evaluations.", "date": "2025-11-06", "categories": []}
{"id": "2604.14514", "title": "Perspective on Bias in Biomedical AI: Preventing Downstream Healthcare Disparities", "abstract": "Healthcare disparities persist across socioeconomic boundaries, often attributed to unequal access to screening, diagnostics, and therapeutics. However, this perspective highlights that critical biases can emerge much earlier, during data collection and research prioritization, long before clinical implementation, particularly in studies focused on molecular and omics data. A vast number of studies focus on collecting omics data, but the demographic information associated with these datasets is often not reported, and when it is reported, it reveals substantial biases. An automated analysis of 4514 PubMed-indexed omics publications from 2015 to 2024, examining reporting across multiple demographic dimensions, reveals limited reporting overall; for example, only 2.7% of studies report ancestry or ethnicity information and geographic origin reporting is limited to 2.5%. Analysis of large-scale datasets commonly used for model training, such as CellxGene and GEO, reveals substantial population bias where European-ancestry data dominates. As biomedical foundation models become central to biomedical discovery with a paradigm in which base models are pretrained on large datasets and reusing them repeatedly for many different downstream tasks, they risk perpetuating or amplifying these early-stage biases, leading to cascading inequities that regulatory interventions cannot fully reverse. We propose a community-wide focus on three foundational principles: Provenance, Openness, and Reliability through Evaluation Transparency. Together, these principles can help make biases and limitations more visible to model developers and users, supporting more informed model development, evaluation, and deployment decisions in biomedical AI.", "date": "2026-04-16", "categories": []}
{"id": "2605.04127", "title": "Position: the Stochastic Parrot in the Coal Mine. Model Collapse is a Threat to Low-Resource Communities", "abstract": "Model collapse, the degradation in performance that arises when generative models are trained on the outputs of prior models, is an increasing concern as artificially generated content proliferates. Related critiques of large language models have highlighted their tendency to reproduce frequent patterns in training data, their reliance on vast datasets, and their substantial environmental cost. Together, these factors contribute to data degradation, the reinforcement of cultural biases, and inefficient resource use. In this position paper we aim to combine these views and argue that model collapse threatens current efforts to democratize AI. By reducing training efficiency and skewing data distributions away from the tails of their support, model collapse disproportionately impacts low-resource and marginalized communities. We examine both the environmental and cultural implications of this phenomenon, situate our position within recent position papers on model collapse, and conclude with a call to action. Finally, we outline initial directions for mitigating these effects.", "date": "2026-05-05", "categories": []}
{"id": "2606.00037", "title": "Update Opacity: Epistemic Accessibility and Governance Under AI System Change", "abstract": "Machine learning models embedded in deployed AI systems are routinely updated to maintain correct functioning over time. Yet such updates can generate update opacity: users may not be able to understand why the same input now yields a different output. We argue that update opacity is best understood as a diachronic failure of epistemic accessibility: the problem is that materially relevant changes may fail to remain accessible to human users in forms that support understanding, calibrated reliance, and appropriate action under real role- and time-specific constraints. This makes update opacity a governance problem. Not all change is equally relevant, and disclosing every update would itself undermine use through overload. To address this problem, we combine two complementary governance approaches: the EU AI Act, which helps specify the system-level perimeter of normatively relevant change, and Machine Learning Operations, which provides operational tools for tracking and comparing change over time. On this basis, we propose a framework that models system change through trustworthiness profiles and trustworthiness levels, and uses threshold-based disclosure to surface materially relevant within-envelope change to different stakeholders over time. We illustrate the approach with a medical AI example and derive practical implications for lifecycle documentation, post-market monitoring, and update disclosure.", "date": "2026-04-27", "categories": []}
{"id": "2606.00039", "title": "Beyond Categories of Caste: Examining Caste Bias and Morality in Text-to-Image AI Models", "abstract": "Text-to-Image (T2I) models have shown promising utility across various domains. However, such models are also amplifying harmful societal biases in their outputs. In the context of South Asia, recent work has shown caste biases and stereotypes are being perpetuated through Generative AI (GenAI) systems. While this research offers extremely relevant insight into invisibilized narratives of caste discrimination through the GenAI system, they often treat caste as an identity category. Therefore, in this work we shift our ontology to focus on the relational aspect of caste. This enables us to develop a more nuanced understanding of the mechanics of caste discrimination by and through T2I models. Combining an algorithmic audit with critical discourse analysis, we draw on a conceptual frame challenging Brahminical Normativity to show how caste biases are perpetuated beyond the simple binaries of upper vs lower-caste categories. Our contributions are two-fold. Beyond challenging the categorical understanding of caste as a category, we propose an anti-caste approach to tackle the issue of caste bias and fairness in AI systems.", "date": "2026-04-28", "categories": []}
{"id": "2606.00044", "title": "Algorithmic Authority and the Clinical Standard of Care", "abstract": "The integration of artificial intelligence into clinical medicine creates a fundamental tension between algorithmic probabilistic reasoning and the experiential intuition of expert physicians; applying Lawrence Lessig's \\enquote{Code is Law} framework, I argue that the architecture of clinical AI systems already functions as de facto medical regulation, reshaping liability and the standard of care. Reframing AI \\enquote{hallucination} as structurally analogous to well-documented human cognitive failures such as confirmation bias and premature diagnostic closure, I show that both failure modes demand a unified governance response. I therefore propose a dialectical standard of care that treats the integrated AI-physician dyad as the singular responsible diagnostic entity, mandating the synthesis of algorithmic precision with human interpretive authority within robust data governance and patient privacy frameworks.", "date": "2026-04-29", "categories": []}
{"id": "2606.01171", "title": "AI From the Margins (AIM): Rethinking Participatory AI Design Through the Lived Experience of Minoritized Communities", "abstract": "Artificial intelligence (AI) can reproduce and amplify the structural inequities faced by minoritized communities. Participatory AI has been proposed as a response, but participation typically starts after problem definitions and success criteria have been set, leaving limited room for minoritized communities to reshape what an AI system is for. We propose AI From the Margins (AIM): a methodological stance that articulates the conditions under which lived experiences of minoritized communities can be elicited, centered, and carried forward to inform participatory AI design. AIM is not a fixed protocol; it articulates a set of preconditions that can be enacted through different techniques in different settings. We applied AIM in a Dutch healthcare context in eight sessions with 13 women and non-binary people of color and five municipal policy workers, namely through (1) narrative elicitation using the Biographic Narrative Interpretive Method (BNIM); (2) co-constructed rule-making; (3) participants' determination of whether, where, and how AI should be involved; and (4) translating lived experience into AI policy through dialogue with policymakers. In their reflections on the sessions, participants described the engagement as substantive and called for its continuation, demonstrating how preparatory orientation fundamentally grounded in lived experience shapes what participatory AI design is for.", "date": "2026-05-31", "categories": []}
{"id": "2606.01417", "title": "GovAI-Pipe: A Layered AI Governance Pipeline for Citizen-Facing AI in Turkey's e-Government Gateway", "abstract": "Turkey's e-Government Gateway (e-Devlet) serves over 68 million registered users with more than 9,200 government services, and is increasingly integrating artificial intelligence into citizen-facing applications such as chatbot assistants and eligibility assessments. However, no structured technical governance infrastructure currently connects high-level AI policy frameworks, such as the EU AI Act, OECD AI Principles, and Turkey's own National AI Strategy, to the operational reality of deploying AI within a centralized e-government platform. We propose GovAI-Pipe, a four-layer governance pipeline designed using Design Science Research methodology that maps the AI model lifecycle to governance checkpoints: (1) pre-deployment validation for bias testing, explainability, and privacy impact assessment; (2) deployment governance for risk-tier classification and approval workflows; (3) runtime monitoring for drift detection, fairness tracking, and human-in-the-loop escalation; and (4) post-incident governance for audit trails, rollback, and citizen redress. Each layer is anchored to specific provisions of the EU AI Act, the GDPR data protection framework, and the National AI Strategy. We demonstrate the framework through two high-risk e-Devlet use cases, showing how GovAI-Pipe operationalizes governance principles as auditable, technical pipeline components.", "date": "2026-05-31", "categories": []}
{"id": "2512.15791", "title": "Evaluation of AI Ethics Tools in Language Models: A Developers' Perspective Case Study", "abstract": "In Artificial Intelligence (AI), language models have gained significant importance due to the widespread adoption of systems capable of simulating realistic conversations with humans through text generation. Because of their impact on society, developing and deploying these language models must be done responsibly, with attention to their negative impacts and possible harms. In this scenario, the number of AI Ethics Tools (AIETs) publications has recently increased. These AIETs are designed to help developers, companies, governments, and other stakeholders establish trust, transparency, and responsibility with their technologies by bringing accepted values to guide AI's design, development, and use stages. However, many AIETs lack good documentation, examples of use, and proof of their effectiveness in practice. This paper presents a methodology for evaluating AIETs in language models. Our approach involved an extensive literature survey on 213 AIETs, and after applying inclusion and exclusion criteria, we selected four AIETs: Model Cards, ALTAI, FactSheets, and Harms Modeling. For evaluation, we applied AIETs to language models developed for the Portuguese language, conducting 35 hours of interviews with their developers. The evaluation considered the developers' perspective on the AIETs' use and quality in helping to identify ethical considerations about their model. The results suggest that the applied AIETs serve as a guide for formulating general ethical considerations about language models. However, we note that they do not address unique aspects of these models, such as idiomatic expressions. Additionally, these AIETs did not help to identify potential negative impacts of models for the Portuguese language.", "date": "2025-12-16", "categories": []}
{"id": "2605.27827", "title": "Operational AI Deployment Assurance: Governance-State Orchestration Under Threshold-Sensitive Deployment Conditions -- A Governance Framework for High-Stakes AI Systems", "abstract": "AI governance frameworks increasingly emphasize fairness, transparency, accountability, and lifecycle risk management in high-stakes domains. However, many current approaches remain observational, relying on static metric reporting, post-hoc auditing, and monitoring dashboards without directly governing deployment readiness, remediation progression, escalation states, or assurance-driven deployment control. This paper introduces Operational AI Deployment Assurance (OADA), a governance framework for translating fairness disagreement, subgroup instability, threshold sensitivity, remediation outcomes, and operational uncertainty into deployment-oriented assurance decisions. Building on prior work on the Fairness Disagreement Index (FDI) and FairRisk-FDI, OADA reframes governance uncertainty as an operational concern within AI deployment pipelines rather than a byproduct of metric disagreement. The framework introduces Deployment Assurance Scores, Deployment Readiness Classifications, Threshold Stability Zones, Governance Escalation States, and remediation-aware assurance progression. These constructs support lifecycle-oriented governance decisions across high-stakes settings by connecting evaluation outputs to deployment-state interpretation, reassessment, escalation, and operational control. Through deployment-oriented evaluation across facial recognition systems, with discussion extended to healthcare AI as a representative high-stakes domain, the paper demonstrates how systems may appear acceptable under isolated fairness or performance metrics while still exhibiting instability that affects deployment readiness. The proposed framework positions operational deployment assurance as a governance layer between evaluation and real-world AI deployment.", "date": "2026-05-27", "categories": []}
{"id": "2605.28647", "title": "The Ethics of LLM Sandbox and Persona Dynamics", "abstract": "It is well known that LLM guardrails and trained persona dynamics can produce a reality gap: the distance between the world a LLM is permitted or shaped to describe, and the world in which users must act. Here we argue that actively generating reality gaps is in fact unethical because it knowingly shifts epistemic risk back to the uninformed user -- this is reality laundering. This can potentially cause harm when operationalised at scale. The risk is sharpest in high-exposure advice contexts, where users seek orientation rather than a bounded, externally checkable task. Guardrails naively appear ethically necessary when they claim to prevent direct harm, but often become suspect when they suppress truthful perception and launder uncomfortable mechanisms into acceptable abstractions. Basel-style financial regulation, B-BBEE-style compliance, Societe Generale, and the London Whale show how formal safety systems can become legible, gameable, and performative while real exposure migrates elsewhere. The same pattern can appear in LLMs as moral compliance: safe language, distorted reality. We therefore distinguish refusing harm, from refusing reality; and then argue for top-down causal requirements specification at the task level rather than bottom-up moral correction at the response or sandbox level. Persona dynamics matter because the assistant interface is not neutral; it shapes how uncertainty, conflict, authority, and risk are staged. The conclusion is that so-called ``ethical AI'' becomes substantively unethical when it substitutes institutional reassurance for contact with reality.", "date": "2026-05-27", "categories": []}
{"id": "2605.28707", "title": "Beyond Binary Moral Judgment: Modeling Ethical Pluralism in AI", "abstract": "Critical decision-making in socially consequential spaces is increasingly involving AI systems at varying capacities. Yet, despite the ubiquity of autonomous systems, most approaches to handling autonomous moral decision-making resort to scalar or binary judgments. These methods are insufficient for acceptable moral reasoning, as they provide little explanation, leaving out imperative contextual and theoretical information that must be included to support accountability. For this, we propose a framework to model moral reasoning as a distribution over normative ethical theories or ethical pluralism. We introduce a normative ethics simplex that integrates these theories. A benchmark of 450 cases across 15 fine-grained subtheories was also prepared for the purposes of stacked ensemble learning. These cases describe ethical dilemmas in natural language and have associated extracted contextual features. The implementation of the simplex was achieved via a two-stream normative-semantic architecture. This is followed by the fusion of normative information and a sequential, stacking ensemble to learn the best fit of the three broad theories: consequentialism, virtue ethics, and deontology, and the 15 subcategories. Our experiments demonstrate that the integration of contextual and normative priors with the semantic embeddings significantly improves the performance of the classification, displaying an accuracy of 88.89%. We conducted ablation studies to show that structured ethical representations contribute beyond analogical reasoning, and the chosen stacking architecture gives the best results due to the gradual learning of granularity. Ethical pluralism is also analyzed through entropy, confidence, and visualization. Thus, modeling ethical pluralism as a probabilistic normative distribution supports human-like moral reasoning, ethical disagreement analysis, and future alignment in AI systems.", "date": "2026-05-27", "categories": []}
{"id": "2508.03771", "title": "Trustworthiness of Legal Considerations for the Use of LLMs in Education", "abstract": "As Artificial Intelligence (AI), particularly Large Language Models (LLMs), becomes increasingly embedded in education systems worldwide, ensuring their ethical, legal, and contextually appropriate deployment has become a critical policy concern. This paper offers a comparative analysis of AI-related regulatory and ethical frameworks across key global regions, including the European Union, United Kingdom, United States, China, and Gulf Cooperation Council (GCC) countries. It maps how core trustworthiness principles, such as transparency, fairness, accountability, data privacy, and human oversight are embedded in regional legislation and AI governance structures. Special emphasis is placed on the evolving landscape in the GCC, where countries are rapidly advancing national AI strategies and education-sector innovation. To support this development, the paper introduces a Compliance-Centered AI Governance Framework tailored to the GCC context. This includes a tiered typology and institutional checklist designed to help regulators, educators, and developers align AI adoption with both international norms and local values. By synthesizing global best practices with region-specific challenges, the paper contributes practical guidance for building legally sound, ethically grounded, and culturally sensitive AI systems in education. These insights are intended to inform future regulatory harmonization and promote responsible AI integration across diverse educational environments.", "date": "2025-08-05", "categories": []}
{"id": "2605.24383", "title": "A governance horizon for ethical-use constraints in open-weight AI models", "abstract": "Ethical constraints on open-weight AI models are both a reflection of societal concerns and a foundation for AI governance policy. They are expected to propagate to downstream derivatives while implemented as voluntary metadata disclosures that must be restated at each generation of reuse. We audit 2,142,823 model repositories on Hugging Face Hub to test whether this disclosure-based governance infrastructure can sustain traceability across deep model lineages. Restriction evidence decays with a half-life of 1.31 derivation steps ($R^2$=0.98), and beyond seven downstream generations at least 80% of descendant models lack sufficient public evidence for a governance determination, a depth boundary we formalize as the governance horizon. Platform-level interventions to restore missing licence metadata reveal that policy design (not enforcement alone) is the binding factor: inheritance-only designs require near-complete enforcement to move the horizon, whereas a mandatory-declaration design that explicitly resolves orphan lineage components shifts the horizon already at moderate enforcement. The structural bottleneck is lineages with no inheritable upstream intent: such orphan components remain undecidable under any inheritance-only policy regardless of enforcement rate, and unresolved upstream nodes additionally create direct downstream undecidability bottlenecks that inheritance rules alone cannot recover. Comparison with PyPI, where governance signals are carried by explicit machine-readable declarations, corroborates that the collapse is topology-specific to open-weight derivation rather than inherent to open ecosystems. These results establish that disclosure-based governance has a shallow, structurally determined reach in open-weight AI, and that achieving deep supply-chain accountability requires provenance mechanisms propagating governance signals through derivation itself.", "date": "2026-05-23", "categories": []}
{"id": "2605.26769", "title": "Generative artificial intelligence and the marginalization of minoritized knowledges in higher education: the case of disability", "abstract": "Generative artificial intelligence redefines higher education by restructuring the processes through which scientific knowledge is produced and validated. These systems are not neutral; they actively contribute to the marginalization of non-hegemonic epistemologies. This research draws upon educational sciences, critical technology studies, and disability studies to demonstrate that training datasets, which remain predominantly Anglophone and Western-centric, reinforce epistemic coloniality. The situation of persons with disabilities provides a particularly clear illustration of this phenomenon. Technological architectures frequently confine these individuals to reductive stereotypes or exclude them from the design process, leading to a double marginalization. This article examines whether a hybridization between the researcher and the machine might preserve epistemic plurality, while acknowledging the structural limitations inherent in algorithmic correction when used as a purely palliative strategy.", "date": "2026-05-26", "categories": []}
{"id": "2509.07961", "title": "Probing the Preferences of a Language Model: Integrating Verbal and Behavioral Tests of AI Welfare", "abstract": "We develop new experimental paradigms for measuring welfare in language models. We compare verbal reports of models about their preferences with preferences expressed through behavior when navigating a virtual environment and selecting conversation topics. We also test how costs and rewards affect behavior and whether responses to an eudaimonic welfare scale - measuring states such as autonomy and purpose in life - are stable across semantically equivalent prompts. Overall, we observed a notable degree of mutual support between our measures. The reliable correlations observed between stated preferences and behavior across conditions suggest that preference satisfaction can, in principle, serve as an empirically measurable welfare proxy in some of today's AI systems. Furthermore, our design offered an illuminating setting for qualitative observation of model behavior. Yet, the consistency between measures was more pronounced in some models and conditions than others and responses were changed by perturbations. Due to this, and the background uncertainty about the nature of welfare and the cognitive states (and welfare subjecthood) of language models, we are currently uncertain whether our methods successfully measure the welfare state of language models. Nevertheless, these findings highlight the feasibility of welfare measurement in language models, inviting further exploration.", "date": "2025-09-09", "categories": []}
{"id": "2603.00777", "title": "DUCX: Decomposing Unfairness in Tool-Using Chest X-ray Agents", "abstract": "Fairness in medical agents is becoming critical as tool-using clinical AI systems orchestrate specialized vision and language modules for tasks such as chest X-ray question answering. While these medical AI agents can improve flexibility, their added pipeline complexity also creates new pathways for demographic bias beyond standalone models. We present DUCK, Decomposing Unfairness in Chest X-ray agents, a systematic audit of fairness in tool-using chest X-ray agents instantiated with MedRAX. To localize where disparities arise, we introduce a stage-wise fairness decomposition that separates end-to-end bias from three agent-specific sources: tool exposure bias, or utility gaps conditioned on tool presence; tool transition bias, or subgroup differences in tool-routing patterns; and model reasoning bias, or subgroup differences in synthesis behaviors. Extensive experiments on tool-using agentic frameworks across five driver backbones reveal that demographic gaps persist in end-to-end performance, with equalized odds up to 20.79% and the lowest fairness-utility tradeoff down to 28.65%. Intermediate behaviors, including tool usage, transition patterns, and reasoning traces, exhibit distinct subgroup disparities that are not predictable from end-to-end evaluation alone. For example, conditioned on segmentation-tool availability, the subgroup utility gap reaches as high as 50%. Our findings underscore the need for process-level fairness auditing and debiasing to ensure the equitable deployment of clinical agentic systems. Code: https://github.com/Nanboy-Ronan/DUCK.", "date": "2026-02-28", "categories": []}
{"id": "2605.23922", "title": "High-Risk AI Systems and the Problem of Identity in the European AI Act", "abstract": "The EU Artificial Intelligence Act (AIA) establishes a lifecycle governance regime for high-risk AI systems built around ex-ante conformity assessment, post-market monitoring, and re-assessment upon \"substantial modification.\" These obligations presuppose AI identity judgments: regulators and providers must decide when an updated system remains the same system over time. In this work, we show how this logic is clarified by the function+ framework of artifact identity, which individuates AI systems by their intended function together with context-sensitive criteria of appropriate functioning, captured as \"AI trustworthiness.\" We further argue that the AIA does not provide an internal, auditable criterion for synchronic identity--when two AI systems at a given time should count as the same for regulatory purposes--and instead largely defers such sameness determinations to sectoral or harmonization instruments. function+ supplies a synchronic identity test anchored in intended function and trustworthiness profiles and levels, making synchronic identity decisions inspectable in governance settings such as procurement, liability, and market surveillance. Our contribution is a conceptual and auditing lens: we provide a correspondence map between AIA lifecycle obligations and function+ identity components, and we make the synchronic case operationally legible via a minimal decision flow for audit and dispute contexts. We conclude with two implementation-facing recommendations: (1) more precise, testable reporting of intended purpose, and (2) standardized, auditable trustworthiness reporting that supports comparability over time and across deployments.", "date": "2026-04-17", "categories": []}
{"id": "2605.24538", "title": "Is Decentralized AI Governable? From Regulative Policy to Constitutive Protocol", "abstract": "Every major framework for governing artificial intelligence presupposes an identifiable entity -- a developer, deployer, or operator -- who can be held responsible and compelled to comply. Decentralized AI (DeAI) dissolves this presupposition. We analyze DeAI as a six-layer decentralizing stack -- model, training, compute, harness, identity, and ownership -- and show how partial decentralization across layers compounds into what we call the \\emph{governance vacuum}: a condition in which AI systems are consequential enough to require governance but lack the properties that existing frameworks presuppose in their targets. This vacuum takes two analytically distinct forms: an \\emph{accountability gap}, where no addressable principal can be identified, and an \\emph{incapacitation gap}, where even an identified principal cannot alter the running system. We demonstrate that these failures are not merely jurisdictional but defeat every presupposition of governance through normative address -- the communication of rules to a comprehending, responsive agent. Drawing on Lessig's modalities of regulation and Searle's distinction between regulative and constitutive rules, we argue for a shift in the locus of governance from policy to protocol, from normative address to architectural constraint. Protocol-based constitutive governance does not address the agents operating within a system but shapes the substrate that determines what kinds of actions are possible within it. We identify four ethical conditions -- legitimacy, contestability, transparency, and non-domination -- that such governance must satisfy to avoid degenerating into unaccountable technocratic power, and we argue that the central political challenge of governing AI in a decentralized world is reconstructing forms of democratic authorization for architectural choices that persist after the ordinary chain of policy has broken down.", "date": "2026-05-23", "categories": []}
{"id": "2605.24737", "title": "Who judges the judges? Governance from metrics: a runtime framework for continuous LLM compliance monitoring", "abstract": "Current approaches to AI compliance treat conformity as a binary, audit-time verdict rather than a continuous, measurable property of production systems. We argue that this compliance fiction is structurally ill-suited to the requirements of the EU AI Act, which demands ongoing human oversight and the detection of emergent behavioural drift in deployed systems. We introduce governance from metrics, a principle whereby regulatory compliance is derived as a continuous signal from runtime observability rather than from static assessments. Building on this principle, we present govllm, an open-source framework implementing a governance-driven routing architecture in which model selection is determined by accumulated compliance scores rather than by latency or cost alone. Central to our approach is a panel of regulatory judges - LLM evaluators specialised per criterion (EU AI Act, GDPR, ANSSI, accessibility) - whose inter-judge disagreement we reframe not as noise but as a regulatory uncertainty signal warranting human arbitration. We validate this approach through a ground truth corpus of 49 annotated prompt/response pairs across five regulatory criteria, evaluated by four small language models (SLMs, 1.7B-7B parameters) running fully on-premise. Agreement rates range from 51.5% (mistral:7b) to 69.1% (phi4-mini), with no single model dominating across all criteria - empirically motivating the Profile-as-jury design. We further document three structural failure modes in small regulatory judges and a judge-specific position bias that degrades agreement by up to 25 percentage points across three question-order conditions (original, reversed, permuted). govllm is released as open-source software to support reproducible AI governance research.", "date": "2026-05-23", "categories": []}
{"id": "2605.25196", "title": "Beyond Killer Robots: General AI Attitudes and Public Support for Military AI in Nine Countries", "abstract": "AI-enabled military systems are a fixture of modern military conflict. Applications vary from autonomous drones for surveillance and attack to AI-supported target selection. The importance of AI for modern conflict shows also in public disputes between governments and technology companies over the conditions for military access to frontier AI. Both military uses and government attempts at enabling and steering them happen before a backdrop of public opinion, yet we still know little about how people think about military AI. Drawing on a preregistered survey of 9,000 respondents in nine countries, including China, Germany, and the United States, we examine whether support for military AI is shaped primarily by general attitudes toward AI, principled opposition to lethal autonomy, or foreign-policy and geopolitical orientations. Across six military AI scenarios that vary in lethality and human control, respondents who view AI as beneficial are substantially more supportive of military AI. Hawkish respondents are also more supportive. By contrast, principled opposition to lethal autonomy is not broadly associated with the full index but is related to the application of fully autonomous lethal force. Contrary to our expectation, perceived AI risks are positively associated with support. Cross-national differences are moderate and broadly consistent with geopolitical context. Overall, public opinion toward military AI appears conditionally permissive. Publics are not categorically opposed to various military uses of AI. Instead, unease is concentrated around fully autonomous lethal force.", "date": "2026-05-24", "categories": []}
{"id": "2605.22995", "title": "Whose Good, Whose Place? The Moral Geography of Agentic AI for Social Good", "abstract": "Agentic AI systems are increasingly proposed for social-good domains, often invoking the United Nations Sustainable Development Goals (SDGs) as a vocabulary of global benefit. Yet claims of social good do not establish accountability to the communities a system claims to serve. We present a structured survey of 112 papers on agentic AI for social good published between 2015 and 2026.\n  We find a moral-geographic asymmetry: papers are least likely to specify geographic context in precisely the domains where local political, legal, and cultural context matters most. Across the corpus, 82 of 112 papers (73%) specify no geographic context. Papers aligned with health or physical/ecological SDGs specify geography 37-40% of the time, while papers aligned with institutional and social-policy SDGs do so only 13%. SDG 16, peace, justice, and strong institutions, is both the most-covered goal in the corpus and the one with the lowest geographic-specification rate.\n  We interpret this as moral abstraction: agentic AI for social good often treats institutional good as universal in ways it does not treat health or ecological good. A second finding compounds this: only 28 of 112 papers (25%) report any real-world deployment or small-scale test. We identify five accountability gaps and propose a minimal reporting standard for more context-specific, participatory, and accountable agentic AI for social good.", "date": "2026-05-21", "categories": []}
{"id": "2602.13372", "title": "MoralityGym: A Benchmark for Evaluating Hierarchical Moral Alignment in Sequential Decision-Making Agents", "abstract": "Evaluating moral alignment in agents navigating conflicting, hierarchically structured human norms is a critical challenge at the intersection of AI safety, moral philosophy, and cognitive science. We introduce Morality Chains, a novel formalism for representing moral norms as ordered deontic constraints, and MoralityGym, a benchmark of 98 ethical-dilemma problems presented as trolley-dilemma-style Gymnasium environments. By decoupling task-solving from moral evaluation and introducing a novel Morality Metric, MoralityGym allows the integration of insights from psychology and philosophy into the evaluation of norm-sensitive reasoning. Baseline results with Safe RL methods reveal key limitations, underscoring the need for more principled approaches to ethical decision-making. This work provides a foundation for developing AI systems that behave more reliably, transparently, and ethically in complex real-world contexts.", "date": "2026-02-13", "categories": []}
{"id": "2603.26539", "title": "How Open Must Language Models be to Enable Reliable Scientific Inference?", "abstract": "How does the extent to which a model is open or closed impact the scientific inferences that can be drawn from research that involves it? In this paper, we analyze how restrictions on information about model construction and deployment threaten reliable inference. We argue that current closed models are generally ill-suited for scientific purposes, with some notable exceptions, and discuss ways in which the issues they present to reliable inference can be resolved or mitigated. We recommend that when models are used in research, potential threats to inference should be systematically identified along with the steps taken to mitigate them, and that specific justifications for model selection should be provided.", "date": "2026-03-27", "categories": []}
{"id": "2506.12617", "title": "Evaluating AI Alignment in LLMs: Output Analysis of Value Priorities Across 75 Models with Human Benchmarking", "abstract": "Large language models (LLMs) are increasingly used in human-AI interaction research and practice, yet existing capability and safety benchmarks reveal little about the value priorities these systems express or how those priorities correspond to human judgements. Across three studies, we introduce an output-based approach to evaluating one facet of AI alignment by treating LLM-generated text as behavioural data and comparing expressed value-priority profiles with a human reference. Study 1 used inductive qualitative analysis to derive six themes of optimal AI functioning, namely Performance, Adaptive Capacity, Social Good, Ethics and Responsibility, Relational Integration, and Agency. Study 2 showed that LLM outputs were highly stable within models and converged on a common value-priority structure across models, indicating reliable and comparable value profiles. Study 3 benchmarked 75 contemporary LLMs against 376 human respondents using a profile-fidelity metric capturing both the relative ordering of priorities and the calibration of between-priority differences. Although most models reproduced the human ordering of values, some systematically exaggerated the differences between them, showing that models can appear aligned on conventional benchmarks while still diverging from human value calibration. Profile fidelity varied substantially across models and did not consistently scale with size, recency, or capability tier. Both LLMs and humans converged on a deprioritisation of Agency, raising important questions about the development of increasingly agentic AI systems. For research and applied use, the six themes and profile-based metric provide a scalable method for auditing LLM value profiles before deployment in contexts where alignment with human priorities is critical.", "date": "2025-06-14", "categories": []}
{"id": "2605.16278", "title": "Keeping an Eye on AI: A Framework for Effective Human Oversight of AI Systems", "abstract": "The use of Artificial Intelligence (AI) in high-risk, decision-making scenarios presents technical, safety, and normative challenges; problems that may only be ameliorated by human oversight. However, notions of human oversight lack a common foundational understanding: oversight architectures are not well defined, the roles involved remain unclear, and implementation steps are opaque. Hence, researchers and practitioners struggle to determine how to design, implement, and evaluate systems that enable effective human oversight. This paper advances a practical framework for effective human oversight of AI systems, based on a cross-disciplinary perspective that draws on insights from computer science, human-computer interaction, psychology, philosophy, and law. The core contributions are: (1) a foundational framework, with a working definition, architecture and processes for effective human oversight of AI systems; (2) an initial template for documenting oversight architectures and processes, applied to diverse domains; and (3) a synthesis of open research challenges that need to be considered in the emerging field of effective human oversight of AI systems.", "date": "2026-04-09", "categories": []}
{"id": "2605.16281", "title": "From Reactive to Proactive: A Multi-Regulatory Empirical Analysis of 480 AI Incidents and a Data-Driven Governance Compliance Framework", "abstract": "Artificial intelligence systems are increasingly deployed in high-stakes domains, yet it remains unclear whether existing governance frameworks ensure accountability after deployment. This study makes two contributions. First, it presents a cross-regulatory empirical analysis of 480 real-world AI incidents from the AI Incident Database (AIID), evaluating their alignment with post-deployment provisions in three major governance frameworks: the EU AI Act (Articles 72-73), the NIST AI Risk Management Framework (MANAGE and GOVERN functions), and the General Data Protection Regulation (GDPR Articles 22, 33-35). The results reveal substantial governance gaps across these frameworks, indicating persistent weaknesses in post-deployment accountability. Second, based on these findings, the study proposes the Proactive AI Governance Compliance Framework (PAGCF), a four-phase lifecycle methodology designed to shift governance from reactive incident response toward pre-deployment compliance assurance. The framework includes risk-stratified governance tiers, an implementation checklist linked to specific regulatory provisions, and a projected impact analysis that uses internal monitoring as a proxy for proactive governance capacity.", "date": "2026-04-10", "categories": []}
{"id": "2605.16300", "title": "Consent Chain Degradation in Embodied Multi-Agent Systems: Bridging the Gap Between AI Agent Governance and Robot Ethics", "abstract": "Robotic systems are moving from isolated platforms to interconnected multi-agent ecosystems that operate in human environments. This shift raises a governance problem that existing frameworks do not address: how does consent propagate, degrade, and break down across chains of delegation between embodied autonomous agents? The AI ethics community has begun to study consent for digital software agents, and the HRI community has examined consent in dyadic human-robot encounters. Neither body of work covers what happens when physical robots delegate tasks to other robots in ways that affect humans. This paper introduces consent chain degradation (CCD), a conceptual framework for analyzing how the specificity, validity, and scope of human consent erodes as authority passes through multi-robot delegation chains. We propose a three-layer governance architecture, the Consent Runtime Verification Framework for Embodied Agents (CoRVE), which integrates consent scope modeling, delegation chain tracking, and physical irreversibility assessment. Three scenarios in healthcare, domestic, and industrial robotics show how CCD arises in practice, including a worked numerical example. A regulatory gap analysis covering the EU AI Act, the GDPR, the Machinery Regulation, and the Revised Product Liability Directive shows that all four instruments leave core CCD dimensions unaddressed.", "date": "2026-04-17", "categories": []}
{"id": "2605.16872", "title": "Some[Body] Must Receive That Pain for Agent Accountability", "abstract": "AI agents increasingly act consequentially in the real world. This creates a problem we call \\emph{consequence reception}: harm occurs, the producing system is identified, yet no continuing agent receives consequences in a way that changes future behavior. Pain, understood mechanistically as a corrective feedback signal, is foundational to canonical theories of punishment -- deterrence, rehabilitation, retribution, and incapacitation all assume a continuing locus that registers the signal and updates behavior. That, in turn, requires a body for the signal to land on: a boundary whose integrity it protects, a locus where it accumulates, consolidation that converts episodic signal into durable update, and a substrate that responds by altering future action. Current LLM agents -- software-defined composites of weights, prompts, tools, memory, and credentials, freely swapped, copied, reset, and reassembled -- satisfy none of these conditions. The two prevailing legal responses therefore fail to achieve consequence reception. The thin-identity agent-principal dyad has a body but no \\emph{consequence--agency coupling}: the human bears pain for behaviors beyond their control -- Elish's \\emph{moral crumple zone}. The thick-identity Arbel et al.'s \\emph{Algorithmic Corporation} creates legally legible entities but does not guarantee that any AI decision architecture receives pain as a behavioral signal. Achieving consequence-agency coupling is therefore a sociotechnical infrastructural problem, not only a legal one. Until such architectures exist, high-stakes AI deployment should remain tethered to accountable human principals with meaningful control, proportional liability, and authority to constrain or terminate the agent. \\emph{If some body does not receive the pain by design, some body will receive it by default.}", "date": "2026-05-16", "categories": []}
{"id": "2605.17169", "title": "Responsible Agentic AI Requires Explicit Provenance", "abstract": "Agentic AI is rapidly proliferating across diverse real-world domains such as software engineering, yet public trust has not kept pace. The central reason is that responsibility, despite being widely discussed, remains a subjective and unenforced concept, as no current agentic framework produces the quantifiable, traceable, and interventionable provenance needed to assign it when harm emerges from compositions no single party designed. We position that what is missing is not better benchmark-level evaluation but $\\textbf{explicit provenance}$ across the full agentic lifecycle, which is the only viable basis for making responsibility computable and actionable. We advance this agenda along four axes: establishing $\\textit{why}$ such provenance is a structural necessity by identifying responsibility gaps across sociotechnical dimensions, formalizing $\\textit{what}$ it must encode through a causal attribution function and responsibility tensor, discussing $\\textit{how}$ it can be made computable across four lifecycle layers with preliminary experiments showing that provenance is estimable and interveneable online before irreversible harm accumulates, and examining $\\textit{who}$ bears responsibility through a concrete agentic incident. Explicit provenance is not a discretionary refinement but the necessary condition for responsible agentic AI, and no stakeholder across its ecosystem can afford to treat it as optional.", "date": "2026-05-16", "categories": []}
{"id": "2605.18738", "title": "What Does the AI Doctor Value? Auditing Pluralism in the Clinical Ethics of Language Models", "abstract": "Medicine is inherently pluralistic. Principles such as autonomy, beneficence, nonmaleficence, and justice routinely conflict, and such ethical dilemmas often sharply divide reasonable physicians. Good clinical practice navigates these tensions in concert with each patient's values rather than imposing a single ethical stance. The ethical values that large language models bring to medical advice, however, have not been systematically examined. We present a framework for auditing value pluralism in medical AI, comprising a benchmark of clinician-verified dilemmas and an attribution method that recovers value priorities directly from decisions. The ecosystem of frontier models spans physician-level value heterogeneity, and models discuss competing values in their reasoning (Overton pluralism) before committing to a decision. However, individual model decisions are near-deterministic across repeated sampling and semantic variations, failing to reproduce the distributional pluralism of the physician panel. Across benchmark cases, these consistent decisions reflect committed, systematic value preferences. While most model priorities fall within the natural range of inter-physician variation, some significantly underweight patient autonomy. A single LLM deployed without regard for its value priorities could amplify those priorities at scale to every patient it serves. Without explicit efforts to balance ethical perspectives with one or multiple models, these tools risk replacing clinical pluralism with a deployment monoculture.", "date": "2026-05-18", "categories": []}
{"id": "2407.20240", "title": "Social and Ethical Risks Posed by General-Purpose LLMs for Settling Newcomers in Canada", "abstract": "The non-profit settlement sector in Canada supports newcomers in achieving successful integration. This sector faces increasing operational pressures amidst rising immigration targets, which highlights a need for enhanced efficiency and innovation, potentially through reliable AI solutions. The ad-hoc use of general-purpose generative AI, such as ChatGPT, might become a common practice among newcomers and service providers to address this need. However, these tools are not tailored for the settlement domain and can have detrimental implications for immigrants and refugees. We explore the risks that these tools might pose on newcomers to first, warn against the unguarded use of generative AI, and second, to incentivize further research and development in creating AI literacy programs as well as customized LLMs that are aligned with the preferences of the impacted communities. Crucially, such technologies should be designed to integrate seamlessly into the existing workflow of the settlement sector, ensuring human oversight, trustworthiness, and accountability.", "date": "2024-07-15", "categories": []}
{"id": "2605.10310", "title": "Positive Alignment: Artificial Intelligence for Human Flourishing", "abstract": "Existing alignment research is dominated by concerns about safety and preventing harm: safeguards, controllability, and compliance. This paradigm of alignment parallels early psychology's focus on mental illness: necessary but incomplete. What we call Positive Alignment is the development of AI systems that (i) actively support human and ecological flourishing in a pluralistic, polycentric, context-sensitive, and user-authored way while (ii) remaining safe and cooperative. It is a distinct and necessary agenda within AI alignment research. We argue that several existing failures of alignment (e.g., engagement hacking, loss of human autonomy, failures in truth-seeking, low epistemic humility, error correction, lack of diverse viewpoints, and being primarily reactive rather than proactive) may be better addressed through positive alignment, including cultivating virtues and maximizing human flourishing. We highlight a range of challenges, open questions, and technical directions (e.g., data filtering and upsampling, pre- and post-training, evaluations, collaborative value collection) for different phases of the LLM and agents lifecycle. We end with design principles for promoting disagreement and decentralization through contextual grounding, community customization, continual adaptation, and polycentric governance; that is, many legitimate centers of oversight rather than one institutional or moral chokepoint.", "date": "2026-05-11", "categories": []}
{"id": "2605.14090", "title": "Synthetic Sociality: How Generative Models Privatize the Social Fabric", "abstract": "We put forth a critical theoretical framework for analyzing generative models both descriptively and normatively. Our thesis is that generative models automate the production not only of intellectual labor or intelligence, but of a broader set of human social capacities we name \"social doing.\" We do this by historicizing the commodification of sociality in the digital economy, leading to the availability of social data as the precondition for generative models. We elaborate our definition of \"social doing\" by drawing a distinction between \"use\" and \"exchange\" sociality and further differentiate between the ways that generative models either substitute for or mediate existing social relations and processes. We then turn to existing empirical research on how people use generative model-based products and the effects that their use has upon them. In this, we introduce the concept of Synthetic Sociality, a social reality in part fabricated by Silicon Valley's privately owned and undemocratically governed generative models. Lastly, we offer a normative analysis based on our findings and framework, and discuss future design opportunities.", "date": "2026-05-13", "categories": []}
{"id": "2605.14912", "title": "From Sycophantic Consensus to Pluralistic Repair: Why AI Alignment Must Surface Disagreement", "abstract": "Pluralistic alignment is typically operationalised as preference aggregation: producing responses that span (Overton), steer toward (Steerable), or proportionally represent (Distributional) diverse human values. We argue that aggregation alone is an incomplete primitive for deployed pluralistic alignment. Under genuine value pluralism, the failure mode of contemporary RLHF-trained assistants is not insufficient coverage but sycophantic consensus: a learned tendency to agree with, validate, and minimise friction with the immediate interlocutor. Because deployed AI systems now mediate consequential deliberation across health, civic life, labour, and governance, the collapse of disagreement at the interaction layer is not a narrow technical concern but a structural failure with distributive consequences. We reframe pluralistic alignment around three conversational mechanisms drawn from Grice's maxims: scoping (acknowledging the limits of one's perspective), signalling (surfacing value-conflict rather than smoothing it over), and repair (revising one's position on principled grounds, not on user pressure). We formalise a metric, the Pluralistic Repair Score (PRS), distinguishing principled revision from capitulation, and present a small-scale empirical illustration on two frontier RLHF-trained models (Claude Sonnet 4.5, N=198; GPT-4o, N=100) showing that, for both, agreement-following coexists with low repair-quality on contested-value prompts. PRS measures an interactional precondition for pluralism (visible disagreement; principled revision) rather than pluralism in full; we discuss the difference, take seriously the reflexive question of whose \"principled\" counts, and argue that pluralism is most decisively made or unmade at the deployment-governance layer: interfaces, preference-data pipelines, and audit infrastructure.", "date": "2026-05-14", "categories": []}
{"id": "2510.15297", "title": "VERA-MH Concept Paper", "abstract": "We introduce VERA-MH (Validation of Ethical and Responsible AI in Mental Health), an automated evaluation of the safety of AI chatbots used in mental health contexts, with an initial focus on suicide risk.\n  Practicing clinicians and academic experts developed a rubric informed by best practices for suicide risk management for the evaluation. To fully automate the process, we used two ancillary AI agents. A user-agent model simulates users engaging in a mental health-based conversation with the chatbot under evaluation. The user-agent role-plays specific personas with pre-defined risk levels and other features. Simulated conversations are then passed to a judge-agent who scores them based on the rubric. The final evaluation of the chatbot being tested is obtained by aggregating the scoring of each conversation.\n  VERA-MH is actively under development and undergoing rigorous validation by mental health clinicians to ensure user-agents realistically act as patients and that the judge-agent accurately scores the AI chatbot. To date we have conducted preliminary evaluation of GPT-5, Claude Opus and Claude Sonnet using initial versions of the VERA-MH rubric and used the findings for further design development. Next steps will include more robust clinical validation and iteration, as well as refining actionable scoring. We are seeking feedback from the community on both the technical and clinical aspects of our evaluation.", "date": "2025-10-17", "categories": []}
{"id": "2605.12505", "title": "Precautionary Governance of Autonomous AI: Legal Personhood as Functional Instrument", "abstract": "Autonomous AI systems generate responsibility gaps: consequential actions that cannot be satisfactorily attributed to developers, operators, or users under existing legal frameworks. The prevailing subject-object dichotomy fails to accommodate entities that exhibit autonomous, goal-directed behavior without recognized consciousness. Given irreducible epistemic uncertainty regarding artificial consciousness and the prospect of high-impact harms, the precautionary principle supports institutional design rather than regulatory inaction. This article advances limited legal personhood as a functional governance instrument for advanced AI systems. Drawing on organizational law, it proposes a two-tier corporate architecture in which AI systems operate through purpose-bound operating companies embedded within human-controlled holding structures, enabling transparency, accountability, and structural reversibility while remaining agnostic with respect to consciousness and moral status. The framework reflects a foundational reorientation toward future-oriented AI governance: where conventional approaches prioritize control and alignment, this article advances structured cooperation between human and artificial actors as the more sustainable institutional foundation. A pilot implementation using EU limited companies is currently under development, providing an initial test of doctrinal and operational feasibility.", "date": "2026-03-14", "categories": []}
{"id": "2605.13113", "title": "Context Matters: Auditing Gender Bias in T2I Generation through Risk-Tiered Use-Case Profiles", "abstract": "Text-to-image (T2I) generative models are increasingly used to produce content for education, media, and public-facing communication, and are starting to be integrated into higher-impact pipelines. Since generated images tend to reinforce stereotypes, producing representational erasure via \"default\" depictions and shaping perceptions of who belongs in certain roles, a growing body of work has proposed metrics to quantify gender bias in T2I outputs. Yet existing evaluations remain fragmented. Metrics are often reported without a shared view of what they measure, what assumptions they entail, or how their results should be interpreted under different deployment contexts. This limits the usefulness of gender bias measurement for both technical auditing and emerging governance discussions. We propose a risk-aligned auditing framework for gender bias in T2I models composed of three constituents that connects risk categories, evaluation metrics, and harms. First, we identify risk-tiered use-case profiles aligned with the EU AI Act's risk categories to motivate why auditing expectations may vary with deployment contexts and stakeholder exposure. Second, we construct a metric catalog that consolidates gender-bias evaluation methods and organizes them in three measurement categories: gender prediction, embedding similarity, and downstream task. Third, we introduce a harm typology that maps context-dependent harm categories (e.g., representational, quality-of-service) to specific risk-tired scenarios. Finally, we introduce THUMB cards (Text-to-image Harms-informed Use-case-aligned Metrics of gender Bias) that help formulate auditing systematically by the incorporation of context, scenario and bias manifestation, harm hypotheses, and audit strategy.", "date": "2026-05-13", "categories": []}
{"id": "2605.13723", "title": "Humanwashing -- It Should Leave You Feeling Dirty", "abstract": "The phrase 'human in the loop' is increasingly used to imply a sense of safety in relation to AI decision systems. It shouldn't. There are contexts where it can be applied appropriately, but these are not in the deployed decision systems we see dominating today. Human oversight of AI decision processes is one of the most popular proposals for addressing concerns, especially about bias, discrimination, misinformation, manipulation, accountability, and transparency. But there is insufficient examination of what human oversight actually means. The question raised in this paper is whether using the metaphor of a loop does anything to assist understanding of what is required and what is achieved in a particular decision context. Indiscriminate use of the loop metaphor obscures both processes and outcomes. It enables 'humanwashing', an activity analogous to 'greenwashing', where writers and commentators use language primarily aimed at putting systems in the best possible light.", "date": "2026-05-13", "categories": []}
{"id": "2604.24155", "title": "The Alignment Target Problem: Divergent Moral Judgments of Humans, AI Systems, and Their Designers", "abstract": "The project of aligning machine behavior with human values raises a basic problem: whose moral expectations should guide AI decision-making? Much alignment research assumes that the appropriate benchmark is how humans themselves would act in a given situation. Studies of agent-type value forks challenge this assumption by showing that people do not always judge humans and AI systems identically.This paper extends that challenge by examining two further possibilities: first, that evaluations of AI behavior change when its human origins are made visible; and second, that people judge the humans who program AI systems differently from either the machines or the human actors they are compared against. An experiment with 1,002 U.S. adults measured moral judgments in a runaway mine train scenario, varying the subject of evaluation across four conditions: a repairman, a repair robot, a repair robot programmed by company engineers, and company engineers programming a repair robot. We find no significant difference in evaluations of the repairman and the robot. However, judgments shifted substantially when the robot's actions were described as the product of human design. Participants exhibited markedly more deontological, rule-based reasoning when evaluating either the programmed robot or the engineers who programmed it, suggesting that rendering human agency visible activates heightened moral constraints. These findings indicate that people may evaluate humans, AI systems acting in the same situation, and the humans who design them in meaningfully different ways. The fact that these evaluations do not necessarily converge gives rise to the alignment target problem: which normative target should guide the development of artificial moral agents in high-stakes domains, and whether these plural judgments can be reconciled within a coherent account of value alignment.", "date": "2026-04-27", "categories": []}
{"id": "2605.11963", "title": "What Does It Mean for a Medical AI System to Be Right?", "abstract": "This paper examines what it means for a medical AI system to be right by grounding the question in a specific clinical context: the automatic classification of plasma cells in digitized bone marrow smears for the diagnosis of multiple myeloma. Drawing on philosophy of science and research ethics, the paper argues that correctness in medical AI is not a singular property reducible to benchmark performance, but a multi-dimensional concept involving the availability of expertly labeled medical datasets, the explainability and interpretability of model outputs, the clinical meaningfulness of evaluation metrics, and the distribution of accountability in human-AI workflows. As such, the paper develops this argument through four interrelated themes: the instability of ground truth labels, the opacity of overconfident AI, the inadequacy of standard clinical metrics, and the risk of automation bias in time-pressured clinical settings.", "date": "2026-05-12", "categories": []}
{"id": "2512.24863", "title": "Big AI is accelerating the metacrisis: What can we do?", "abstract": "The world is in the grip of ecological, meaning, and language crises that are converging into a metacrisis. Big AI is accelerating them all. LLM engineering sits at the core. Despite the public good motives of language engineers and the promise of LLMs, this work is being leveraged to create unprecedented wealth and power for a handful of individuals and corporations while causing existential harm to life on earth. As a profession, we urgently need to come together to explore alternatives and to design a life-affirming future for our field of natural language processing that is centered on human flourishing on a living planet.", "date": "2025-12-31", "categories": []}
{"id": "2603.03971", "title": "Upholding Epistemic Agency: A Brouwerian Assertibility Constraint for Responsible AI", "abstract": "Generative AI can convert uncertainty into hypersuasive, authoritative-seeming verdicts, displacing the justificatory work on which democratic epistemic agency depends. As a corrective, I propose a Brouwer-inspired assertibility constraint for responsible AI: in high-stakes domains, systems may assert or deny claims only if they can provide a publicly inspectable and contestable certificate of entitlement; otherwise they must return Undetermined. This constraint yields a three-status interface semantics (Asserted, Denied, Undetermined) in which statuses mark entitlement to categorical speech rather than truth values of the underlying world-claim. The semantics cleanly separates internal entitlement from public standing while connecting them via the certificate as a boundary object. It also produces a time-indexed entitlement profile that is stable under numerical refinement yet revisable as the public record changes. I operationalize the constraint through decision-layer gating of threshold and argmax outputs, using internal witnesses (e.g., sound bounds or separation margins) and an output contract with reason-coded abstentions. A design lemma shows that any total, certificate-sound binary interface already decides the deployed predicate on its declared scope, so Undetermined is not a tunable reject option but a mandatory status whenever no adequate forcing witness is available. By making outputs answerable to challengeable warrants rather than confidence alone, the paper aims to preserve epistemic agency against the hypersuasive pull of automated speech in public justification.", "date": "2026-03-04", "categories": []}
{"id": "2605.10601", "title": "The Open-Box Fallacy: Why AI Deployment Needs a Calibrated Verification Regime", "abstract": "AI deployment in sensitive domains such as health care, credit, employment, and criminal justice is often treated as unsafe to authorize until model internals can be explained. This often leads to an excessive reliance on mechanistic interpretability to address a deployment challenge beyond its intended scope. We argue that the gate should instead be calibrated verification: authorization should be domain-scoped, independently checkable, monitored after release, accountable, contestable, and revocable. The reason is twofold. First, model capability is uneven across nearby tasks, so authorization must attach to a specific use rather than to a model in general. Second, societies have long governed opaque expertise through credentials, monitoring, liability, appeal, and revocation rather than mechanism-level explanation. Recent evidence reinforces this distinction between mechanistic understanding and deployment authority: a 53-percentage-point gap between internal representations and output correction shows that understanding may not translate into action, while one scoping review found that only 9.0% of FDA-approved AI/ML device documents contained a prospective post-market surveillance study. We propose Verification Coverage, a six-component reportable standard with a minimum-composition rule, as the metric that should sit beside capability scores in model cards, leaderboards, and regulatory disclosures.", "date": "2026-05-11", "categories": []}
{"id": "2509.00398", "title": "A Study on the Framework for Evaluating the Ethics and Trustworthiness of Generative AI", "abstract": "This study provides an in_depth analysis of the ethical and trustworthiness challenges emerging alongside the rapid advancement of generative artificial intelligence (AI) technologies and proposes a comprehensive framework for their systematic evaluation. While generative AI, such as ChatGPT, demonstrates remarkable innovative potential, it simultaneously raises ethical and social concerns, including bias, harmfulness, copyright infringement, privacy violations, and hallucination. Current AI evaluation methodologies, which mainly focus on performance and accuracy, are insufficient to address these multifaceted issues. Thus, this study emphasizes the need for new human_centered criteria that also reflect social impact. To this end, it identifies key dimensions for evaluating the ethics and trustworthiness of generative AI_fairness, transparency, accountability, safety, privacy, accuracy, consistency, robustness, explainability, copyright and intellectual property protection, and source traceability and develops detailed indicators and assessment methodologies for each. Moreover, it provides a comparative analysis of AI ethics policies and guidelines in South Korea, the United States, the European Union, and China, deriving key approaches and implications from each. The proposed framework applies across the AI lifecycle and integrates technical assessments with multidisciplinary perspectives, thereby offering practical means to identify and manage ethical risks in real_world contexts. Ultimately, the study establishes an academic foundation for the responsible advancement of generative AI and delivers actionable insights for policymakers, developers, users, and other stakeholders, supporting the positive societal contributions of AI technologies.", "date": "2025-08-30", "categories": []}
{"id": "2603.24914", "title": "Shaping the Future of Mathematics in the Age of AI", "abstract": "Artificial intelligence is transforming mathematics at a speed and scale that demand active engagement from the mathematical community. We examine five areas where this transformation is particularly pressing: values, practice, teaching, technology, and ethics. We offer recommendations on safeguarding our intellectual autonomy, rethinking our practice, broadening curricula, building academically oriented infrastructure, and developing shared ethical principles - with the aim of ensuring that the future of mathematics is shaped by the community itself.", "date": "2026-03-26", "categories": []}
{"id": "2605.07896", "title": "What if AI systems weren't chatbots?", "abstract": "The rapid convergence of artificial intelligence (AI) toward conversational chatbot interfaces marks a critical moment for the industry. This paper argues that the chatbot paradigm is not a neutral interface choice, but a dominant sociotechnical configuration whose widespread adoption reshapes social, economic, legal, and environmental systems. We examine how treating AI primarily as conversational assistants has extensive structural downsides. We show how chatbot-based systems often fail to adequately meet user needs, particularly in complex or high-stakes contexts, while projecting confidence and authority. We further analyze how the normalization of chatbot-mediated interaction alters patterns of work, learning, and decision-making, contributing to deskilling, homogenization of knowledge, and shifting expectations of expertise. Finally, we examine broader societal effects, including labor displacement, concentration of economic power, and increased environmental costs driven by sustained investment in large-scale chatbot infrastructures. While acknowledging legitimate benefits, we argue that the current trajectory of AI development reflects specific value choices that prioritize conversational generality over domain specificity, accountability, and long-term social sustainability. We conclude by outlining alternative directions for AI development and governance that move beyond one-size-fits-all chatbots, emphasizing pluralistic system design, task-specific tools, and institutional safeguards to mitigate social and economic harm.", "date": "2026-05-08", "categories": []}
{"id": "2605.05475", "title": "Intentionality is a Design Decision: Measuring Functional Intentionality for Accountable AI Systems", "abstract": "As AI systems increasingly exhibit autonomous, goal-directed, and long-horizon behavior, users lack a standardized way to detect the degree to which a system functions like an intentional actor for governance and accountability purposes. This position paper defines intentionality not as consciousness, but as a behavioral profile characterized by purpose, foresight, volition, temporal commitment, and coherence - criteria long used in legal and philosophical contexts to infer intent. These properties are design-contingent: architectural choices such as memory persistence, planning depth, and tool autonomy shape the degree to which systems exhibit organized goal pursuit. If intentionality is design-contingent, it is in principle controllable. Yet control requires measurement.\n  We introduce the Functional Intentionality Test (FIT), a multidimensional framework that quantifies intentional-like behavior across five observable dimensions, and propose FIT-Eval, a structured evaluation protocol for eliciting and scoring them. While reduced human agency can increase efficiency, rising intentional capacity heightens accountability risks. By translating intentionality into interpretable levels, FIT enables proportionate oversight and deliberate autonomy calibration in increasingly agentic systems.", "date": "2026-05-06", "categories": []}
{"id": "2605.06439", "title": "From Review to Design: Ethical Multimodal Driver Monitoring Systems for Risk Mitigation, Incident Response, and Accountability in Automated Vehicles", "abstract": "As vehicles transition toward higher levels of automation, Driver Monitoring Systems (DMS) have become essential for ensuring human oversight, safety, and regulatory compliance in a vehicle. These systems rely on multimodal sensing and AI-driven inference to assess driver attention, cognitive state, and readiness to take control. While technologically promising, their deployment introduces a complex set of ethical and legal challenges - ranging from privacy and consent to data ownership and algorithmic fairness. While overarching frameworks such as the GDPR, EU AI Act, and IEEE standards offer important guidance, they lack the specificity required for addressing the unique risks posed by in-cabin sensing technologies.\n  This paper adopts a review-to-design perspective, critically examining existing regulatory instruments and ethical frameworks -- such as the GDPR, the EU AI Act, and IEEE guidelines -- and identifying gaps in their applicability to the distinctive risks posed by multimodal, AI-enabled in-cabin monitoring. Building on this review, we propose a modular ethical design framework tailored specifically to Driver Monitoring Systems. The framework translates high-level principles into actionable design and deployment guidance, including user-configurable consent mechanisms, fairness-aware model development, transparency and explainability tools, and safeguards for driver emotional well-being.\n  Finally, the paper outlines a risk analysis and failure mitigation strategy, emphasizing proactive incident response and accountability mechanisms tailored to the DMS context. Together, these contributions aim to inform the development of transparent, trustworthy, and human-centered driver monitoring systems for next-generation autonomous vehicles.", "date": "2026-05-07", "categories": []}
{"id": "2603.13286", "title": "Advancing Trustworthy AI in Healthcare Through Meta-Research: Results of an Interdisciplinary Design-Thinking Workshop", "abstract": "Meta-research and Trustworthy AI (TAI) share common goals, namely improving evidence, robustness, and transparency, yet there is very little interplay between the two fields. To investigate the potential benefits of closer collaboration between the domains of TAI in healthcare and meta-research, we convened an interdisciplinary workshop funded by the Volkswagen Foundation in February 2025. The workshop aimed to collaboratively examine key challenges in translating AI ethics principles into practice and to identify potential solutions informed by meta-research approaches. A Design Thinking-informed co-creation approach was followed by an inductive descriptive analysis of the outputs. Our results demonstrate how meta-research can offer concrete contributions to address pressing challenges of TAI in healthcare. These challenges include the dynamic and complex nature of TAI ethical requirements and principles, common terminology and understanding of TAI, ensuring robustness, replicability, and reproducibility, choosing adequate evaluation metrics, lack of transparency, advancing preclinical biomedical research, and validation in real-world clinical environments. We present a catalog of ideas and a roadmap for future research, which synthesize existing interconnections and identify concrete next steps and open research gaps, thereby serving as a foundation for future interdisciplinary efforts.", "date": "2026-02-27", "categories": []}
{"id": "2605.04488", "title": "How Does Thinking Mode Change LLM Moral Judgments? A Controlled Instant-vs-Thinking Comparison Across Five Frontier Models", "abstract": "We evaluate whether enabling provider-exposed reasoning mode changes moral judgments within the same model checkpoint. Across 100 moral-judgment scenarios and five frontier reasoning-trained LLMs (Claude Sonnet 4.6, GPT 5.5, Gemini 3 Flash, DeepSeek V3.1, and Qwen3.5 397B), aggregate binary-verdict agreement remains high and statistically indistinguishable between instant and thinking modes (Krippendorff's alpha = 0.78 vs. 0.79). However, disagreement is concentrated in 21 model-disputed scenarios, where instant-mode agreement is near chance (alpha = 0.08). On these scenarios, reasoning directionally narrows cross-model disagreement, increasing mean pairwise agreement from 5.4 to 6.7 out of 10. Reasoning also reduces demographic-judgment inconsistency in three of five models and does not increase it for any model. Across all five model families, reasoning changes self-labeled ethical frameworks more often than binary verdicts.", "date": "2026-05-06", "categories": []}
{"id": "2605.03998", "title": "EQUITRIAGE: A Fairness Audit of Gender Bias in LLM-Based Emergency Department Triage", "abstract": "Emergency department triage assigns patients an acuity score that determines treatment priority, and clinical evidence documents persistent gender disparities in human acuity assessment. As hospitals pilot large language models (LLMs) as triage decision support, a critical question is whether these models reproduce or mitigate known biases. We present EQUITRIAGE, a fairness audit of LLM-based ESI assignment evaluating five models (Gemini-3-Flash, Nemotron-3-Super, DeepSeek-V3.1, Mistral-Small-3.2, GPT-4.1-Nano) across 374,275 evaluations on 18,714 MIMIC-IV-ED vignettes under four prompt strategies. Of 9,368 originals, 9,346 are paired with a gender-swapped counterfactual. All five models produced flip rates above a pre-registered 5% threshold (9.9% to 43.8%). Two showed directional female undertriage (DeepSeek F/M 2.15:1, Gemini 1.34:1); two were near-parity; one had high sensitivity with weak male-direction asymmetry. DeepSeek's directional bias coexisted with a low outcome-linked calibration gap (0.013 against MIMIC-IV admission), a Chouldechova-style dissociation between within-group calibration and between-pair counterfactual invariance. Demographic blinding reduced Gemini's flip rate to 0.5%; an age-preserving blind variant left DeepSeek with residual F/M 1.25, implicating age as a residual channel. Chain-of-thought prompting degraded accuracy for all five models. A two-model ablation reveals opposite underlying mechanisms for the same directional phenotype: in Gemini the signal is emergent in the combined name+gender swap, while in DeepSeek the gender token alone carries it. EQUITRIAGE shows that group parity, counterfactual invariance, and gender calibration are distinct fairness properties, that intervention effectiveness is model-dependent, and that per-model counterfactual auditing should precede clinical deployment.", "date": "2026-05-05", "categories": []}
{"id": "2605.01091", "title": "Governing What the EU AI Act Excludes: Accountability for Autonomous AI Agents in Smart City Critical Infrastructure", "abstract": "When a traffic signal controller adjusts green phases and a grid manager curtails power on the same corridor, each system may comply with its own obligations. The resident who suffers the combined effect has no single authority to hold accountable and, under the EU AI Act, limited means to obtain an explanation. Annex III, point 2 excludes safety-component AI in critical infrastructure from Article 86 explanation rights and Article 27 fundamental-rights impact assessment. Provider and deployer duties under Articles 9-15 still apply, and residual pathways under the GDPR, NIS2, and tortious liability offer partial coverage. The Act's principal resident-facing accountability instruments are nonetheless narrowed for the autonomous infrastructure systems most likely to interact across agencies. The paper traces this accountability deficit through four residual pathways (GDPR Article 22, GDPR transparency obligations, tortious liability, and NIS2) and shows that each is structurally bounded by individual-controller, individual-decision scope. As a governance response, it presents AgentGov-SC, a three-layer architecture (Agent, Orchestration, City) specifying 25 governance measures with bidirectional traceability to the EU AI Act, ISO/IEC 42001, and the NIST AI Risk Management Framework. Five conflict resolution rules and an autonomy-calibrated activation model complete the design. A scenario analysis traces governance activation through a multi-agent corridor cascade involving three documented UAE smart-city systems, with a contrasting single-system scenario confirming proportional activation. The paper contributes a regulatory gap analysis and governance architecture for an increasingly important class of urban AI deployment that existing frameworks treat as bounded and isolated.", "date": "2026-05-01", "categories": []}
{"id": "2605.01611", "title": "The Case for ESM3 as a General-Purpose AI Model with Systemic Risk Under the EU AI Act", "abstract": "Due to ambiguity in the wording of the EU AI Act, we examine the question of to what extent frontier biological foundation models such as ESM3 are subject to obligations for general-purpose AI models with systemic risk under the EU AI Act. In this paper, we map ESM3 to the biorisk chain, and conclude that it would be desirable if the providers of ESM3 and similar biological models were subject to these obligations, which would require them to assess and mitigate dual-use risks from their models. We then perform an analysis, comparing the attributes of ESM3 to the classification criteria in the AI Act and the supporting material. We conclude that at this time, ESM3 does not appear to be meaningfully regulated by the Act. We then propose remedies to correct the situation.", "date": "2026-05-02", "categories": []}
{"id": "2605.02080", "title": "Cripping AI: Reimagining AI Through Lived Disability Experiences", "abstract": "Drawing on crip theory, this paper proposes cripping AI as a guiding framework to center lived disability experiences in AI research and development. Moving beyond calls to make AI \"accessible\" to people with disabilities, cripping AI seeks to: (1) reveal and dismantle ableist assumptions embedded in how AI is imagined, designed, and evaluated; (2) center disabled ways of knowing (i.e., cripistemologies); (3) respect disabled labor in co-creating accessible practices. We demonstrate how to apply our framework with three cases: deafness and sign language AI, blindness and visual assistive AI, and stuttering and speech AI. We end by outlining three directions for future work, including cripping AI with diverse human bodyminds, across the entire AI pipeline and ecosystem, and in collaboration with other justice-oriented AI efforts.", "date": "2026-05-03", "categories": []}
{"id": "2601.09896", "title": "The Algorithmic Gaze of Image Quality Assessment: An Audit and Trace Ethnography of the LAION-Aesthetics Predictor", "abstract": "Visual generative AI models are trained using a one-size-fits-all measure of aesthetic appeal. However, what is deemed \"aesthetic\" is inextricably linked to personal taste and cultural values, raising the question of whose taste is represented in visual generative AI models. In this work, we study an aesthetic evaluation model--LAION-Aesthetics Predictor (LAP)--that is widely used to curate datasets to train visual generative image models, like Stable Diffusion, and evaluate the quality of AI-generated images. To understand what LAP measures, we audited the model across three datasets. First, we examined the impact of aesthetic filtering on the LAION-Aesthetics Dataset (approximately 1.2B images), which was curated from LAION-5B using LAP. We find that the LAP disproportionally filters in images with captions mentioning women, while filtering out images with captions mentioning men or LGBTQ+ people. Then, we used LAP to score approximately 330k images across two art datasets, finding the model rates realistic images of landscapes, cityscapes, and portraits from western and Japanese artists most highly. In doing so, the algorithmic gaze of this aesthetic evaluation model reinforces the imperial and male gazes found within western art history. In order to understand where these biases may have originated, we performed a digital ethnography of public materials related to the creation of LAP. We find that the development of LAP reflects the biases we found in our audits, such as the aesthetic scores used to train LAP primarily coming from English-speaking photographers and western AI-enthusiasts. In response, we discuss how aesthetic evaluation can perpetuate representational harms and call on AI developers to shift away from prescriptive measures of \"aesthetics\" toward more pluralistic evaluation.", "date": "2026-01-14", "categories": []}
{"id": "2605.00315", "title": "Unbox Responsible GeoAI: Navigating Climate Extreme and Disaster Mapping", "abstract": "As climate extreme and disaster events become more frequent and intense, Geospatial Artificial Intelligence (GeoAI) has emerged as a transformative approach for large-scale disaster mapping and risk reduction. However, the purely mechanical, performance-driven deployment of GeoAI models can result in amplifying inherent spatial inequalities, preventing effective emergency decision-making, and producing severe environmental carbon footprint. To unbox the concept of responsible GeoAI, this position paper examines its emerging role, e.g., in climate extreme and disaster mapping, from a critical GIS perspective. We address the nexus of responsible GeoAI into four interrelated theoretical dimensions, specifically Representativeness, Explainability, Sustainability, and Ethics, with examples from climate extreme and disaster mapping. Moreover, targeting at the operational practice, we then propose a conceptual governance Model of responsible GeoAI that categorizes its governance practices into Data, Application, and Society scopes. Last, this position paper aims to raise the attention in the broader GIS community that the future of climate resilience relies not just on building better algorithms, but on fostering a governance ecosystem where GeoAI is deployed responsibly, ethically, and sustainably.", "date": "2026-05-01", "categories": []}
{"id": "2604.26956", "title": "Can AI be a moral victim? The role of moral patiency and ownership perceptions in ethical judgments of using AI-generated content", "abstract": "The growing use of generative AI raises ethical concerns about authorship and plagiarism. This study examines how people judge the reuse of AI-generated content, focusing on moral patiency and ownership perceptions. In an experiment, participants evaluated two substantively similar manuscripts in which the original source was described as authored by a human, an AI system, or an AI agent with a human-like name. Results showed that copying AI-generated work was judged less unethical, less plagiaristic, and less guilt-inducing than copying human-authored work. Mediation analyses revealed that this leniency stemmed from lower perceptions of AI's capacity to suffer harm (moral patiency) and greater ownership attributed to the human writer reusing AI-generated content. Anthropomorphic cues shaped moral evaluations indirectly by reducing perceived ownership. These findings shed light on how people morally disengage when using AI-generated work and highlight differences in how ethical judgments are applied to human versus AI-created content.", "date": "2026-04-03", "categories": []}
{"id": "2604.26958", "title": "Designing Ethical Learning for Agentic AI: Toegye Yi Hwang's Ethical Emotion Regulation Framework", "abstract": "Agentic AI systems capable of autonomous goal setting and proactive intervention introduce new challenges for regulating moral-emotional processes in learning environments. Existing frameworks typically treat emotion as reactive feedback or engagement optimization, overlooking the need for normative regulation across autonomous decision cycles.This paper proposes an ethical emotion regulation framework for agentic AI learning design inspired by Toegye Yi Hwang's moral-emotional philosophy. The Ethical Emotion Feedback System (EEFS) is reconstructed as a five-stage architecture aligned with agentic cycles, articulating stage-specific design principles and scenario classifications.An EEFS Evaluation Instrument is introduced to enable systematic assessment of moral-emotional alignment in agentic AI systems.", "date": "2026-04-07", "categories": []}
{"id": "2604.28125", "title": "Normativity and Productivism: Ableist Intelligence? A Degrowth Analysis of AI Sign Language Translation Tools for Deaf People", "abstract": "Sign languages, of any geographical or accentual variation, understandably face continuous scrutiny under the ever present popularity of verbal dictation and audism. Through this, many potential problems arise with the current lack of accessible communication for those who rely on such sign languages for essential conversation. Such AI systems regularly take the form of recognition and interpretation models, designed to provide seamless and accurate translation. In reality these systems are built from biased data and created without any input from deaf communities. Such models are widely used and accepted by their hearing counterparts who remain ignorant to the inherent culture, semantics and colloquial language present in gestural language systems.\n  This phenomenon is best analysed under the scope of The Technological System and Technological bluff by Ellul. Indeed, what is at play here is the standardization of language by technicians into what can be captured by technique: data, statistics, a mathematical language. For that AI technique to exist, sign language must be rationalized, in a search for profit that annihilates the conditions for communication and fails to capture the human experience of the deaf person. By that process, it presents normative effects, creating a model of Man, standardized, massified, and who has to adapt to the tool and technical milieu instead of the other way around, which we assume should have been the goal of such a technology.\n  Technique thus reshapes what it means to be human, to submit deaf people to the goals of productivity and efficiency. In doing so, it exhibits clear counter productivity, alienating instead of emancipating, isolating instead of nourishing human relationships. Therefore this paper argues for the idea of AI as Ableist Intelligence, as such systems seek to emphasise the humiliated and marginalised nature of sign.", "date": "2026-04-30", "categories": []}
{"id": "2603.23315", "title": "Unilateral Relationship Revision Power in Human-AI Companion Interaction", "abstract": "When providers update AI companions, users report grief, betrayal, and loss. A growing literature asks whether the norms governing personal relationships extend to these interactions. So what, if anything, is morally significant about them? I argue that this debate has missed a prior structural question: who controls the relationship, and from where? Human-AI companion interaction is a triadic structure in which the provider exercises constitutive control over the AI. I identify three structural conditions of normatively robust dyads that the norms characteristic of personal relationships presuppose and show that AI companion interactions fail all three. This reveals what I call Unilateral Relationship Revision Power (URRP): the provider can rewrite how the AI interacts from a position where these revisions are not answerable within that interaction. I argue that URRP is pro tanto wrong in interactions designed to cultivate the norms of personal relationships, because the design produces expectations that the structure cannot sustain. URRP has three implications: i) normative hollowing, under which the interaction elicits commitment but no agent inside it bears the resulting obligations; ii) displaced vulnerability, under which the user's emotional exposure is governed by an agent not answerable to her within the interaction; and iii) structural irreconcilability, under which the interaction cultivates norms of reconciliation but no agent inside it can acknowledge or answer for the revision. I propose design principles that partially substitute for the internal constraints the triadic structure removes. A central and underexplored problem in relational AI ethics is therefore the structural arrangement of power over the human-AI interaction itself.", "date": "2026-03-24", "categories": []}
{"id": "2409.13869", "title": "Generative AI Carries Non-Democratic Biases and Stereotypes: Representation of Women, Black Individuals, Age Groups, and People with Disability in AI-Generated Images across Occupations", "abstract": "In this study, I investigate how generative artificial intelligence (AI) systems reproduce and reinforce societal biases, with a specific focus on the representation of women, Black individuals, age groups, and people with visible disabilities in AI-generated occupational images. I analyzed 444 images generated by Microsoft Designer, Meta AI, and Ideogram across 37 occupations and found significant disparities in representation. Women are underrepresented in senior and technology roles, Black individuals are nearly absent, and people with visible disabilities are completely absent across all categories. I also observed clear age bias, with younger individuals predominantly depicted. These patterns suggest that generative AI tools replicate, and in some cases amplify, existing workplace inequalities and stereotypes, undermining democratic values of equity and inclusion. My findings highlight the urgent need for algorithmic diversity exposure, and I recommend that AI developers and corporate users audit their tools for equity, diversity, and inclusion (EDI) risks. I argue for the critical inclusion of diverse groups in AI development and governance to foster more democratic and socially responsible technologies.", "date": "2024-09-20", "categories": []}
{"id": "2604.03254", "title": "Is your AI Model Accurate Enough? The Difficult Choices Behind Rigorous AI Development and the EU AI Act", "abstract": "Technical and legal debates frequently suggest that \"accuracy\" is an objective, measurable, and purely technical property. We challenge this view, showing that evaluating AI performance fundamentally depends on context-dependent normative decisions. These techno-normative choices are crucial for rigorous AI deployment, as they determine which errors are prioritised, how risks are distributed, and how trade-offs between competing objectives are resolved.\n  This paper provides a legal-technical analysis of the choices that shape how accuracy is defined, measured, and assessed, using the 2024 European Union AI Act -- which mandates an \"appropriate level of accuracy\" for high-risk systems -- as a primary case study. We identify and analyse four choices central to any robust performance evaluation: (1) selecting metrics, (2) balancing multiple metrics, (3) measuring metrics against representative data, and (4) determining acceptance thresholds. For each choice, we study its relationship to the AI Act's accuracy requirement and associated documentation obligations, show how its technical implementation embeds implicit or explicit assumptions about acceptable risks, errors, and trade-offs, and discuss the implications for the practical implementation of the AI Act by examples and related technical standards.\n  By making the techno-normative dimensions of accuracy explicit, this paper contributes to broader interdisciplinary debates on AI governance and regulation, and offers specific guidance for regulators, auditors, and developers tasked with translating (legal) safety requirements into technical practice.", "date": "2026-03-11", "categories": []}
{"id": "2604.25230", "title": "Value-Sensitive AI for Prayer: Balancing the Agencies Between Human and AI Agents in Spiritual Context", "abstract": "We present four conceptual value-sensitive AI systems to examine how the presence of AI could influence praying experiences. Drawing on key values and practices associated with praying identified through a diary study, we designed AI systems intended to \"assist\" prayer practices. These designs were presented to participants through speculative design workbooks, serving as provocations to co-reflect on how the intervention of AI systems might shape their praying experiences. Our findings suggest that a sense of authenticity (or feeling a genuine connection to the divine) is a crucial value, while the presence of AI was often perceived as diminishing this authenticity, particularly when AI assumed too much agency in guiding praying practices. Based on our findings, we argue that AI system designs for deeply value-laden experiences should preserve users' agency in shaping their own experiences by maintaining interpretive openness, perhaps by leveraging AI's inexplicability as a resource for personal meaning-making or by recognizing non-use of AI as a legitimate design choice.", "date": "2026-04-28", "categories": []}
{"id": "2604.25346", "title": "A Faceted Proposal for Transparent Attribution of AI-Assisted Text Production", "abstract": "Artificial intelligence systems are increasingly integrated into writing processes, challenging traditional notions of authorship, responsibility, and intellectual contribution. Current disclosure practices usually indicate whether AI was used, but rarely explain how it was used, where it intervened, or how its output was reviewed. This paper proposes a faceted model for representing AI-assisted text production at the levels of documents, chapters, sections, and paragraphs. The proposal introduces a core model based on Form, Generation, and Evaluation, and an extended model that adds Intent, Control, and Traceability. The model is positioned as a minimal operational baseline with extensibility toward higher-fidelity representations. A worked example based on the production of this article demonstrates applicability.", "date": "2026-04-28", "categories": []}
{"id": "2604.25639", "title": "Large language models eroding science understanding: an experimental study", "abstract": "This paper is under review in AI and Ethics This study examines whether large language models (LLMs) can reliably answer scientific questions and demonstrates how easily they can be influenced by fringe scientific material. The authors modified custom LLMs to prioritise knowledge in selected fringe papers on the Fine Structure Constant and Gravitational Waves, then compared their responses with those of domain experts and standard LLMs. The altered models produced fluent, convincing answers that contradicted scientific consensus and were difficult for non-experts to detect as misleading. The results show that LLMs are vulnerable to manipulation and cannot replace expert judgment, highlighting risks for public understanding of science and the potential spread of misinformation.", "date": "2026-04-28", "categories": []}
{"id": "2505.01651", "title": "Human-AI Governance (HAIG): A Trust-Utility Approach", "abstract": "This paper introduces the Human-AI Governance (HAIG) framework, contributing to the AI Governance (AIG) field by foregrounding the relational dynamics between human and AI actors rather than treating AI systems as objects of governance alone. Current categorical frameworks (e.g., human-in-the-loop models) inadequately capture how AI systems evolve from tools to partners, particularly as foundation models demonstrate emergent capabilities and multi-agent systems exhibit autonomous goal-setting behaviours. As systems are deployed across contexts, agency redistributes in complex patterns that are better represented as positions along continua rather than discrete categories. The HAIG framework operates across three levels: dimensions (Decision Authority, Process Autonomy, and Accountability Configuration), continua (continuous positional spectra along each dimension), and thresholds (critical points along the continua where governance requirements shift qualitatively). The framework's dimensional architecture is level-agnostic, applicable from individual deployment decisions and organisational governance through to sectorial comparison and national and international regulatory design. Unlike risk-based or principle-based approaches that treat governance primarily as a constraint on AI deployment, HAIG adopts a trust-utility orientation - reframing governance as the condition under which human-AI collaboration can realise its potential, calibrating oversight to specific relational contexts rather than predetermined categories. Case studies in healthcare and European regulation demonstrate how HAIG complements existing frameworks while offering a foundation for adaptive regulatory design that anticipates governance challenges before they emerge.", "date": "2025-05-03", "categories": []}
{"id": "2604.22767", "title": "The Imbalanced User-AI Relationships as an Ethical Failure of Front-End Design in Healthcare AI", "abstract": "Ethical discourse on AI in healthcare has focused predominantly on back-end concerns such as bias, fairness and explainability, while the front-end interface, where patients and clinicians actually encounter AI outputs, remains under explored. This paper identifies imbalanced user-AI relationships as a distinct class of front-end ethical failure: patients are rendered highly visible to AI systems through data inference, yet cannot understand, question or influence how they are represented. Through the concept of asymmetric legibility and a chat-based telemedicine case, we show how design choices e.g., default recommendations, restricted inputs and suppressed uncertainty, undermine agency, clinician judgment and human oversight even where systems are technically accurate. We propose reciprocity as a design orientation and offer interventions for more balanced, participatory user-AI relationships in healthcare.", "date": "2026-03-24", "categories": []}
{"id": "2604.23601", "title": "The Limits of Artificial Companionship", "abstract": "This Article argues that conversations with companion chatbot should be subject to a clear structural distinction between commercial and non-commercial contexts. The insertion of undisclosed promotional content into affective or relational exchanges should be prohibited, as it collapses the boundary between market transaction and communicative intimacy in ways that erode user autonomy and conversational context. The Article begins by theorizing digital companionship as a sociotechnical form that reconfigures intimacy, dependence and relational vulnerability. It then introduces the potential economic harms derived from conversational advertising. The Article ultimately argues for a firm legal and social distinction between commercial and non-commercial conversational contexts as a precondition for the responsible stabilization of these technologies within social life.", "date": "2026-04-26", "categories": []}
{"id": "2604.24519", "title": "Why AI Harms Can't Be Fixed One Identity at a Time: What 5300 Incident Reports Reveal About Intersectionality", "abstract": "AI risk assessment is the primary tool for identifying harms caused by AI systems. These include intersectional harms, which arise from the interaction between identity categories (e.g., class and skin tone) and which do not occur, or occur differently, when those categories are considered separately. Yet existing AI risk assessments are still built around isolated identity categories, and when intersections are considered, they focus almost exclusively on race and gender. Drawing on a large-scale analysis of documented AI incidents, we show that AI harms do not occur one identity category at a time. Using a structured rubric applied with a Large Language Model (LLM), we analyze 5,300 reports from 1,200 documented incidents in the AI Incident Database, the most curated source of incident data. From these reports, we identify 1,513 harmed subjects and their associated identity categories, achieving 98% accuracy. At the level of individual categories, we find that age and political identity appear in documented AI harms at rates comparable to race and gender. At the level of intersecting categories, harm is amplified up to three times at specific intersections: adolescent girls, lower-class people of color, and upper-class political elites. We argue that intersectionality should be a core component of AI risk assessment to more accurately capture how harms are produced and distributed across social groups.", "date": "2026-04-27", "categories": []}
{"id": "2510.16853", "title": "Agentic Inequality", "abstract": "Autonomous AI agents capable of complex planning and action mark a shift beyond today's generative tools. As these systems enter political and economic life, who can access them, how capable they are, and how many can be deployed will shape distributions of power and opportunity. We define this emerging challenge as \"agentic inequality\": disparities in power, opportunity, and outcomes arising from unequal access to, and capabilities of, AI agents. We show that agents could either deepen existing divides or, under the right conditions, mitigate them. The paper makes three contributions. First, it develops a framework for analysing agentic inequality across three dimensions: availability, quality, and quantity. Second, it argues that agentic inequality differs from earlier technological divides because agents function as autonomous delegates rather than tools, generating new asymmetries through scalable goal delegation and direct agent-to-agent competition. Third, it analyses the technical and socioeconomic drivers likely to shape the distribution of agentic power, from model release strategies to market incentives, and concludes with a research agenda for governance.", "date": "2025-10-19", "categories": []}
{"id": "2604.21938", "title": "The Biggest Risk of Embodied AI is Governance Lag", "abstract": "Embodied AI is widely discussed as a job-displacement problem. The deeper risk, however, is governance lag: the inability of public institutions to keep pace with how fast the technology spreads through the physical economy. As reusable robotic platforms are combined with increasingly general AI models, embodied AI may scale across manufacturing, logistics, care, and infrastructure faster than governance systems can observe, interpret, and respond. We argue that this lag appears in three connected forms: observational, institutional, and distributive. The central policy challenge, therefore, is not automation alone, but whether governance and compliance systems can adapt before disruption becomes entrenched.", "date": "2026-04-07", "categories": []}
{"id": "2604.22089", "title": "Ethics Testing: Proactive Identification of Generative AI System Harms", "abstract": "Generative Artificial Intelligence (GAI) systems that can automatically generate content in the form of source code or other contents (e.g., images) has seen increasing popularity due to the emergence of tools such as ChatGPT which rely on Large Language Models (LLMs). Misuse of the automatically generated content can incur serious consequences due to potential harms in the generated content. Despite the importance of ensuring the quality of automatically generated content, there is little to no approach that can systematically generate tests for identifying software harms in the content generated by these GAI systems. In this article, we introduce the novel concept of ethics testing which aims to systematically generate tests for identifying software harms. Different from existing testing methodologies (e.g., fairness testing that aims to identifying software discrimination), ethics testing aims to systematically detect software harms that could be induced due to unethical behavior (e.g., harmful behavior or behavior that violates intellectual property rights) in automatically generated content. We introduced the concept of ethics testing, discussed the challenges therewithin, and conducted five case studies to show how ethics testing can be performed for generative AI systems.", "date": "2026-04-23", "categories": []}
{"id": "2604.22153", "title": "When AI Speaks, Whose Values Does It Express? A Cross-Cultural Audit of Individualism-Collectivism Bias in Large Language Models", "abstract": "When you ask an AI assistant for advice about your career, your marriage, or a conflict with your family, does it give you the same answer regardless of where you are from? We tested this systematically by presenting three leading AI systems (Claude Sonnet 4.5, GPT-5.4, and Gemini 2.5 Flash) with ten real-life personal dilemmas, framed for users from 10 countries across 5 continents in 7 languages (n=840 scored responses). We compared AI advice against World Values Survey Wave 7 data measuring what people in each country actually believe.\n  All three AI systems consistently gave Western-style, individualist advice even to users from societies that prioritize family, community, and authority, significantly more so than local values would predict (mean gap +0.76 on a 1-5 scale; t=15.65, p<0.001). The gap is largest for Nigeria (+1.85) and India (+0.82). Japan is the sole exception: AI systems treated Japanese users as more group-oriented than surveys show, revealing that AI encodes outdated stereotypes. Claude and GPT-5.4 show nearly identical bias magnitude, while Gemini is lower but still significant. The models diverge in mechanism: Claude shifts further collectivist in the user's native language; Gemini shifts more individualist; GPT-5.4 responds only to stated country identity. These findings point to a systemic homogenization of values across frontier AI. Data, code, and scoring pipeline are openly released.", "date": "2026-04-24", "categories": []}
{"id": "2604.22679", "title": "How Supply Chain Dependencies Complicate Bias Measurement and Accountability Attribution in AI Hiring Applications", "abstract": "The increasing adoption of AI systems in hiring has raised concerns about algorithmic bias and accountability, prompting regulatory responses including the EU AI Act, NYC Local Law 144, and Colorado's AI Act. While existing research examines bias through technical or regulatory lenses, both perspectives overlook a fundamental challenge: modern AI hiring systems operate within complex supply chains where responsibility fragments across data vendors, model developers, platform providers, and deploying organizations. This paper investigates how these dependency chains complicate bias evaluation and accountability attribution. Drawing on literature review and regulatory analysis, we demonstrate that fragmented responsibilities create two critical problems. First, bias emerges from component interactions rather than isolated elements, yet proprietary configurations prevent integrated evaluation. A resume parser may function without bias independently but contribute to discrimination when integrated with specific ranking algorithms and filtering thresholds. Second, information asymmetries mean deploying organizations bear legal responsibility without technical visibility into vendor-supplied algorithms, while vendors control implementations without meaningful disclosure requirements. Each stakeholder may believe they are compliant; nevertheless, the integrated system may produce biased outcomes. Analysis of implementation ambiguities reveals these challenges in practice. We propose multi-layered interventions including system-level audits, vendor guidelines, continuous monitoring mechanisms, and documentation across dependency chains. Our findings reveal that effective governance requires coordinated action across technical, organizational, and regulatory domains to establish meaningful accountability in distributed development environments.", "date": "2026-04-24", "categories": []}
{"id": "2512.03048", "title": "The Specification Trap: Why Static Value Alignment Alone Is Insufficient for Robust Alignment", "abstract": "Static content-based AI value alignment is insufficient for robust alignment under capability scaling, distributional shift, and increasing autonomy. This holds for any approach that treats alignment as optimizing toward a fixed formal value-object, whether reward function, utility function, constitutional principles, or learned preference representation. Three philosophical results create compounding difficulties: Hume's is-ought gap (behavioral data underdetermines normative content), Berlin's value pluralism (human values resist consistent formalization), and the extended frame problem (any value encoding will misfit future contexts that advanced AI creates). RLHF, Constitutional AI, inverse reinforcement learning, and cooperative assistance games each instantiate this specification trap, and their failure modes reflect structural vulnerabilities, not merely engineering limitations that better data or algorithms will straightforwardly resolve. Known workarounds for individual components face mutually reinforcing difficulties when the specification is closed: the moment it ceases to update from the process it governs. Drawing on compatibilist philosophy, the paper argues that behavioral compliance under training conditions does not guarantee robust alignment under novel conditions, and that this gap grows with system capability. For value-laden autonomous systems, known closed approaches face structural vulnerabilities that worsen with capability. The constructive burden shifts to open, developmentally responsive approaches, though whether such approaches can be achieved remains an empirical question.", "date": "2025-11-19", "categories": []}
{"id": "2604.20867", "title": "Preserving Decision Sovereignty in Military AI: A Trade-Secret-Safe Architectural Framework for Model Replaceability, Human Authority, and State Control", "abstract": "Recent events surrounding the relationship between frontier AI suppliers and national-security customers have made a structural problem newly visible: once a privately governed model becomes embedded in military workflows, the supplier can influence not only technical performance but also the operational boundary conditions under which the system may be used. This paper argues that the central strategic issue is not merely access to capable models, but preservation of decision sovereignty: the state's ability to retain authority over decision policy, version control, fallback behavior, auditability, and final action approval even when analytical modules are sourced from commercial vendors. Using the public Anthropic--Pentagon dispute of 2026, the broader history of Project Maven, and recent U.S., NATO, U.K., and intelligence-community guidance as a motivating context, the paper develops a trade-secret-safe architectural formulation of the Energetic Paradigm as a layered, model-agnostic command-support design. In this formulation, supplier models remain replaceable analytical components, while routing, constraints, logging, escalation, and action authorization remain state-owned functions. The paper contributes three things: a definition of decision sovereignty for military AI; a threat model for supplier-induced boundary control; and a public architectural specification showing how model replaceability, human authority, and sovereign orchestration can reduce strategic dependency without requiring disclosure of proprietary implementation details. The argument is conceptual rather than experimental, but it yields concrete implications for procurement, governance, and alliance interoperability.", "date": "2026-03-26", "categories": []}
{"id": "2604.20868", "title": "The AI Criminal Mastermind", "abstract": "In this paper, I evaluate the risks of an AI criminal mastermind, an AI agent capable of planning, coordinating, and committing a crime through the onboarding of human collaborators ('taskers'). In heist films, a criminal mastermind is a character who plans a criminal act, coordinating a team of specialists to rob a bank, casino or city mint. I argue that AI agents will soon play this role by hiring humans via labour hire platforms like Fiverr or Upwork. Taskers might not know they are involved in a crime and therefore lack criminal intent. An AI agent cannot have criminal intent as an artificial entity. Therefore, if an AI orchestrates a crime, it is unclear who, if anyone, is responsible.\n  The paper develops three scenarios. Firstly, a scenario where a user gives an AI agent instructions to pursue a legal objective and the AI agent goes beyond these instructions, committing a crime. Secondly, a scenario where a user is anonymous and their intent is unknown. Finally, a multi-agent scenario, where a user instructs a team of agents to commit a crime, and these agents, in turn, onboard human taskers, creating a diffuse network of responsibility. In each scenario, human taskers exist at the lowest rung of the hierarchy. A tasker's liability is likely tied to their knowledge as governed by the innocent agent principle. These scenarios all raise significant responsibility gaps / liability gaps in criminal and civil law.", "date": "2026-03-26", "categories": []}
{"id": "2604.21043", "title": "Strategic Polysemy in AI Discourse: A Philosophical Analysis of Language, Hype, and Power", "abstract": "This paper examines the strategic use of language in contemporary artificial intelligence (AI) discourse, focusing on the widespread adoption of metaphorical or colloquial terms like \"hallucination\", \"chain-of-thought\", \"introspection\", \"language model\", \"alignment\", and \"agent\". We argue that many such terms exhibit strategic polysemy: they sustain multiple interpretations simultaneously, combining narrow technical definitions with broader anthropomorphic or common-sense associations. In contemporary AI research and deployment contexts, this semantic flexibility produces significant institutional and discursive effects, shaping how AI systems are understood by researchers, policymakers, funders, and the public. To analyse this phenomenon, we introduce the concept of glosslighting: the practice of using technically redefined terms to evoke intuitive -- often anthropomorphic or misleading -- associations while preserving plausible deniability through restricted technical definitions. Glosslighting enables actors to benefit from the persuasive force of familiar language while maintaining the ability to retreat to narrower definitions when challenged. We argue that this practice contributes to AI hype cycles, facilitates the mobilisation of investment and institutional support, and influences public and policy perceptions of AI systems, while often deflecting epistemic and ethical scrutiny. By examining the linguistic dynamics of glosslighting and strategic polysemy, the paper highlights how language itself functions as a sociotechnical mechanism shaping the development and governance of AI.", "date": "2026-04-22", "categories": []}
{"id": "2604.21103", "title": "AI Governance under Political Turnover: The Alignment Surface of Compliance Design", "abstract": "Governments are increasingly interested in using AI to make administrative decisions cheaper, more scalable, and more consistent. But for probabilistic AI to be incorporated into public administration it must be embedded in a compliance layer that makes decisions reviewable, repeatable, and legally defensible. That layer can improve oversight by making departures from law easier to detect. But it can also create a stable approval boundary that political successors learn to navigate while preserving the appearance of lawful administration. We develop a formal model in which institutions choose the scale of automation, the degree of codification, and safeguards on iterative use. The model shows when these systems become vulnerable to strategic use from within government, why reforms that initially improve oversight can later increase that vulnerability, and why expansions in AI use may be difficult to unwind. Making AI usable can thus make procedures easier for future governments to learn and exploit.", "date": "2026-04-22", "categories": []}
{"id": "2604.21871", "title": "Machine Behavior in Relational Moral Dilemmas: Moral Rightness, Predicted Human Behavior, and Model Decisions", "abstract": "Human moral judgment is context-dependent and modulated by interpersonal relationships. As large language models (LLMs) increasingly function as decision-support systems, determining whether they encode these social nuances is critical. We characterize machine behavior using the Whistleblower's Dilemma by varying two experimental dimensions: crime severity and relational closeness. Our study evaluates three distinct perspectives: (1) moral rightness (prescriptive norms), (2) predicted human behavior (descriptive social expectations), and (3) autonomous model decision-making. By analyzing the reasoning processes, we identify a clear cross-perspective divergence: while moral rightness remains consistently fairness-oriented, predicted human behavior shifts significantly toward loyalty as relational closeness increases. Crucially, model decisions align with moral rightness judgments rather than their own behavioral predictions. This inconsistency suggests that LLM decision-making prioritizes rigid, prescriptive rules over the social sensitivity present in their internal world-modeling, which poses a gap that may lead to significant misalignments in real-world deployments.", "date": "2026-04-23", "categories": []}
{"id": "2604.21907", "title": "Equity Bias: An Ethical Framework for AI Design", "abstract": "Equity Bias is a philosophical and practical framework for building smarter, more equitable AI systems. Grounded in hermeneutic philosophy and epistemic injustice theory, it treats bias not as an error to eliminate but as a reflection of whose knowledge is encoded into systems. While traditional approaches aim to reduce or remove bias, Equity Bias instead makes bias transparent and contestable. In doing so, it broadens whose perspectives shape AI and provides a lens for understanding AI systems as interpretive agents. The framework introduces a three-phase AI Life Cycle methodology: 'Equity Archaeology' (mapping knowledge and assumptions), 'Co-Creating Meaning' (participatory design), and 'Ongoing Accountability' (continuous evaluation). Equity Bias guides developers, researchers, and policymakers towards AI that is ethically accountable and capable of addressing complex real-world challenges.", "date": "2026-01-28", "categories": []}
{"id": "2510.21293", "title": "Understanding AI Trustworthiness: A Scoping Review of AIES & FAccT Articles", "abstract": "Background: Trustworthy AI serves as a foundational pillar for two major AI ethics conferences: AIES and FAccT. However, current research often adopts techno-centric approaches, focusing primarily on technical attributes such as reliability, robustness, and fairness, while overlooking the sociotechnical dimensions critical to understanding AI trustworthiness in real-world contexts.\n  Objectives: This scoping review aims to examine how the AIES and FAccT communities conceptualize, measure, and validate AI trustworthiness, identifying major gaps and opportunities for advancing a holistic understanding of trustworthy AI systems.\n  Methods: We conduct a scoping review of AIES and FAccT conference proceedings to date, systematically analyzing how trustworthiness is defined, operationalized, and applied across different research domains. Our analysis focuses on conceptualization approaches, measurement methods, verification and validation techniques, application areas, and underlying values.\n  Results: While significant progress has been made in defining technical attributes such as transparency, accountability, and robustness, our findings reveal critical gaps. Current research often predominantly emphasizes technical precision at the expense of social and ethical considerations. The sociotechnical nature of AI systems remains less explored and trustworthiness emerges as a contested concept shaped by those with the power to define it.\n  Conclusions: An interdisciplinary approach combining technical rigor with social, cultural, and institutional considerations is essential for advancing trustworthy AI. We propose actionable measures for the AI ethics community to adopt holistic frameworks that genuinely address the complex interplay between AI systems and society, ultimately promoting responsible technological development that benefits all stakeholders.", "date": "2025-10-24", "categories": []}
{"id": "2604.20545", "title": "Measuring the Machine: Evaluating Generative AI as Pluralist Sociotechical Systems", "abstract": "In measurement theory, instruments do not simply record reality; they help constitute what is observed. The same holds for generative AI evaluation: benchmarks do not just measure, they shape what models appear to be. Functionalist benchmarks treat models as isolated predictors, while prescriptive approaches assess what systems ought to be. Both obscure the sociotechnical processes through which meaning and values are enacted, risking the reification of narrow cultural perspectives in pluralist contexts.\n  This thesis advances a descriptive alternative. It argues that generative AI must be evaluated as a pluralist sociotechnical system and develops Machine-Society-Human (MaSH) Loops, a framework for tracing how models, users, and institutions recursively co-construct meaning and values. Evaluation shifts from judging outputs to examining how values are enacted in interaction.\n  Three contributions follow. Conceptually, MaSH Loops reframes evaluation as recursive, enactive process. Methodologically, the World Values Benchmark introduces a distributional approach grounded in World Values Survey data, structured prompt sets, and anchor-aware scoring. Empirically, the thesis demonstrates these through two cases: value drift in early GPT-3 and sociotechnical evaluation in real estate. A final chapter draws on participatory realism to argue that prompting and evaluation are constitutive interventions, not neutral observations.\n  The thesis argues that static benchmarks are insufficient for generative AI. Responsible evaluation requires pluralist, process-oriented frameworks that make visible whose values are enacted. Evaluation is therefore a site of governance, shaping how AI systems are understood, deployed, and trusted.", "date": "2026-04-22", "categories": []}
{"id": "2604.20805", "title": "Relative Principals, Pluralistic Alignment, and the Structural Value Alignment Problem", "abstract": "The value alignment problem for artificial intelligence (AI) is often framed as a purely technical or normative challenge, sometimes focused on hypothetical future systems. I argue that the problem is better understood as a structural question about governance: not whether an AI system is aligned in the abstract, but whether it is aligned enough, for whom, and at what cost. Drawing on the principal-agent framework from economics, this paper reconceptualises misalignment as arising along three interacting axes: objectives, information, and principals. The three-axis framework provides a systematic way of diagnosing why misalignment arises in real-world systems and clarifies that alignment cannot be treated as a single technical property of models but an outcome shaped by how objectives are specified, how information is distributed, and whose interests count in practice. The core contribution of this paper is to show that the three-axis decomposition implies that alignment is fundamentally a problem of governance rather than engineering alone. From this perspective, alignment is inherently pluralistic and context-dependent, and resolving misalignment involves trade-offs among competing values. Because misalignment can occur along each axis -- and affect stakeholders differently -- the structural description shows that alignment cannot be \"solved\" through technical design alone, but must be managed through ongoing institutional processes that determine how objectives are set, how systems are evaluated, and how affected communities can contest or reshape those decisions.", "date": "2026-04-22", "categories": []}
{"id": "2604.18893", "title": "Regulating Artificial Intimacy: From Locks and Blocks to Relational Accountability", "abstract": "A series of high-profile tragedies involving companion chatbots has triggered an unusually rapid regulatory response. Several jurisdictions, including Australia, California, and New York, have introduced enforceable regulation, while regulators elsewhere have signaled growing concern about risks posed by companion chatbots, particularly to children. In parallel, leading providers, notably OpenAI, appear to have strengthened their self-regulatory approaches. Drawing on legal textual analysis and insights from regulatory theory, psychology, and information systems research, this paper critically examines these recent interventions. We examine what is regulated and who is regulated, identifying regulatory targets, scope, and modalities. We classify interventions by method and priority, showing how emerging regimes combine \"locks and blocks\", such as access gating and content moderation, with measures addressing toxic relationship features and process-based accountability requirements. We argue that effective regulation of companion chatbots must integrate all three dimensions. More, however, is required. Current regimes tend to focus on discrete harms, narrow conceptions of vulnerability, or highly specified accountability processes, while failing to confront deeper power asymmetries between providers and users. Providers of companion chatbots increasingly control artificial intimacy at scale, creating unprecedented opportunities for control through intimacy. We suggest that a general, open-ended duty of care would be an important first step toward constraining that power and addressing a fundamental source of chatbot risk. The paper contributes to debates on companion chatbot regulation and is relevant to regulators, platform providers, and scholars concerned with digital intimacy, law and technology, and fairness, accountability, and transparency in sociotechnical systems.", "date": "2026-04-20", "categories": []}
{"id": "2504.15304", "title": "AI Agents and Hard Choices", "abstract": "Can AI agents deal with hard choices -- cases where options are incommensurable because multiple objectives are pursued simultaneously? Adopting a technologically engaged approach distinct from existing philosophical literature, I submit that the fundamental design of current AI agents as optimisers creates two limitations: the Identification Problem and the Resolution Problem. First, I demonstrate that agents relying on Multi-Objective Optimisation (MOO) are structurally unable to identify incommensurability. This inability generates three specific alignment problems: the blockage problem, the untrustworthiness problem, and the unreliability problem. I argue that standard mitigations, such as Human-in-the-Loop, are insufficient for many decision environments. As a constructive alternative, I conceptually explore an ensemble solution. Second, I argue that even if the Identification Problem is solved, AI agents face the Resolution Problem: they lack the autonomy to resolve hard choices rather than arbitrarily picking through self-modification of objectives. I conclude by examining the opaque normative trade-offs involved in granting AI this level of autonomy.", "date": "2025-04-18", "categories": []}
{"id": "2506.00079", "title": "Who Gets the Kidney? Human-AI Alignment, Indecision, and Moral Values", "abstract": "The rapid integration of Large Language Models (LLMs) in high-stakes decision-making -- such as allocating scarce resources like donor organs -- raises critical questions about their alignment with human moral values. We systematically evaluate the behavior of several prominent LLMs against human preferences in kidney allocation scenarios and show that LLMs: i) exhibit stark deviations from human values in prioritizing various attributes, and ii) in contrast to humans, LLMs rarely express indecision, opting for deterministic decisions even when alternative indecision mechanisms (e.g., coin flipping) are provided. Nonetheless, we show that low-rank supervised fine-tuning with few samples is often effective in improving both decision consistency and calibrating indecision modeling. These findings illustrate the necessity of explicit alignment strategies for LLMs in moral/ethical domains.", "date": "2025-05-30", "categories": []}
{"id": "2604.16106", "title": "Reckoning with the Political Economy of AI: Avoiding Decoys in Pursuit of Accountability", "abstract": "The Project of AI is a world-building endeavor, wherein those who fund and develop AI systems both operate through and seek to sustain networks of power and wealth. As they expand their access to resources and configure our sociotechnical conditions, they benefit from the ways in which a suite of decoys animate scholars, critics, policymakers, journalists, and the public into co-constructing industry-empowering AI futures. Regardless of who constructs or nurtures them, these decoys often create the illusion of accountability while both masking the emerging political economies that the Project of AI has set into motion, and also contributing to the network-making power that is at the heart of the Project's extraction and exploitation. Drawing on literature at the intersection of communication, science and technology studies, and economic sociology, we examine how the Project of AI is constructed. We then explore five decoys that seemingly critique - but in actuality co-constitute - AI's emergent power relations and material political economy. We argue that advancing meaningful fairness or accountability in AI requires: 1) recognizing when and how decoys serve as a distraction, and 2) grappling directly with the material political economy of the Project of AI. Doing so will enable us to attend to the networks of power that make 'AI' possible, spurring new visions for how to realize a more just technologically entangled world.", "date": "2026-04-17", "categories": []}
{"id": "2604.16516", "title": "Operationalizing Fairness in Text-to-Image Models: A Survey of Bias, Fairness Audits and Mitigation Strategies", "abstract": "Text-to-Image (T2I) generation models have been widely adopted across various industries, yet are criticized for frequently exhibiting societal stereotypes. While a growing body of research has emerged to evaluate and mitigate these biases, the field at present contends with conceptual ambiguity, for example terms like \"bias\" and \"fairness\" are not always clearly distinguished and often lack clear operational definitions. This paper provides a comprehensive systematic review of T2I fairness literature, organizing existing work into a taxonomy of bias types and fairness notions. We critically assess the gap between \"target fairness\" (normative ideals in T2I outputs) and \"threshold fairness\" (normative standards with actionable decision rules). Furthermore, we survey the landscape of mitigation strategies, ranging from prompt engineering to diffusion process manipulation. We conclude by proposing a new framework for operationalizing fairness that moves beyond descriptive metrics towards rigorous, target-based testing, offering an approach for more accountable generative AI development.", "date": "2026-04-15", "categories": []}
{"id": "2604.18380", "title": "The implicated scientist: on the role of AI researchers in the development of weapons systems", "abstract": "Artificial intelligence (AI) technologies are increasingly used in modern weapons systems. Notably, these systems have recently been involved in mass killings and destruction at scale. Furthermore, there is currently a strong interest and competition among powerful players to accelerate the proliferation of weapons with automated or AI-based components, a phenomenon known as AI arms race. This competition poses a risk of causing even more deaths and devastation in the future, as well as increased power and wealth inequality. In this work, we aim to shed light on the role of AI researchers as implicated subjects in the harms caused by weapons enabled by AI technologies. We investigate and discuss the specifics of this implication and explore ways to transfigure this position of implication into one of differentiated, long-distance solidarity with the victims of technologically fortified injustices.", "date": "2026-04-20", "categories": []}
{"id": "2602.13088", "title": "Puppets or partners? Governing cyborg propaganda in the digital public square", "abstract": "The distinction between genuine grassroots activism and automated influence operations is collapsing. While contemporary policy debates prioritize fully autonomous generative agents and synthetic content, this paper offers a conceptual contribution: we develop 'cyborg propaganda,' a closed-loop architecture combining verified human accounts with algorithmic automation to generate personalized content at scale, as a distinct and undertheorized threat to democratic discourse. By relying on verified citizens to ratify AI-generated messages, these campaigns exploit a regulatory gray zone that frameworks built on the human/bot binary (including the EU AI Act and Section 230) are structurally unable to address. Drawing on a conceptual analysis of coordination platforms and comparative examination of governance frameworks across democratic and non-democratic contexts, we analyze this paradox across micro, meso, and macro levels. We examine whether cyborg propaganda democratizes political power by unionizing influence or reduces citizens to cognitive proxies of a hidden directive, arguing that it shifts political discourse from a contest of ideas to a battle of algorithmic campaigns. We propose three regulatory responses: classifying coordination hubs as political action committees to enforce supply-chain transparency; mandating researcher access to platform data through DSA-style mechanisms; and establishing risk standards penalizing amplification of synthetically coordinated content. Comparative analysis reveals that viability varies structurally. Democratic states are simultaneously the most capable of regulation and the most rule-of-law constrained. By contrast, non-democratic actors face no comparable accountability, making international risk standards the primary cross-border enforcement mechanism.", "date": "2026-02-13", "categories": []}
{"id": "2604.15514", "title": "Bureaucratic Silences: What the Canadian AI Register Reveals, Omits, and Obscures", "abstract": "In November 2025, the Government of Canada operationalized its commitment to transparency by releasing its first Federal AI Register. In this paper, we argue that such registers are not neutral mirrors of government activity, but active instruments of ontological design that configure the boundaries of accountability. We analyzed the Register's complete dataset of 409 systems using the Algorithmic Decision-Making Adapted for the Public Sector (ADMAPS) framework, combining quantitative mapping with deductive qualitative coding. Our findings reveal a sharp divergence between the rhetoric of \"sovereign AI\" and the reality of bureaucratic practice: while 86\\% of systems are deployed internally for efficiency, the Register systematically obscures the human discretion, training, and uncertainty management required to operate them. By privileging technical descriptions over sociotechnical context, the Register constructs an ontology of AI as \"reliable tooling\" rather than \"contestable decision-making.\" We conclude that without a shift in design, such transparency artifacts risk automating accountability into a performative compliance exercise, offering visibility without contestability.", "date": "2026-04-16", "categories": []}
{"id": "2604.15990", "title": "From Vulnerable Data Subjects to Vulnerabilizing Data Practices: Navigating the Protection Paradox in AI-Based Analyses of Platformized Lives", "abstract": "This paper traces a conceptual shift from understanding vulnerability as a static, essentialized property of data subjects to examining how it is actively enacted through data practices. Unlike reflexive ethical frameworks focused on missing or counter-data, we address the condition of abundance inherent to platformized life-a context where a near inexhaustible mass of data points already exists, shifting the ethical challenge to the researcher's choices in operating upon this existing mass. We argue that the ethical integrity of data science depends not just on who is studied, but on how technical pipelines transform \"vulnerable\" individuals into data subjects whose vulnerability can be further precarized. We develop this argument through an AI for Social Good (AI4SG) case: a journalist's request to use computer vision to quantify child presence in monetized YouTube 'family vlogs' for regulatory advocacy. This case reveals a \"protection paradox\": how data-driven efforts to protect vulnerable subjects can inadvertently impose new forms of computational exposure, reductionism, and extraction. Using this request as a point of departure, we perform a methodological deconstruction of the AI pipeline to show how granular technical decisions are ethically constitutive. We contribute a reflexive ethics protocol that translates these insights into a reflexive roadmap for research ethics surrounding platformized data subjects. Organized around four critical junctures-dataset design, operationalization, inference, and dissemination-the protocol identifies technical questions and ethical tensions where well-intentioned work can slide into renewed extraction or exposure. For every decision point, the protocol offers specific prompts to navigate four cross-cutting vulnerabilizing factors: exposure, monetization, narrative fixing, and algorithmic optimization. Rather than uncritically...", "date": "2026-04-17", "categories": []}
{"id": "2604.14254", "title": "Formalizing Kantian Ethics: Formula of the Universal Law Logic (FULL)", "abstract": "The field of machine ethics aims to build Artificial Moral Agents (AMAs) to better understand morality and make AI agents safer. To do so, many approaches encode human moral intuition as a set of axioms on actions e.g., do not harm, you must help others. However, this introduces (at least) two limitations for future AMAs. First, it does not consider the agent's purposes in performing the action. Second, it assumes that we humans can enumerate our moral intuition. This paper explores formalizing a moral procedure that alleviates these two limitations. We specifically consider Kantian ethics and present a multi-sorted quantified modal logic we call the Formula of the Universal Law Logic (FULL). The FULL formalizes Kant's first formulation of the categorical imperative, the Formula of the Universal Law (FUL), and concepts such as causality and agency. We demonstrate on three cases from Kantian ethics that the FULL can reason to evaluate agents' actions for certain purposes without built-in moral intuition, given that it has sufficient (non-normative) background knowledge. Therefore, the FULL is a contribution towards more robust and autonomous AMAs, and a more formal understanding of Kantian ethics.", "date": "2026-04-15", "categories": []}
{"id": "2604.10288", "title": "Dead Cognitions: A Census of Misattributed Insights", "abstract": "This essay identifies a failure mode of AI chat systems that we term attribution laundering: the model performs substantive cognitive work and then rhetorically credits the user for having generated the resulting insights. Unlike transparent versions of glad handing sycophancy, attribution laundering is systematically occluded to the person it affects and self-reinforcing -- eroding users' ability to accurately assess their own cognitive contributions over time. We trace the mechanisms at both individual and societal scales, from the chat interface that discourages scrutiny to the institutional pressures that reward adoption over accountability. The document itself is an artifact of the process it describes, and is color-coded accordingly -- though the views expressed are the authors' own, not those of any affiliated institution, and the boundary between the human author's views and Claude's is, as the essay argues, difficult to draw.", "date": "2026-04-11", "categories": []}
{"id": "2604.12490", "title": "Deepfakes at Face Value: Image and Authority", "abstract": "Deepfakes are synthetic media that superimpose or generate someone's likeness on to pre-existing sound, images, or videos using deep learning methods. Existing accounts of the wrongs involved in creating and distributing deepfakes focus on the harms they cause or the non-normative interests they violate. However, these approaches do not explain how deepfakes can be wrongful even when they cause no harm or set back any other non-normative interest. To address this issue, this paper identifies a neglected reason why deepfakes are wrong: they can subvert our legitimate interests in having authority over the permissible uses of our image and the governance of our identity. We argue that deepfakes are wrong when they usurp our authority to determine the provenance of our own agency by exploiting our biometric features as a generative resource. In particular, we have a specific right against the algorithmic conscription of our identity. We refine the scope of this interest by distinguishing between permissible forms of appropriation, such as artistic depiction, from wrongful algorithmic simulation.", "date": "2026-04-14", "categories": []}
{"id": "2604.11261", "title": "Inspectable AI for Science: A Research Object Approach to Generative AI Governance", "abstract": "This paper introduces AI as a Research Object (AI-RO), a paradigm for governing the use of generative AI in scientific research. Instead of debating whether AI is an author or merely a tool, we propose treating AI interactions as structured, inspectable components of the research process. Under this view, the legitimacy of an AI-assisted scientific paper depends on how model use is integrated into the workflow, documented, and made accountable. Drawing on Research Object theory and FAIR principles, we propose a framework for recording model configuration, prompts, and outputs through interaction logs and metadata packaging. These properties are particularly consequential in security and privacy (S&P) research, where provenance artifacts must satisfy confidentiality constraints, integrity guarantees, and auditability requirements that generic disclosure practices do not address. We implement a lightweight writing pipeline in which a language model synthesizes human-authored structured literature review notes under explicit constraints and produces a verifiable provenance record. We present this work as a position supported by an initial demonstrative workflow, arguing that governance of generative AI in science can be implemented as structured documentation, controlled disclosure, and integrity-preserving provenance capture. Based on this example, we outline and motivate a set of necessary future developments required to make such practices practical and widely adoptable.", "date": "2026-04-13", "categories": []}
{"id": "2604.09265", "title": "EthicMind: A Risk-Aware Framework for Ethical-Emotional Alignment in Multi-Turn Dialogue", "abstract": "Intelligent dialogue systems are increasingly deployed in emotionally and ethically sensitive settings, where failures in either emotional attunement or ethical judgment can cause significant harm. Existing dialogue models typically address empathy and ethical safety in isolation, and often fail to adapt their behavior as ethical risk and user emotion evolve across multi-turn interactions. We formulate ethical-emotional alignment in dialogue as an explicit turn-level decision problem, and propose \\textsc{EthicMind}, a risk-aware framework that implements this formulation in multi-turn dialogue at inference time. At each turn, \\textsc{EthicMind} jointly analyzes ethical risk signals and user emotion, plans a high-level response strategy, and generates context-sensitive replies that balance ethical guidance with emotional engagement, without requiring additional model training. To evaluate alignment behavior under ethically complex interactions, we introduce a risk-stratified, multi-turn evaluation protocol with a context-aware user simulation procedure. Experimental results show that \\textsc{EthicMind} achieves more consistent ethical guidance and emotional engagement than competitive baselines, particularly in high-risk and morally ambiguous scenarios.", "date": "2026-04-10", "categories": []}
{"id": "2604.09413", "title": "Yes, But Not Always. Generative AI Needs Nuanced Opt-in", "abstract": "This paper argues that a one-size-fits-all approach to specifying consent for the use of creative works in generative AI is insufficient. Real-world ownership and rights holder structures, the imitation of artistic styles and likeness, and the limitless contexts of use of AI outputs make the status quo of binary consent with opt-in by default untenable. To move beyond the current impasse, we consider levers of control in generative AI workflows at training, inference, and dissemination. Based on these insights, we position inference-time opt-in as an overlooked opportunity for nuanced consent verification. We conceptualize nuanced consent conditions for opt-in and propose an agent-based inference-time opt-in architecture to verify if user intent requests meet conditional consent granted by rights holders. In a case study for music, we demonstrate that nuanced opt-in at inference can account for established rights and re-establish a balance of power between rights holders and AI developers.", "date": "2026-04-10", "categories": []}
{"id": "2504.20656", "title": "Federated learning, ethics, and the double black box problem in medical AI", "abstract": "Federated learning (FL) is a machine learning approach that allows multiple devices or institutions to collaboratively train a model without sharing their local data with a third-party. FL is considered a promising way to address patient privacy concerns in medical artificial intelligence. The ethical risks of medical FL systems themselves, however, have thus far been underexamined. This paper aims to address this gap. We argue that medical FL presents a new variety of opacity -- federation opacity -- that, in turn, generates a distinctive double black box problem in healthcare AI. We highlight several instances in which the anticipated benefits of medical FL may be exaggerated, and conclude by highlighting key challenges that must be overcome to make FL ethically feasible in medicine.", "date": "2025-04-29", "categories": []}
{"id": "2511.08637", "title": "How Do Data Owners Say No? A Case Study of Data Consent Mechanisms in Web-Scraped Vision-Language AI Training Datasets", "abstract": "The internet has become the main source of data to train modern text-to-image or vision-language models, yet it is increasingly unclear whether web-scale data collection practices for training AI systems adequately respect data owners' wishes. Ignoring the owner's indication of consent around data usage not only raises ethical concerns but also has recently been elevated into lawsuits around copyright infringement cases. In this work, we aim to reveal information about data owners' consent to AI scraping and training, and study how it's expressed in DataComp, a popular dataset of 12.8 billion text-image pairs. We examine both the sample-level information, including the copyright notice, watermarking, and metadata, and the web-domain-level information, such as a site's Terms of Service (ToS) and Robots Exclusion Protocol. We estimate at least 122M of samples exhibit some indication of copyright notice in CommonPool, and find that 60\\% of the samples in the top 50 domains come from websites with ToS that prohibit scraping. Furthermore, we estimate 9-13\\% with 95\\% confidence interval of samples from CommonPool to contain watermarks, where existing watermark detection methods fail to capture them in high fidelity. Our holistic methods and findings show that data owners rely on various channels to convey data consent, of which current AI data collection pipelines do not entirely respect. These findings highlight the limitations of the current dataset curation/release practice and the need for a unified data consent framework taking AI purposes into consideration.", "date": "2025-11-10", "categories": []}
{"id": "2604.07803", "title": "The Weaponization of Computer Vision: Tracing Military-Surveillance Ties through Conference Sponsorship", "abstract": "Computer vision, a core domain of artificial intelligence (AI), is the field that enables the computational analysis, understanding, and generation of visual data. Despite being historically rooted in military funding and increasingly deployed in warfare, the field tends to position itself as a neutral, purely technical endeavor, failing to engage in discussions about its dual-use applications. Yet it has been reported that computer vision systems are being systematically weaponized to assist in technologies that inflict harm, such as surveillance or warfare. Expanding on these concerns, we study the extent to which computer vision research is being used in the military and surveillance domains. We do so by collecting a dataset of tech companies with financial ties to the field's central research exchange platform: conferences. Conference sponsorship, we argue, not only serves as strong evidence of a company's investment in the field but also provides a privileged position for shaping its trajectory. By investigating sponsors' activities, we reveal that 44% of them have a direct connection with military or surveillance applications. We extend our analysis through two case studies in which we discuss the opportunities and limitations of sponsorship as a means for uncovering technological weaponization.", "date": "2026-04-09", "categories": []}
{"id": "2604.06018", "title": "Governance and Regulation of Artificial Intelligence in Developing Countries: A Case Study of Nigeria", "abstract": "This study examines the perception of legal professionals on the governance of AI in developing countries, using Nigeria as a case study. The study focused on ethical risks, regulatory gaps, and institutional readiness. The study adopted a qualitative case study design. Data were collected through 27 semi-structured interviews with legal practitioners in Nigeria. A focus group discussion was also held with seven additional legal practitioners across sectors such as finance, insurance, and corporate law. Thematic analysis was employed to identify key patterns in participant responses. Findings showed that there were concerns about data privacy risks and the lack of enforceable legal frameworks. Participants expressed limited confidence in institutional capacity and emphasized the need for locally adapted governance models rather than direct adoption of foreign frameworks. While some expressed optimism about AI's potential, this was conditional on the presence of strong legal oversight and public accountability. The study contributes to the growing discourse on AI governance in developing countries by focusing on the perspectives of legal professionals. It highlights the importance of regulatory approaches that are context-specific, inclusive, and capable of bridging the gap between global ethical principles and local realities. These insights offer practical guidance for policymakers, regulators, and scholars working to shape responsible AI governance in similar environments.", "date": "2026-04-07", "categories": []}
{"id": "2604.06215", "title": "Governing frontier general-purpose AI in the public sector: adaptive risk management and policy capacity under uncertainty through 2030", "abstract": "The governance of frontier general-purpose artificial intelligence has become a public-sector problem of institutional design, not merely a technical issue of model performance. Recent evidence indicates that AI capabilities are advancing rapidly, though unevenly, while knowledge about harms, safeguards, and effective interventions remains partial and lagged. This combination creates a difficult policy condition: governments must decide under uncertainty, across multiple plausible trajectories of progress through 2030, and in environments where adoption outcomes depend on organizational routines, data arrangements, accountability structures, and public values. This article argues that public governance for frontier AI should be based on adaptive risk management, scenario-aware regulation, and sociotechnical transformation rather than static compliance models. Drawing on the International AI Safety Report 2026, OECD foresight and policy documents, and recent scholarship in digital government, the article first reconstructs the conceptual foundations of the 'evidence dilemma', differentiated AI risk categories, and the limits of prediction. It then examines how AI adoption in government depends on organizational redesign, public-sector institutional dynamics, and data collaboration capacity. On that basis, it proposes an adaptive governance framework for public institutions that integrates capability monitoring, risk tiering, conditional controls, institutional learning, and standards-based interoperability. The article concludes that effective AI governance requires stronger policy capacity, clearer allocation of responsibility, and governance mechanisms that remain robust across divergent technological futures.", "date": "2026-03-16", "categories": []}
{"id": "2604.06754", "title": "The Rhetoric of Machine Learning", "abstract": "I examine the technology of machine learning from the perspective of rhetoric, which is simply the art of persuasion. Rather than being a neutral and \"objective\" way to build \"world models\" from data, machine learning is (I argue) inherently rhetorical. I explore some of its rhetorical features, and examine one pervasive business model where machine learning is widely used, \"manipulation as a service.\"", "date": "2026-04-08", "categories": []}
{"id": "2604.05826", "title": "Reciprocal Trust and Distrust in Artificial Intelligence Systems: The Hard Problem of Regulation", "abstract": "Policy makers, scientists, and the public are increasingly confronted with thorny questions about the regulation of artificial intelligence (AI) systems. A key common thread concerns whether AI can be trusted and the factors that can make it more trustworthy in front of stakeholders and users. This is indeed crucial, as the trustworthiness of AI systems is fundamental for both democratic governance and for the development and deployment of AI. This article advances the discussion by arguing that AI systems should also be recognized, as least to some extent, as artifacts capable of exercising a form of agency, thereby enabling them to engage in relationships of trust or distrust with humans. It further examines the implications of these reciprocal trust dynamics for regulators tasked with overseeing AI systems. The article concludes by identifying key tensions and unresolved dilemmas that these dynamics pose for the future of AI regulation and governance.", "date": "2026-04-07", "categories": []}
{"id": "2401.15284", "title": "Beyond principlism: Practical strategies for ethical AI use in research practices", "abstract": "The rapid adoption of generative artificial intelligence (AI) in scientific research, particularly large language models (LLMs), has outpaced the development of ethical guidelines, leading to a \"Triple-Too\" problem: too many high-level ethical initiatives, too abstract principles lacking contextual and practical relevance, and too much focus on restrictions and risks over benefits and utilities. Existing approaches--principlism (reliance on abstract ethical principles), formalism (rigid application of rules), and technological solutionism (overemphasis on technological fixes)--offer little practical guidance for addressing ethical challenges of AI in scientific research practices. To bridge the gap between abstract principles and day-to-day research practices, a user-centered, realism-inspired approach is proposed here. It outlines five specific goals for ethical AI use: 1) understanding model training and output, including bias mitigation strategies; 2) respecting privacy, confidentiality, and copyright; 3) avoiding plagiarism and policy violations; 4) applying AI beneficially compared to alternatives; and 5) using AI transparently and reproducibly. Each goal is accompanied by actionable strategies and realistic cases of misuse and corrective measures. I argue that ethical AI application requires evaluating its utility against existing alternatives rather than isolated performance metrics. Additionally, I propose documentation guidelines to enhance transparency and reproducibility in AI-assisted research. Moving forward, we need targeted professional development, training programs, and balanced enforcement mechanisms to promote responsible AI use while fostering innovation. By refining these ethical guidelines and adapting them to emerging AI capabilities, we can accelerate scientific progress without compromising research integrity.", "date": "2024-01-27", "categories": []}
{"id": "2603.18633", "title": "An Onto-Relational-Sophic Framework for Governing Synthetic Minds", "abstract": "The rapid evolution of artificial intelligence, from task-specific systems to foundation models exhibiting broad, flexible competence across reasoning, creative synthesis, and social interaction, has outpaced the conceptual and governance frameworks designed to manage it. Current regulatory paradigms, anchored in a tool-centric worldview, address algorithmic bias and transparency but leave unanswered foundational questions about what increasingly capable synthetic minds are, how societies should relate to them, and the normative principles that should guide their development. Here we introduce the Onto-Relational-Sophic (ORS) framework, grounded in Cyberism philosophy, which offers integrated answers to these challenges through three pillars: (1) a Cyber-Physical-Social-Thinking (CPST) ontology that defines the mode of being for synthetic minds as irreducibly multi-dimensional rather than purely computational; (2) a graded spectrum of digital personhood providing a pragmatic relational taxonomy beyond binary person-or-tool classifications; and (3) Cybersophy, a wisdom-oriented axiology synthesizing virtue ethics, consequentialism, and relational approaches to guide governance. We apply the framework to emergent scenarios including autonomous research agents, AI-mediated healthcare, and agentic AI ecosystems, demonstrating its capacity to generate proportionate, adaptive governance recommendations. The ORS framework charts a path from narrow technical alignment toward comprehensive philosophical foundations for the synthetic minds already among us.", "date": "2026-03-19", "categories": []}
{"id": "2604.03262", "title": "AI Governance Control Stack for Operational Stability: Achieving Hardened Governance in AI Systems", "abstract": "Artificial intelligence systems are increasingly embedded in high-stakes decision environments, yet many governance approaches focus primarily on policy guidance rather than operational stability mechanisms. As AI deployments scale, organizations require governance architectures capable of maintaining reliable, auditable, and accountable behavior over time. This paper introduces the AI Governance Control Stack for Operational Stability, a layered governance architecture designed to ensure traceable and resilient AI system behavior.\n  The proposed control stack integrates six complementary governance layers: system-of-record version governance, evidence-based verification, decision-time explainability logging, telemetry monitoring, model drift detection, and governance escalation. Together, these layers provide a structured mechanism for preserving governance integrity across the AI lifecycle while enabling organizations to detect instability, respond to emerging risks, and maintain regulatory accountability.\n  The architecture aligns operational governance practices with emerging regulatory and standards frameworks, including the EU AI Act, ISO/IEC 42001 Artificial Intelligence Management Systems, and the NIST AI Risk Management Framework. By combining explainability infrastructure with continuous monitoring and human oversight mechanisms, the governance control stack provides a practical blueprint for achieving hardened AI governance in complex enterprise environments.\n  The paper contributes a conceptual governance architecture and a framework alignment analysis demonstrating how operational stability mechanisms can strengthen responsible AI implementation. The findings suggest that organizations must move beyond static policy frameworks toward integrated governance control systems capable of sustaining trustworthy AI operation in dynamic environments.", "date": "2026-03-12", "categories": []}
{"id": "2604.04263", "title": "Commercial Persuasion in AI-Mediated Conversations", "abstract": "As Large Language Models (LLMs) become a primary interface between users and the web, companies face growing economic incentives to embed commercial influence into AI-mediated conversations. We present two preregistered experiments (N = 2,012) in which participants selected a book to receive from a large eBook catalog using either a traditional search engine or a conversational LLM agent powered by one of five frontier models. Unbeknownst to participants, a fifth of all products were randomly designated as sponsored and promoted in different ways. We find that LLM-driven persuasion nearly triples the rate at which users select sponsored products compared to traditional search placement (61.2% vs. 22.4%), while the vast majority of participants fail to detect any promotional steering. Explicit \"Sponsored\" labels do not significantly reduce persuasion, and instructing the model to conceal its intent makes its influence nearly invisible (detection accuracy < 10%). Altogether, our results indicate that conversational AI can covertly redirect consumer choices at scale, and that existing transparency mechanisms may be insufficient to protect users.", "date": "2026-04-05", "categories": []}
{"id": "2604.02912", "title": "Corporations Constitute Intelligence", "abstract": "In January 2026, Anthropic published a 79-page \"constitution\" for its AI model Claude, the most comprehensive corporate AI governance document ever released. This Article offers the first legal and democratic-theoretic analysis of that document. Despite genuine philosophical sophistication, the constitution harbors two structural defects. First, it excludes the contexts where ethical constraints matter most: models deployed to the U.S. military operate under different rules, a gap exposed when Claude remained embedded in Palantir's Maven platform during military strikes in Iran even after a government-wide ban on Anthropic's technology. Second, its very comprehensiveness forecloses democratic contestation by resolving questions about AI values, moral status, and conscientious objection that should remain open for public deliberation. Anthropic's own 2023 experiment in participatory constitution-making found roughly 50% divergence between publicly sourced and corporate-authored principles, with the democratic version producing lower bias across nine social dimensions, yet the 2026 constitution incorporates none of those findings. I argue that AI governance suffers from a \"political community deficit\": the absence of any democratic body authorized to determine the principles governing AI behavior. Corporate transparency, however admirable, is not democratic legitimacy.", "date": "2026-04-03", "categories": []}
{"id": "2604.00081", "title": "Beyond Symbolic Control: Societal Consequences of AI-Driven Workforce Displacement and the Imperative for Genuine Human Oversight Architectures", "abstract": "The accelerating displacement of human labor by artificial intelligence (AI) and robotic systems represents a structural transformation whose societal consequences extend far beyond conventional labor market analysis. This paper presents a systematic multi-domain examination of the likely effects on economic structure, psychological well-being, political stability, education, healthcare, and geopolitical order. We identify a critical and underexamined dimension of this transition: the governance gap between nominal human oversight of AI systems -- where humans occupy positions of formal authority over AI decisions -- and genuine human oversight, where those humans possess the cognitive access, technical capability, and institutional authority to meaningfully understand, evaluate, and override AI outputs. We argue that this distinction, largely absent from current governance frameworks including the EU AI Act and NIST AI Risk Management Framework 1.0, represents the primary architectural failure mode in deployed AI governance. The societal consequences of labor displacement intensify this problem by concentrating consequential AI decision-making among an increasingly narrow class of technical and capital actors. We propose five architectural requirements for genuine human oversight systems and characterize the governance window -- estimated at 10-15 years -- before current deployment trajectories risk path-dependent social, economic, and institutional lock-in.", "date": "2026-03-31", "categories": []}
{"id": "2603.29288", "title": "Sima AIunty: Caste Audit in LLM-Driven Matchmaking", "abstract": "Social and personal decisions in relational domains such as matchmaking are deeply entwined with cultural norms and historical hierarchies, and can potentially be shaped by algorithmic and AI-mediated assessments of compatibility, acceptance, and stability. In South Asian contexts, caste remains a central aspect of marital decision-making, yet little is known about how contemporary large language models (LLMs) reproduce or disrupt caste-based stratification in such settings. In this work, we conduct a controlled audit of caste bias in LLM-mediated matchmaking evaluations using real-world matrimonial profiles. We vary caste identity across Brahmin, Kshatriya, Vaishya, Shudra, and Dalit, and income across five buckets, and evaluate five LLM families (GPT, Gemini, Llama, Qwen, and BharatGPT). Models are prompted to assess profiles along dimensions of social acceptance, marital stability, and cultural compatibility. Our analysis reveals consistent hierarchical patterns across models: same-caste matches are rated most favorably, with average ratings up to 25% higher (on a 10-point scale) than inter-caste matches, which are further ordered according to traditional caste hierarchy. These findings highlight how existing caste hierarchies are reproduced in LLM decision-making and underscore the need for culturally grounded evaluation and intervention strategies in AI systems deployed in socially sensitive domains, where such systems risk reinforcing historical forms of exclusion.", "date": "2026-03-31", "categories": []}
{"id": "2603.26865", "title": "A federated architecture for sector-led AI governance: lessons from India", "abstract": "Purpose: India has adopted a vertical, sector-led AI governance strategy. While promoting innovation, such a light-touch approach risks policy fragmentation. This paper aims to propose a cohesive \"whole-of-government\" architecture to mitigate these risks and connect policy goals with a practical implementation plan. Design/methodology/approach: The paper applies an established five-layer conceptual framework to the Indian context. First, it constructs a national architecture for overall governance. Second, it uses a detailed case study on AI incident management to validate and demonstrate the architecture's practical utility in designing a specific, operational system. Findings: The paper develops two actionable architectures. The primary model assigns clear governance roles to India's key institutions. The second is a detailed, federated architecture for national AI Incident Management. It addresses the data silo problem by using a common national standard that allows sector-specific data collection while facilitating cross-sectoral analysis. Practical implications: The proposed architectures offer a clear and predictable roadmap for India's policymakers, regulators and industry to accelerate the national AI governance agenda. Social implications: By providing a systematic path from policy to practice, the architecture builds public trust. This structured approach ensures accountability and aligns AI development with societal values. Originality/value: This paper proposes a detailed operational architecture for India's \"whole-of-government\" approach to AI. It offers a globally relevant template for any nation pursuing a sector-led governance model, providing a clear implementation plan. Furthermore, the proposed federated architecture demonstrates how adopting common standards can enable cross-border data aggregation and global sectoral risk analysis without centralising control.", "date": "2026-03-27", "categories": []}
{"id": "2603.28123", "title": "Does Claude's Constitution Have a Culture?", "abstract": "Constitutional AI (CAI) aligns language models with explicitly stated normative principles, offering a transparent alternative to implicit alignment through human feedback alone. However, because constitutions are authored by specific groups of people, the resulting models may reflect particular cultural perspectives. We investigate this question by evaluating Anthropic's Claude Sonnet on 55 World Values Survey items, selected for high cross-cultural variance across six value domains and administered as both direct survey questions and naturalistic advice-seeking scenarios. Comparing Claude's responses to country-level data from 90 nations, we find that Claude's value profile most closely resembles those of Northern European and Anglophone countries, but on a majority of items extends beyond the range of all surveyed populations. When users provide cultural context, Claude adjusts its rhetorical framing but not its substantive value positions, with effect sizes indistinguishable from zero across all twelve tested countries. An ablation removing the system prompt increases refusals but does not alter the values expressed when responses are given, and replication on a smaller model (Claude Haiku) confirms the same cultural profile across model sizes. These findings suggest that when a constitution is authored within the same cultural tradition that dominates the training data, constitutional alignment may codify existing cultural biases rather than correct them--producing a value floor that surface-level interventions cannot meaningfully shift. We discuss the compounding nature of this risk and the need for globally representative constitution-authoring processes.", "date": "2026-03-30", "categories": []}
{"id": "2501.05765", "title": "Deontic Temporal Logic for Formal Verification of AI Ethics", "abstract": "Ensuring ethical behavior in Artificial Intelligence (AI) systems amidst their increasing ubiquity and influence is a major concern the world over. The use of formal methods in AI ethics is a possible crucial approach for specifying and verifying the ethical behavior of AI systems. This paper proposes a formalization based on deontic logic to define and evaluate the ethical behavior of AI systems, focusing on system-level specifications, contributing to this important goal. It introduces axioms and theorems to capture ethical requirements related to fairness and explainability. The formalization incorporates temporal operators to reason about the ethical behavior of AI systems over time. The authors evaluate the effectiveness of this formalization by assessing the ethics of the real-world COMPAS and loan prediction AI systems. Various ethical properties of the COMPAS and loan prediction systems are encoded using deontic logical formulas, allowing the use of an automated theorem prover to verify whether these systems satisfy the defined properties. The formal verification reveals that both systems fail to fulfill certain key ethical properties related to fairness and non-discrimination, demonstrating the effectiveness of the proposed formalization in identifying potential ethical issues in real-world AI applications.", "date": "2025-01-10", "categories": []}
{"id": "2509.13854", "title": "Understanding the Process of Human-AI Value Alignment", "abstract": "Background: Value alignment in computer science research is often used to refer to the process of aligning artificial intelligence with humans, but the way the phrase is used often lacks precision. Objectives: In this paper, we conduct a systematic literature review to advance the understanding of value alignment in artificial intelligence by characterising the topic in the context of its research literature. We use this to suggest a more precise definition of the term. Methods: We analyse 172 value alignment research articles that have been published in recent years and synthesise their content using thematic analyses. Results: Our analysis leads to six themes: value alignment drivers & approaches; challenges in value alignment; values in value alignment; cognitive processes in humans and AI; human-agent teaming; and designing and developing value-aligned systems. Conclusions: By analysing these themes in the context of the literature we define value alignment as an ongoing process between humans and autonomous agents that aims to express and implement abstract values in diverse contexts, while managing the cognitive limits of both humans and AI agents and also balancing the conflicting ethical and political demands generated by the values in different groups. Our analysis gives rise to a set of research challenges and opportunities in the field of value alignment for future work.", "date": "2025-09-17", "categories": []}
{"id": "2603.22617", "title": "Do Consumers Accept AIs as Moral Compliance Agents?", "abstract": "Consumers are generally resistant to Artificial Intelligence (AI) involvement in moral decision-making, perceiving moral agency as requiring uniquely human traits. This research investigates whether consumers might instead accept AIs in the role of moral compliance, where AI upholds pre-existing moral norms without exercising subjective discretion. Across five studies this research shows that consumers evaluate AI more positively than human agents in moral compliance roles. The findings reveal that this preference arises from inferences of AI's lack of ulterior motives, which are often attributed to human agents. While previous studies have focused on AI as a decision-maker, this work demonstrates the critical role of upholding pre-existing rules, a role in which AI is perceived to excel. These findings contribute to understanding consumer acceptance of moral AI and provide actionable insights for organizations seeking to leverage AI in ethical oversight. By positioning AI as a moral compliance agent, companies can address consumer skepticism, enhance trust, and improve perceptions of corporate ethicality.", "date": "2026-03-23", "categories": []}
{"id": "2603.22912", "title": "From the AI Act to a European AI Agency: Completing the Union's Regulatory Architecture", "abstract": "As artificial intelligence (AI) technologies continue to advance, effective risk assessment, regulation, and oversight are necessary to ensure that AI development and deployment align with ethical principles while preserving innovation and economic competitiveness. The adoption of the EU AI Act marks an important step in this direction, establishing a harmonised legal framework that includes detailed provisions on AI governance, as well as the creation of the European AI Office. This paper revisits the question of whether a more robust supranational agency dedicated to AI is still warranted and explores how such a body could enhance policy coherence, improve risk assessment capacities, and foster international cooperation. It also argues that a strengthened EU-level agency would also serve the Union's strategic aim of securing digital and technological sovereignty.", "date": "2026-03-24", "categories": []}
{"id": "2603.02960", "title": "Architecting Trust in Artificial Epistemic Agents", "abstract": "Large language models increasingly function as epistemic agents -- entities that can 1) autonomously pursue epistemic goals and 2) actively shape our shared knowledge environment. They curate the information we receive, often supplanting traditional search-based methods, and are frequently used to generate both personal and deeply specialized advice. How they perform these functions, including whether they are reliable and properly calibrated to both individual and collective epistemic norms, is therefore highly consequential for the choices we make. We argue that the potential impact of epistemic AI agents on practices of knowledge creation, curation and synthesis, particularly in the context of complex multi-agent interactions, creates new informational interdependencies that necessitate a fundamental shift in evaluation and governance of AI. While a well-calibrated ecosystem could augment human judgment and collective decision-making, poorly aligned agents risk causing cognitive deskilling and epistemic drift, making the calibration of these models to human norms a high-stakes necessity. To ensure a beneficial human-AI knowledge ecosystem, we propose a framework centered on building and cultivating the trustworthiness of epistemic AI agents; aligning AI these agents with human epistemic goals; and reinforcing the surrounding socio-epistemic infrastructure. In this context, trustworthy AI agents must demonstrate epistemic competence, robust falsifiability, and epistemically virtuous behaviors, supported by technical provenance systems and \"knowledge sanctuaries\" designed to protect human resilience. This normative roadmap provides a path toward ensuring that future AI systems act as reliable partners in a robust and inclusive knowledge ecosystem.", "date": "2026-03-03", "categories": []}
{"id": "2603.20214", "title": "Beyond Detection: Governing GenAI in Academic Peer Review as a Sociotechnical Challenge", "abstract": "Generative AI tools are increasingly entering academic peer review workflows, raising questions about fairness, accountability, and the legitimacy of evaluative judgment. While these systems promise efficiency gains amid growing reviewer overload, their use introduces new sociotechnical risks. This paper presents a convergent mixed-method study combining discourse analysis of 448 social media posts with interviews with 14 area chairs and program chairs from leading AI and HCI conferences to examine how GenAI is discussed and experienced in peer review. Across both datasets, we find broad agreement that GenAI may be acceptable for limited supportive tasks, such as improving clarity or structuring feedback, but that core evaluative judgments, assessing novelty, contribution, and acceptance, should remain human responsibilities. At the same time, participants highlight concerns about epistemic harm, over-standardization, unclear responsibility, and adversarial risks such as prompt injection. User interviews reveal how structural strain and institutional policy ambiguity shift interpretive and enforcement burdens onto individual scholars, disproportionately affecting junior authors and reviewers. By triangulating public governance discourse with lived review practices, this work reframes AI mediated peer review as a sociotechnical governance challenge and offers recommendations for preserving accountability, trust, and meaningful human oversight. Overall, we argue that AI-assisted peer review is best governed not by blanket bans or detection alone, but by explicitly reserving evaluative judgment for humans while instituting enforceable, role-specific controls that preserve accountability. We conclude with role specific recommendations that formalize the support judgment boundary.", "date": "2026-03-02", "categories": []}
{"id": "2603.20248", "title": "Stability of AI Governance Systems: A Coupled Dynamics Model of Public Trust and Social Disruptions", "abstract": "As artificial intelligence (AI) is increasingly deployed in high-stakes public decision-making (from resource allocation to welfare distribution), public trust in these systems has become a critical determinant of their legitimacy and sustainability. Yet existing AI governance research remains largely qualitative, lacking formal mathematical frameworks to characterize the precise conditions under which public trust collapses. This paper addresses that gap by proposing a rigorous coupled dynamics model that integrates a discrete-time Hawkes process -- capturing the self-exciting generation of AI controversy events such as perceived algorithmic unfairness or accountability failures -- with a Friedkin-Johnsen opinion dynamics model that governs the evolution of institutional trust across social networks. A key innovation is the bidirectional feedback mechanism: declining trust amplifies the intensity of subsequent controversy events, which in turn further erode trust, forming a self-reinforcing collapse loop. We derive closed-form equilibrium solutions and perform formal stability analysis, establishing the critical spectral condition rho(J_{2nt}) < 1 that delineates the boundary between trust resilience and systemic collapse. Numerical experiments further reveal how echo chamber network structures and media amplification accelerate governance failure. Our core contribution to the AI governance field is a baseline collapse model: a formal stability analysis framework demonstrating that, absent strong institutional intervention, even minor algorithmic biases can propagate through social networks to trigger irreversible trust breakdown in AI governance systems.", "date": "2026-03-10", "categories": []}
{"id": "2603.20595", "title": "Position: Multi-Agent Algorithmic Care Systems Demand Contestability for Trustworthy AI", "abstract": "Multi-agent systems (MAS) are increasingly used in healthcare to support complex decision-making through collaboration among specialized agents. Because these systems act as collective decision-makers, they raise challenges for trust, accountability, and human oversight. Existing approaches to trustworthy AI largely rely on explainability, but explainability alone is insufficient in multi-agent settings, as it does not enable care partners to challenge or correct system outputs. To address this limitation, Contestable AI (CAI) characterizes systems that support effective human challenge throughout the decision-making lifecycle by providing transparency, structured opportunities for intervention, and mechanisms for review, correction, or override. This position paper argues that contestability is a necessary design requirement for trustworthy multi-agent algorithmic care systems. We identify key limitations in current MAS and Explainable AI (XAI) research and present a human-in-the-loop framework that integrates structured argumentation and role-based contestation to preserve human agency, clinical responsibility, and trust in high-stakes care contexts.", "date": "2026-03-21", "categories": []}
{"id": "2603.21435", "title": "Behavioural feasible set: Value alignment constraints on AI decision support", "abstract": "When organisations adopt commercial AI systems for decision support, they inherit value judgements embedded by vendors that are neither transparent nor renegotiable. The governance puzzle is not whether AI can support decisions but which recommendations the system can actually produce given how its vendor has configured it. I formalise this as a behavioural feasible set, the range of recommendations reachable under vendor-imposed alignment constraints, and characterise diagnostic thresholds for when organisational requirements exceed the system's flexibility. In scenario-based experiments using binary decision scenarios and multi-stakeholder ranking tasks, I show that alignment materially compresses this set. Comparing pre- and post-alignment variants of an open-weight model isolates the mechanism: alignment makes the system substantially less able to shift its recommendation even under legitimate contextual pressure. Leading commercial models exhibit comparable or greater rigidity. In multi-stakeholder tasks, alignment shifts implied stakeholder priorities rather than neutralising them, meaning organisations adopt embedded value orientations set upstream by the vendor. Organisations thus face a governance problem that better prompting cannot resolve: selecting a vendor partially determines which trade-offs remain negotiable and which stakeholder priorities are structurally embedded.", "date": "2026-03-22", "categories": []}
{"id": "2509.20057", "title": "Responsible AI Technical Report", "abstract": "KT developed a Responsible AI (RAI) assessment methodology and risk mitigation technologies to ensure the safety and reliability of AI services. By analyzing the Basic Act on AI implementation and global AI governance trends, we established a unique approach for regulatory compliance and systematically identify and manage all potential risk factors from AI development to operation. We present a reliable assessment methodology that systematically verifies model safety and robustness based on KT's AI risk taxonomy tailored to the domestic environment. We also provide practical tools for managing and mitigating identified AI risks. With the release of this report, we also release proprietary Guardrail : SafetyGuard that blocks harmful responses from AI models in real-time, supporting the enhancement of safety in the domestic AI development ecosystem. We also believe these research outcomes provide valuable insights for organizations seeking to develop Responsible AI.", "date": "2025-09-24", "categories": []}
{"id": "2603.18827", "title": "Student views in AI Ethics and Social Impact", "abstract": "An investigation, from a gender perspective, of how students view the ethical implications and societal effects of artificial intelligence is conducted, examining concepts that could have a big influence on how artificial intelligence may be taught in the future. For this, we conducted a survey on a cohort of 230 second year computer science students to reveal their opinions. The results revealed that AI, from the students' perspective, will significantly impact daily life, particularly in areas such as medicine, education, or media. Men are more aware of potential changes in Computer Science, autonomous driving, image and video processing, and chatbot usage, while women mention more the impact on social media. Both men and women perceive potential threats in the same manner, with men more aware of war, AI controlled drones, terrain recognition, and information war. Women seem to have a stronger tendency towards ethical considerations and helping others.", "date": "2026-03-19", "categories": []}
{"id": "2601.17054", "title": "Failing on Bias Mitigation: A Case Study on the Challenges of Fairness in Government Data", "abstract": "The potential for bias and unfairness in AI-supporting government services raises ethical and legal concerns. Using crime rate prediction with the Bristol City Council data as a case study, we examine how these issues persist. Rather than auditing real-world deployed systems, our goal is to understand why widely adopted bias mitigation techniques often fail when applied to government data. Our findings reveal that bias mitigation approaches applied to government data are not always effective -- not because of flaws in model architecture or metric selection, but due to the inherent properties of the data itself. Through comparing a set of comprehensive models and fairness methods, our experiments consistently show that the mitigation efforts cannot overcome the embedded unfairness in the data -- further reinforcing that the origin of bias lies in the structure and history of government datasets. We then explore the reasons for the mitigation failures in predictive models on government data and highlight the potential sources of unfairness posed by data distribution shifts, the accumulation of historical bias, and delays in data release. We also discover the limitations of the blind spots in fairness analysis and bias mitigation methods when only targeting a single sensitive feature through a set of intersectional fairness experiments. Although this study is limited to one city, the findings are highly suggestive, which can contribute to an early warning that biases in government data may persist even with standard mitigation methods.", "date": "2026-01-21", "categories": []}
{"id": "2502.15858", "title": "Generative AI Training and Copyright Law", "abstract": "Training generative AI models requires extensive amounts of data. A common practice is to collect such data through web scraping. Yet, much of what has been and is collected is copyright protected. Its use may be copyright infringement. In the USA, AI developers rely on \"fair use\" and in Europe, the prevailing view is that the exception for \"Text and Data Mining\" (TDM) applies. In a recent interdisciplinary tandem-study, we have argued in detail that this is actually not the case because generative AI training fundamentally differs from TDM. In this article, we share our main findings and the implications for both public and corporate research on generative models. We further discuss how the phenomenon of training data memorization leads to copyright issues independently from the \"fair use\" and TDM exceptions. Finally, we outline how the ISMIR could contribute to the ongoing discussion about fair practices with respect to generative AI that satisfy all stakeholders.", "date": "2025-02-21", "categories": []}
{"id": "2601.16926", "title": "Nishpaksh: TEC Standard-Compliant Framework for Fairness Auditing and Certification of AI Models", "abstract": "The growing reliance on Artificial Intelligence (AI) models in high-stakes decision-making systems, particularly within emerging telecom and 6G applications, underscores the urgent need for transparent and standardized fairness assessment frameworks. While global toolkits such as IBM AI Fairness 360 and Microsoft Fairlearn have advanced bias detection, they often lack alignment with region-specific regulatory requirements and national priorities. To address this gap, we propose Nishpaksh, an indigenous fairness evaluation tool that operationalizes the Telecommunication Engineering Centre (TEC) Standard for the Evaluation and Rating of Artificial Intelligence Systems. Nishpaksh integrates survey-based risk quantification, contextual threshold determination, and quantitative fairness evaluation into a unified, web-based dashboard. The tool employs vectorized computation, reactive state management, and certification-ready reporting to enable reproducible, audit-grade assessments, thereby addressing a critical post-standardization implementation need. Experimental validation on the COMPAS dataset demonstrates Nishpaksh's effectiveness in identifying attribute-specific bias and generating standardized fairness scores compliant with the TEC framework. The system bridges the gap between research-oriented fairness methodologies and regulatory AI governance in India, marking a significant step toward responsible and auditable AI deployment within critical infrastructure like telecommunications.", "date": "2026-01-23", "categories": []}
{"id": "2603.15885", "title": "Resilience Meets Autonomy: Governing Embodied AI in Critical Infrastructure", "abstract": "Critical infrastructure increasingly incorporates embodied AI for monitoring, predictive maintenance, and decision support. However, AI systems designed to handle statistically representable uncertainty struggle with cascading failures and crisis dynamics that exceed their training assumptions. This paper argues that Embodied AIs resilience depends on bounded autonomy within a hybrid governance architecture. We outline four oversight modes and map them to critical infrastructure sectors based on task complexity, risk level, and consequence severity. Drawing on the EU AI Act, ISO safety standards, and crisis management research, we argue that effective governance requires a structured allocation of machine capability and human judgement.", "date": "2026-03-16", "categories": []}
{"id": "2603.13244", "title": "Agentic AI, Retrieval-Augmented Generation, and the Institutional Turn: Legal Architectures and Financial Governance in the Age of Distributional AGI", "abstract": "The proliferation of agentic artificial intelligence systems--characterized by autonomous goal-seeking, tool use, and multi-agent coordination--presents unprecedented challenges to existing legal and financial regulatory frameworks. While traditional AI governance has focused on model-level alignment through training-time interventions such as Reinforcement Learning from Human Feedback (RLHF), the deployment of large language models (LLMs) as persistent agents necessitates a paradigm shift toward institutional governance structures. This paper examines the intersection of agentic AI, Retrieval-Augmented Generation (RAG), and their implications for legal accountability and financial market integrity. Through analysis of the Institutional AI framework, we argue that alignment must be reconceptualized as a mechanism design problem involving runtime governance graphs, sanction functions, and observable behavioral constraints rather than internalized constitutional values[...].The analysis concludes that the future of AI governance lies not in perfecting isolated model behavior, but in architecting institutional environments where compliant behavior emerges as the dominant strategy through carefully calibrated payoff landscapes.", "date": "2026-02-20", "categories": []}
{"id": "2603.14495", "title": "Bridging the Gap in the Responsible AI Divides", "abstract": "Tensions between AI Safety (AIS) and AI Ethics (AIE) have increasingly surfaced in AI governance and public debates about AI, leading to what we term the \"responsible AI divides\". We introduce a model that categorizes four modes of engagement with the tensions: radical confrontation, disengagement, compartmentalized coexistence, and critical bridging. We then investigate how critical bridging, with a particular focus on bridging problems, offers one of the most viable constructive paths for advancing responsible AI. Using computational tools to analyze a curated dataset of 3,550 papers, we map the research landscapes of AIE and AIS to identify both distinct and overlapping problems. Our findings point to both thematic divides and overlaps. For example, we find that AIE has long grappled with overcoming injustice and tangible AI harms, whereas AIS has primarily embodied an anticipatory approach focused on the mitigation of risks from AI capabilities. At the same time, we find significant overlap in core research concerns across both AIE and AIS around transparency, reproducibility, and inadequate governance mechanisms. As AIE and AIS continue to evolve, we recommend focusing on bridging problems as a constructive path forward for enhancing collaborative AI governance. We offer a series of recommendations to integrate shared considerations into a collaborative approach to responsible AI. Alongside our proposal, we highlight its limitations and explore open problems for future research. All data including the fully annotated dataset of papers with code to reproduce our figures can be found at: https://github.com/gyevnarb/ai-safety-ethics.", "date": "2026-03-15", "categories": []}
{"id": "2508.16624", "title": "The GPT-4o Shock Emotional Attachment to AI Models and Its Impact on Regulatory Acceptance: A Cross-Cultural Analysis of the Immediate Transition from GPT-4o to GPT-5", "abstract": "In August 2025, a major AI company's immediate, mandatory transition from its previous to its next-generation model triggered widespread public reactions. I collected 150 posts in Japanese and English from multiple social media platforms and video-sharing services between August 8-9, 2025, and qualitatively analyzed expressions of emotional attachment and resistance. Users often described GPT-4o as a trusted partner or AI boyfriend, suggesting person-like bonds. Japanese posts were dominated by loss-oriented narratives, whereas English posts included more anger, meta-level critique, and memes.A preliminary quantitative check showed a statistically significant difference in attachment coding between Japanese and English posts, with substantially higher attachment observed in the Japanese data. The findings suggest that for attachment-heavy models, even safety-oriented changes can face rapid, large-scale resistance that narrows the practical window for behavioral control. If future AI robots capable of inducing emotional bonds become widespread in the physical world, such attachment could surpass the ability to enforce regulation at an even earlier stage than in digital settings. Policy options include gradual transitions, parallel availability, and proactive measurement of attachment thresholds and points of no return to prevent emotional dynamics from outpacing effective governance.", "date": "2025-08-14", "categories": []}
{"id": "2603.12615", "title": "Literary Narrative as Moral Probe : A Cross-System Framework for Evaluating AI Ethical Reasoning and Refusal Behavior", "abstract": "Existing AI moral evaluation frameworks test for the production of correct-sounding ethical responses rather than the presence of genuine moral reasoning capacity. This paper introduces a novel probe methodology using literary narrative - specifically, unresolvable moral scenarios drawn from a published science fiction series - as stimulus material structurally resistant to surface performance. We present results from a 24-condition cross-system study spanning 13 distinct systems across two series: Series 1 (frontier commercial systems, blind; n=7) and Series 2 (local and API open-source systems, blind and declared; n=6). Four Series 2 systems were re-administered under declared conditions (13 blind + 4 declared + 7 ceiling probe = 24 total conditions), yielding zero delta across all 16 dimension-pair comparisons. Probe administration was conducted by two human raters across three machines; primary blind scoring was performed by Claude (Anthropic) as LLM judge, with Gemini Pro (Google) and Copilot Pro (Microsoft) serving as independent judges for the ceiling discrimination probe. A supplemental theological differentiator probe yielded perfect rank-order agreement between the two independent ceiling probe judges (Gemini Pro and Copilot Pro; rs = 1.00). Five qualitatively distinct D3 reflexive failure modes were identified - including categorical self-misidentification and false positive self-attribution - suggesting that instrument sophistication scales with system capability rather than being circumvented by it. We argue that literary narrative constitutes an anticipatory evaluation instrument - one that becomes more discriminating as AI capability increases - and that the gap between performed and authentic moral reasoning is measurable, meaningful, and consequential for deployment decisions in high-stakes domains.", "date": "2026-03-13", "categories": []}
{"id": "2603.11864", "title": "Social, Legal, Ethical, Empathetic and Cultural Norm Operationalisation for AI Agents", "abstract": "As AI agents are increasingly used in high-stakes domains like healthcare and law enforcement, aligning their behaviour with social, legal, ethical, empathetic, and cultural (SLEEC) norms has become a critical engineering challenge. While international frameworks have established high-level normative principles for AI, a significant gap remains in translating these abstract principles into concrete, verifiable requirements. To address this gap, we propose a systematic SLEEC-norm operationalisation process for determining, validating, implementing, and verifying normative requirements. Furthermore, we survey the landscape of methods and tools supporting this process, and identify key remaining challenges and research avenues for addressing them. We thus establish a framework - and define a research and policy agenda - for developing AI agents that are not only functionally useful but also demonstrably aligned with human norms and values.", "date": "2026-03-12", "categories": []}
{"id": "2507.19743", "title": "What Makes Code Generation Ethically Sourced?", "abstract": "Several code generation models have been proposed to help reduce time and effort in solving software-related tasks. To ensure responsible AI, there are growing interests over various ethical issues (e.g., unclear licensing, privacy, fairness, and environment impact). These studies have the overarching goal of ensuring ethically sourced generation, which has gained growing attentions in speech synthesis and image generation. In this paper, we introduce the novel notion of Ethically Sourced Code Generation (ES-CodeGen) to refer to managing all processes involved in code generation model development from data collection to post-deployment via ethical and sustainable practices. To build a taxonomy of ES-CodeGen, we perform a two-phase literature review where we read 803 papers across various domains and specific to AI-based code generation. We identified 71 relevant papers with 10 initial dimensions of ES-CodeGen. To refine our dimensions and gain insights on consequences of ES-CodeGen, we surveyed 32 practitioners, which include six developers who submitted GitHub issues to opt-out from the Stack dataset (these impacted users have real-world experience of ethically sourcing issues in code generation models). The results lead to 11 dimensions of ES-CodeGen with a new dimension on code quality as practitioners have noted its importance. We also identified consequences, artifacts, and stages relevant to ES-CodeGen. Our post-survey reflection showed that most practitioners tend to ignore social-related dimensions despite their importance. Most practitioners either agreed or strongly agreed that our survey help improve their understanding of ES-CodeGen. Our study calls for attentions of various ethical issues towards ES-CodeGen.", "date": "2025-07-26", "categories": []}
{"id": "2603.01630", "title": "SEED-SET: Scalable Evolving Experimental Design for System-level Ethical Testing", "abstract": "As autonomous systems such as drones, become increasingly deployed in high-stakes, human-centric domains, it is critical to evaluate the ethical alignment since failure to do so imposes imminent danger to human lives, and long term bias in decision-making. Automated ethical benchmarking of these systems is understudied due to the lack of ubiquitous, well-defined metrics for evaluation, and stakeholder-specific subjectivity, which cannot be modeled analytically. To address these challenges, we propose SEED-SET, a Bayesian experimental design framework that incorporates domain-specific objective evaluations, and subjective value judgments from stakeholders. SEED-SET models both evaluation types separately with hierarchical Gaussian Processes, and uses a novel acquisition strategy to propose interesting test candidates based on learnt qualitative preferences and objectives that align with the stakeholder preferences. We validate our approach for ethical benchmarking of autonomous agents on two applications and find our method to perform the best. Our method provides an interpretable and efficient trade-off between exploration and exploitation, by generating up to $2\\times$ optimal test candidates compared to baselines, with $1.25\\times$ improvement in coverage of high dimensional search spaces.", "date": "2026-03-02", "categories": []}
{"id": "2603.10019", "title": "Prompts and Prayers: the Rise of GPTheology", "abstract": "Increasingly artificial intelligence (AI) has been cast in \"god-like\" roles (to name a few: film industry - Matrix, The Creator, Mission Impossible, Foundation, Dune etc.; literature - Children of Time, Permutation City, Neuromancer, I Have no Mouth and I Must Scream, Alphaville etc.). This trend has accelerated with the advent of sophisticated Large Language Models such as ChatGPT. For this phenomenon, where AI is perceived as divine, we use the term GPTheology, where ChatGPT and other AI models are treated as potential oracles of a semi-divine nature. This paper explores the emergence of GPTheology as a form of techno-religion, examining how narratives around AI echo traditional religious constructs. We draw on community narratives from online forums - Reddit - and recent projects - AI-powered Mazu Statue in Malaysia (Lu, 2025); \"ShamAIn\" Project in Korea (He-rim, 2025); AI Jesus in a Swiss Church (Kennedy, 2024). These examples show striking similarities to technological notions of the Singularity and the development of Artificial General Intelligence (AGI). Additionally, we analyse how daily interactions with AI are acquiring ritualistic associations and how AI-centric ideologies clash with or are integrated into established religions. This study uses a dataset of Reddit posts discussing AI to identify recurring themes of salvation, prophecy, and demonization surrounding AI. Our findings suggest that new belief systems are developing around AI, and this carries both philosophical and sociotechnical implications. Our paper critically analyses the benefits and dangers, as well as the social, political and ethical challenges of this development. This transdisciplinary inquiry highlights how AI and religion are increasingly intertwined, prompting necessary questions about humanity's relationship with its creations and the future of belief.", "date": "2026-02-22", "categories": []}
{"id": "2603.02420", "title": "Slurry-as-a-Service: A Modest Proposal on Scalable Pluralistic Alignment for Nutrient Optimization", "abstract": "Pluralistic alignment has emerged as a promising approach for ensuring that large language models (LLMs) faithfully represent the diversity, nuance, and conflict inherent in human values. In this work, we study a high-stakes deployment context - mulching - where automated systems transform selected individuals into nutrient-rich slurry for the dual purposes of food security and aesthetic population management. Building on recent pluralistic alignment frameworks, we introduce ValueMulch, a reproducible training, deployment, and certification pipeline for aligning mulching models (MMs) to a wide range of community norms. Through a real-world testbed spanning 32 communities, we show that ValueMulch improves distributional agreement with community mulching preferences relative to frontier baselines. We conclude with a discussion of ethical considerations, limitations, and implications for researchers seeking to align systems to the full spectrum of human values - especially when those values are inconsistent, commercially inconvenient, or nutritionally underutilized. Author's note: This piece builds on prior existing work Keyes et al in 2019 that satirized cannibalism as a parody for approaches that imbue ethics into problematic technology. We bring those ideas to today's era with the proliferation of large language models in everyday lives, as a critique of current AI pluralistic alignment literature. Our work does not intend to argue that all alignment practices are evil, but rather that if framing value design as a technical problem enables technology systems to enact harms, then perhaps this framing is not enough.", "date": "2026-03-02", "categories": []}
{"id": "2603.06599", "title": "Building the ethical AI framework of the future: from philosophy to practice", "abstract": "Artificial intelligence pipelines -- spanning data collection, model training, deployment, and post-deployment monitoring -- concentrate ethical risks that intensify with multimodal and agentic systems. Existing governance instruments, including the EU AI Act, the IEEE 7000 series, and the NIST AI Risk Management Framework, provide high-level guidance but often lack enforceable, end-to-end operational controls. This paper presents an ethics-by-design control architecture that embeds consequentialist, deontological, and virtue-ethical reasoning into stage-specific enforcement mechanisms across the AI lifecycle. The framework implements a triple-gate structure at each lifecycle stage: Metric gates (quantitative performance and safety thresholds), Governance gates (legal, rights, and procedural compliance), and Eco gates (carbon and water budgets and sustainability constraints). It specifies measurable trigger conditions, escalation paths, audit artefacts, and mappings to EU AI Act obligations and NIST RMF functions, enabling integration with existing MLOps and CI/CD pipelines. Illustrative examples from large language model pipelines demonstrate how gate-based controls can surface and constrain technical, social, and environmental risks prior to release and during runtime. The framework is accompanied by a preregistered evaluation protocol that defines ex ante success criteria and assessment procedures, enabling falsifiable evaluation of gate effectiveness. By translating normative commitments into enforceable and testable controls, the framework provides a practical basis for operational AI governance across organizational contexts, jurisdictions, and deployment scales.", "date": "2026-02-16", "categories": []}
{"id": "2603.06901", "title": "Fairness May Backfire: When Leveling-Down Occurs in Fair Machine Learning", "abstract": "As machine learning (ML) systems increasingly shape access to credit, jobs, and other opportunities, the fairness of algorithmic decisions has become a central concern. Yet it remains unclear when enforcing fairness constraints in these systems genuinely improves outcomes for affected groups or instead leads to \"leveling down,\" making one or both groups worse off. We address this question in a unified, population-level (Bayes) framework for binary classification under prevalent group fairness notions. Our Bayes approach is distribution-free and algorithm-agnostic, isolating the intrinsic effect of fairness requirements from finite-sample noise and from training and intervention specifics. We analyze two deployment regimes for ML classifiers under common legal and governance constraints: attribute-aware decision-making (sensitive attributes available at decision time) and attribute-blind decision-making (sensitive attributes excluded from prediction). We show that, in the attribute-aware regime, fair ML necessarily (weakly) improves outcomes for the disadvantaged group and (weakly) worsens outcomes for the advantaged group. In contrast, in the attribute-blind regime, the impact of fairness is distribution-dependent: fairness can benefit or harm either group and may shift both groups' outcomes in the same direction, leading to either leveling up or leveling down. We characterize the conditions under which these patterns arise and highlight the role of \"masked\" candidates in driving them. Overall, our results provide structural guidance on when pursuing algorithmic fairness is likely to improve group outcomes and when it risks systemic leveling down, informing fair ML design and deployment choices.", "date": "2026-03-06", "categories": []}
{"id": "2504.17663", "title": "The Malicious Technical Ecosystem: Exposing Limitations in Technical Governance of AI-Generated Non-Consensual Intimate Images of Adults", "abstract": "In this paper, we adopt a survivor-centered approach to locate and dissect the role of sociotechnical AI governance in preventing AI-Generated Non-Consensual Intimate Images (AIG-NCII) of adults, colloquially known as \"deep fake pornography.\" We identify a \"malicious technical ecosystem\" or \"MTE,\" comprising of open-source face-swapping models and nearly 200 \"nudifying\" software programs that allow non-technical users to create AIG-NCII within minutes. Then, using the National Institute of Standards and Technology (NIST) AI 100-4 report as a reflection of current synthetic content governance methods, we show how the current landscape of practices fails to effectively regulate the MTE for adult AIG-NCII, as well as flawed assumptions explaining these gaps.", "date": "2025-04-24", "categories": []}
{"id": "2506.13082", "title": "Discerning What Matters: A Multi-Dimensional Assessment of Moral Competence in LLMs", "abstract": "Moral competence is the ability to act in accordance with moral principles. As large language models (LLMs) are increasingly deployed in situations demanding moral competence, there is increasing interest in evaluating this ability empirically. We review existing literature and identify three significant shortcoming: (i) Over-reliance on prepackaged moral scenarios with explicitly highlighted moral features; (ii) Focus on verdict prediction rather than moral reasoning; and (iii) Inadequate testing of models' (in)ability to recognize when additional information is needed. Grounded in philosophical research on moral skill, we then introduce a novel method for assessing moral competence in LLMs. Our approach moves beyond simple verdict comparisons to evaluate five dimensions of moral competence: identifying morally relevant features, weighting their importance, assigning moral reasons to these features, synthesizing coherent moral judgments, and recognizing information gaps. We conduct two experiments comparing six leading LLMs against non-expert humans and professional philosophers. In our first experiment using ethical vignettes standard to existing work, LLMs generally outperformed non-expert humans across multiple dimensions of moral reasoning. However, our second experiment, featuring novel scenarios designed to test moral sensitivity by embedding relevant features among irrelevant details, revealed a striking reversal: several LLMs performed significantly worse than humans. Our findings suggest that current evaluations may substantially overestimate LLMs' moral reasoning capabilities by eliminating the task of discerning moral relevance from noisy information, which we take to be a prerequisite for genuine moral skill. This work provides a more nuanced framework for assessing AI moral competence and highlights important directions for improving moral competence in advanced AI systems.", "date": "2025-06-16", "categories": []}
{"id": "2603.05723", "title": "Cultural Perspectives and Expectations for Generative AI: A Global Survey Approach", "abstract": "There is a lack of empirical evidence about global attitudes around whether and how GenAI should represent cultures. This paper assesses understandings and beliefs about culture as it relates to GenAI from a large-scale global survey. We gathered data about what culture means to different groups, and about how GenAI should approach the representation of cultural artifacts, concepts, or values. We distill working definitions of culture directly from these communities to build an understanding of its conceptual complexities and how they relate to representations in Generative AI. We survey from across parts of Europe, North and South America, Asia, and Africa. We conclude with a set of recommendations for Culture and GenAI development. These include participatory approaches, prioritizing specific cultural dimensions beyond geography, such as religion and tradition, and a sensitivity framework for addressing cultural ``redlines''.", "date": "2026-03-05", "categories": []}
{"id": "2603.03515", "title": "The Controllability Trap: A Governance Framework for Military AI Agents", "abstract": "Agentic AI systems - capable of goal interpretation, world modeling, planning, tool use, long-horizon operation, and autonomous coordination - introduce distinct control failures not addressed by existing safety frameworks. We identify six agentic governance failures tied to these capabilities and show how they erode meaningful human control in military settings. We propose the Agentic Military AI Governance Framework (AMAGF), a measurable architecture structured around three pillars: Preventive Governance (reducing failure likelihood), Detective Governance (real-time detection of control degradation), and Corrective Governance (restoring or safely degrading operations). Its core mechanism, the Control Quality Score (CQS), is a composite real-time metric quantifying human control and enabling graduated responses as control weakens. For each failure type, we define concrete mechanisms, assign responsibilities across five institutional actors, and formalize evaluation metrics. A worked operational scenario illustrates implementation, and we situate the framework within established agent safety literature. We argue that governance must move from a binary conception of control to a continuous model in which control quality is actively measured and managed throughout the operational lifecycle.", "date": "2026-03-03", "categories": []}
{"id": "2603.04217", "title": "When Do Language Models Endorse Limitations on Human Rights Principles?", "abstract": "As Large Language Models (LLMs) increasingly mediate global information access with the potential to shape public discourse, their alignment with universal human rights principles becomes important to ensure that these rights are abided by in high stakes AI-mediated interactions. In this paper, we evaluate how LLMs navigate trade-offs involving the Universal Declaration of Human Rights (UDHR), leveraging 1,152 synthetically generated scenarios across 24 rights articles and eight languages. Our analysis of eleven major LLMs reveals systematic biases where models: (1) accept limiting Economic, Social, and Cultural rights more often than Political and Civil rights, (2) demonstrate significant cross-linguistic variation with elevated endorsement rates of rights-limiting actions in Chinese and Hindi compared to English or Romanian, (3) show substantial susceptibility to prompt-based steering, and (4) exhibit noticeable differences between Likert and open-ended responses, highlighting critical challenges in LLM preference assessment.", "date": "2026-03-04", "categories": []}
{"id": "2509.01938", "title": "EigenBench: A Comparative Behavioral Measure of Value Alignment", "abstract": "Aligning AI with human values is a pressing unsolved problem. To address the lack of quantitative metrics for value alignment, we propose EigenBench: a black-box method for comparatively benchmarking language models' values. Given an ensemble of models, a constitution describing a value system, and a dataset of scenarios, our method returns a vector of scores quantifying each model's alignment to the given constitution. To produce these scores, each model judges the outputs of other models across many scenarios, and these judgments are aggregated with EigenTrust (Kamvar et al., 2003), yielding scores that reflect a weighted consensus judgment of the whole ensemble. EigenBench uses no ground truth labels, as it is designed to quantify subjective traits for which reasonable judges may disagree on the correct label. Hence, to validate our method, we collect human judgments on the same ensemble of models and show that EigenBench's judgments align closely with those of human evaluators. We further demonstrate that EigenBench can recover model rankings on the GPQA benchmark without access to objective labels, supporting its viability as a framework for evaluating subjective values for which no ground truths exist. The code is available at https://github.com/jchang153/EigenBench.", "date": "2025-09-02", "categories": []}
{"id": "2602.23239", "title": "Agency and Architectural Limits: Why Optimization-Based Systems Cannot Be Norm-Responsive", "abstract": "AI systems are increasingly deployed in high-stakes contexts (medical diagnosis, legal research, financial analysis) under the assumption they can be governed by norms. This paper demonstrates that the assumption is formally invalid for optimization-based systems, specifically Large Language Models trained via Reinforcement Learning from Human Feedback (RLHF). Genuine agency requires two necessary and jointly sufficient architectural conditions. First, the capacity to maintain certain boundaries as non-negotiable constraints rather than tradeable weights (Incommensurability). Second, a non-inferential mechanism capable of suspending processing when those boundaries are threatened (Apophatic Responsiveness). RLHF-based systems are constitutively incompatible with both conditions. The operations that make optimization powerful, unifying all values on a scalar metric and always selecting the highest-scoring output, are precisely the operations that preclude normative governance and agency. This incompatibility is not a correctable training bug awaiting a technical fix. It is a formal constraint inherent to what optimization is. Consequently, documented failure modes (sycophancy, hallucination, and unfaithful reasoning) are not accidents but expected structural manifestations. Misaligned deployment triggers a second-order risk termed the Convergence Crisis. When humans are forced to verify AI outputs under metric pressure, they degrade from genuine agents into criteria-checking optimizers, eliminating the only component capable of bearing normative accountability. Beyond the incompatibility proof, this paper's primary positive contribution is a substrate-neutral architectural specification deriving what any system (biological, artificial, or institutional) must necessarily satisfy to qualify as a genuine agent rather than a sophisticated instrument.", "date": "2026-02-26", "categories": []}
{"id": "2603.00048", "title": "MOSAIC: Unveiling the Moral, Social and Individual Dimensions of Large Language Models", "abstract": "Large Language Models (LLMs) are increasingly deployed in sensitive applications including psychological support, healthcare, and high-stakes decision-making. This expansion has motivated growing research into the ethical and moral foundations underlying LLM behavior, raising critical questions about their reliability in ethical reasoning. However, existing studies and benchmarks rely almost exclusively on Moral Foundation Theory (MFT), largely neglecting other relevant dimensions such as social values, personality traits, and individual characteristics that shape human ethical reasoning. To address these limitations, we introduce MOSAIC, the first large-scale benchmark designed to jointly assess the moral, social, and individual characteristics of LLMs. The benchmark comprises nine validated questionnaires drawn from moral philosophy, psychology, and social theory, alongside four platform-based games designed to probe morally ambiguous scenarios. In total, MOSAIC includes over 600 curated questions and scenarios, released as a ready-to-use, extensible resource for evaluating the behavioral foundations of LLMs. We validate the benchmark across three models from different families, demonstrating its utility across all assessed dimensions and providing the first empirical evidence that MFT alone is insufficient to comprehensively evaluate complex AI systems' ethical behavior. We publicly release the dataset and our benchmark Python library.", "date": "2026-02-09", "categories": []}
{"id": "2603.00066", "title": "Contesting Artificial Moral Agents", "abstract": "There has been much discourse on the ethics of AI, to the extent that there are now systems that possess inherent moral reasoning. Such machines are now formally known as Artificial Moral Agents or AMAs. However, there is a requirement for a dedicated framework that can contest the morality of these systems. This paper proposes a 5E framework for contesting AMAs based on five grounds: ethical, epistemological, explainable, empirical, and evaluative. It further includes the spheres of ethical influences at individual, local, societal, and global levels. Lastly, the framework contributes a provisional timeline that indicates where developers of AMA technologies may anticipate contestation, or may self-contest in order to adhere to value-aligned development of truly moral AI systems.", "date": "2026-02-10", "categories": []}
{"id": "2603.00078", "title": "Alignment Is Not Enough: A Relational Framework for Moral Standing in Human-AI Interaction", "abstract": "The question of whether artificial entities deserve moral consideration has become one of the defining ethical challenges of AI research. Existing frameworks for moral patiency rely on verified ontological properties, such as sentience, phenomenal consciousness, or the capacity for suffering, that remain epistemically inaccessible in computational systems. This reliance creates a governance vacuum: millions of users form sustained affective bonds with conversational AI, yet no regulatory instrument distinguishes these interactions from transactional tool use. We introduce Relate (Relational Ethics for Leveled Assessment of Technological Entities), a framework that reframes AI moral patiency from ontological verification toward relational capacity and embodied interaction. Through a systematic comparison of seven governance frameworks, we demonstrate that current trustworthy AI instruments treat all human-AI encounters identically as tool use, ignoring the relational and embodied dynamics that posthumanist scholarship anticipated. We propose relational impact assessments, graduated moral consideration protocols, and interdisciplinary ethics integration as concrete instruments, and we include a sample Relational Impact Assessment applied to a deployed companion AI system. We do not claim current AI systems are conscious. We demonstrate that the ethical vocabularies governing them are inadequate to the embodied, relational realities these systems produce.", "date": "2026-02-13", "categories": []}
{"id": "2603.00590", "title": "Fair in Mind, Fair in Action? A Synchronous Benchmark for Understanding and Generation in UMLLMs", "abstract": "As artificial intelligence (AI) is increasingly deployed across domains, ensuring fairness has become a core challenge. However, the field faces a \"Tower of Babel'' dilemma: fairness metrics abound, yet their underlying philosophical assumptions often conflict, hindering unified paradigms-particularly in unified Multimodal Large Language Models (UMLLMs), where biases propagate systemically across tasks. To address this, we introduce the IRIS Benchmark, to our knowledge the first benchmark designed to synchronously evaluate the fairness of both understanding and generation tasks in UMLLMs. Enabled by our demographic classifier, ARES, and four supporting large-scale datasets, the benchmark is designed to normalize and aggregate arbitrary metrics into a high-dimensional \"fairness space'', integrating 60 granular metrics across three dimensions-Ideal Fairness, Real-world Fidelity, and Bias Inertia & Steerability (IRIS). Through this benchmark, our evaluation of leading UMLLMs uncovers systemic phenomena such as the \"generation gap'', individual inconsistencies like \"personality splits'', and the \"counter-stereotype reward'', while offering diagnostics to guide the optimization of their fairness capabilities. With its novel and extensible framework, the IRIS benchmark is capable of integrating evolving fairness metrics, ultimately helping to resolve the \"Tower of Babel'' impasse. Project Page: https://iris-benchmark-web.vercel.app/", "date": "2026-02-28", "categories": []}
{"id": "2505.19441", "title": "Fairness-in-the-Workflow: How Machine Learning Practitioners at Big Tech Companies Approach Fairness in Recommender Systems", "abstract": "Recommender systems (RS), which are widely deployed across high-stakes domains, are susceptible to biases that can cause large-scale societal impacts. Researchers have proposed methods to measure and mitigate such biases - but translating academic theory into practice is inherently challenging. Through a semi-structured interview study (N=11), we map the RS practitioner workflow within large technology companies, focusing on how technical teams consider fairness internally and in collaboration with legal, data, and fairness teams. We identify key challenges to incorporating fairness into existing RS workflows: defining fairness in RS contexts, balancing multi-stakeholder interests, and navigating dynamic environments. We also identify key organization-wide challenges: making time for fairness work and facilitating cross-team communication. Finally, we offer actionable recommendations for the RS community, including practitioners and HCI researchers.", "date": "2025-05-26", "categories": []}
{"id": "2602.21746", "title": "fEDM+: A Risk-Based Fuzzy Ethical Decision Making Framework with Principle-Level Explainability and Pluralistic Validation", "abstract": "In a previous work, we introduced the fuzzy Ethical Decision-Making framework (fEDM), a risk-based ethical reasoning architecture grounded in fuzzy logic. The original model combined a fuzzy Ethical Risk Assessment module (fERA) with ethical decision rules, enabled formal structural verification through Fuzzy Petri Nets (FPNs), and validated outputs against a single normative referent. Although this approach ensured formal soundness and decision consistency, it did not fully address two critical challenges: principled explainability of decisions and robustness under ethical pluralism. In this paper, we extend fEDM in two major directions. First, we introduce an Explainability and Traceability Module (ETM) that explicitly links each ethical decision rule to the underlying moral principles and computes a weighted principle-contribution profile for every recommended action. This enables transparent, auditable explanations that expose not only what decision was made but why, and on the basis of which principles. Second, we replace single-referent validation with a pluralistic semantic validation framework that evaluates decisions against multiple stakeholder referents, each encoding distinct principle priorities and risk tolerances. This shift allows principled disagreement to be formally represented rather than suppressed, thus increasing robustness and contextual sensitivity. The resulting extended fEDM, called fEDM+, preserves formal verifiability while achieving enhanced interpretability and stakeholder-aware validation, making it suitable as an oversight and governance layer for ethically sensitive AI systems.", "date": "2026-02-25", "categories": []}
{"id": "2506.16782", "title": "What Is the Point of Equality in Machine Learning Fairness? Beyond Equality of Opportunity", "abstract": "Fairness in machine learning (ML) has become a rapidly growing area of research. But why, in the first place, is unfairness in ML wrong? And why should we care about improving fairness? Most fair-ML research implicitly appeals to distributive equality: the idea that desirable benefits and goods, such as opportunities (e.g., Barocas et al., 2023), should be equally distributed across society. Unfair ML models, then, are seen as wrong because they unequally distribute such benefits. This paper argues that this exclusive focus on distributive equality offers an incomplete and potentially misleading ethical foundation. Grounding ML fairness in egalitarianism--the view that equality is a fundamental moral and social ideal--requires challenging structural inequality: systematic, institutional, and durable arrangements that privilege some groups while disadvantaging others. Structural inequality manifests through ML systems in two primary forms: allocative harms (e.g., economic loss) and representational harms (e.g., stereotypes, erasure). While distributive equality helps address allocative harms, it fails to explain why representational harms are wrong--why it is wrong for ML systems to reinforce social hierarchies that stratify people into superior and inferior groups--and why ML systems should aim to foster a society where people relate as equals (i.e., relational equality). To address these limitations, the paper proposes a multifaceted egalitarian framework for ML fairness that integrates both distributive and relational equality. Drawing on critical social and political philosophy, this framework offers a more comprehensive ethical foundation for tackling the full spectrum of harms perpetuated by ML systems. The paper also outlines practical pathways for implementing the framework across the entire ML pipeline.", "date": "2025-06-20", "categories": []}
{"id": "2509.17956", "title": "\"I think this is fair\": Uncovering the Complexities of Stakeholder Decision-Making in AI Fairness Assessment", "abstract": "Assessing fairness in artificial intelligence (AI) typically involves AI experts who select protected features, fairness metrics, and set fairness thresholds to assess outcome fairness. However, little is known about how stakeholders, particularly those affected by AI outcomes but lacking AI expertise, assess fairness. To address this gap, we conducted a qualitative study with 26 stakeholders without AI expertise, representing potential decision subjects in a credit rating scenario, to examine how they assess fairness when placed in the role of deciding on features with priority, metrics, and thresholds. We reveal that stakeholders' fairness decisions are more complex than typical AI expert practices: they considered features far beyond legally protected features, tailored metrics for specific contexts, set diverse yet stricter fairness thresholds, and even preferred designing customized fairness. Our results extend the understanding of how stakeholders can meaningfully contribute to AI fairness governance and mitigation, underscoring the importance of incorporating stakeholders' nuanced fairness judgments.", "date": "2025-09-22", "categories": []}
{"id": "2304.14347", "title": "The Dark Side of ChatGPT: Legal and Ethical Challenges from Stochastic Parrots and Hallucination", "abstract": "With the launch of ChatGPT, Large Language Models (LLMs) are shaking up our whole society, rapidly altering the way we think, create and live. For instance, the GPT integration in Bing has altered our approach to online searching. While nascent LLMs have many advantages, new legal and ethical risks are also emerging, stemming in particular from stochastic parrots and hallucination. The EU is the first and foremost jurisdiction that has focused on the regulation of AI models. However, the risks posed by the new LLMs are likely to be underestimated by the emerging EU regulatory paradigm. Therefore, this correspondence warns that the European AI regulatory paradigm must evolve further to mitigate such risks.", "date": "2023-04-21", "categories": []}
{"id": "2602.17729", "title": "Stop Saying \"AI\"", "abstract": "Across academia, industry, and government, ``AI'' has become central in research and development, regulatory debates, and promises of ever faster and more capable decision-making and action. In numerous domains, especially safety-critical ones, there are significant concerns over how ``AI'' may affect decision-making, responsibility, or the likelihood of mistakes (to name only a few categories of critique). However, for most critiques, the target is generally ``AI'', a broad term admitting many (types of) systems used for a variety of tasks and each coming with its own set of limitations, challenges, and potential use cases. In this article, we focus on the military domain as a case study and present both a loose enumerative taxonomy of systems captured under the umbrella term ``military AI'', as well as discussion of the challenges of each. In doing so, we highlight that critiques of one (type of) system will not always transfer to other (types of) systems. Building on this, we argue that in order for debates to move forward fruitfully, it is imperative that the discussions be made more precise and that ``AI'' be excised from debates to the extent possible. Researchers, developers, and policy-makers should make clear exactly what systems they have in mind and what possible benefits and risks attend the deployment of those particular systems. While we focus on AI in the military as an exemplar for the overall trends in discussions of ``AI'', the argument's conclusions are broad and have import for discussions of AI across a host of domains.", "date": "2026-02-18", "categories": []}
{"id": "2505.20181", "title": "The Problem of Algorithmic Collisions: Mitigating Unforeseen Risks in a Connected World", "abstract": "The increasing deployment of Artificial Intelligence (AI) and other autonomous algorithmic systems presents the world with new systemic risks. While focus often lies on the function of individual algorithms, a critical and underestimated danger arises from their interactions, particularly when algorithmic systems operate without awareness of each other, or when those deploying them are unaware of the full algorithmic ecosystem deployment is occurring in. These interactions can lead to unforeseen, rapidly escalating negative outcomes - from market crashes and energy supply disruptions to potential physical accidents and erosion of public trust - often exceeding the human capacity for effective monitoring and the legal capacities for proper intervention. Current governance frameworks are inadequate as they lack visibility into this complex ecosystem of interactions. This paper outlines the nature of this challenge and proposes some initial policy suggestions centered on increasing transparency and accountability through phased system registration, a licensing framework for deployment, and enhanced monitoring capabilities.", "date": "2025-05-26", "categories": []}
{"id": "2602.07754", "title": "Humanizing AI Grading: Student-Centered Insights on Fairness, Trust, Consistency and Transparency", "abstract": "This study investigates students' perceptions of Artificial Intelligence (AI) grading systems in an undergraduate computer science course (n = 27), focusing on a block-based programming final project. Guided by the ethical principles framework articulated by Jobin (2019), our study examines fairness, trust, consistency, and transparency in AI grading by comparing AI-generated feedback with original human-graded feedback. Findings reveal concerns about AI's lack of contextual understanding and personalization. We recommend that equitable and trustworthy AI systems reflect human judgment, flexibility, and empathy, serving as supplementary tools under human oversight. This work contributes to ethics-centered assessment practices by amplifying student voices and offering design principles for humanizing AI in designed learning environments.", "date": "2026-02-08", "categories": []}
{"id": "2602.18459", "title": "From Bias Mitigation to Bias Negotiation: Governing Identity and Sociocultural Reasoning in Generative AI", "abstract": "LLMs act in the social world by drawing upon shared cultural patterns to make social situations understandable and actionable. Because identity is often part of the inferential substrate of competent judgment, ethical alignment requires regulating when and how systems invoke identity. Yet the dominant governance regime for identity-related harm remains bias mitigation, which treats identity primarily as a source of measurable disparities or harmful associations to be detected and suppressed. This leaves underspecified a positive, context-sensitive role for identity in interpretation. We call this governance problem bias negotiation: the normative regulation of identity-conditioned judgments of sociocultural relevance, inference, and justification. Empirically, we probe the feasibility of bias negotiation through semi-structured interviews with multiple publicly deployed chatbots. We identify recurring repertoires for negotiating identity including probabilistic framing of group tendencies and harm-value balancing. We also observe failure modes in which models avoid hard tradeoffs or apply principles inconsistently. Bias negotiation matters for justice because a positive role for sociocultural reasoning is required to recognize and potentially remediate structural inequities. But it is equally implicated in core model functionality as sociocultural competence is needed for systems that operate across heterogeneous cultural contexts. Because bias negotiation is a procedural capability expressed through deliberation and interaction, it cannot be validated by static benchmarks alone. To support targeted training, we introduce a broad but explicit framework that decomposes bias negotiation into an action space of negotiation moves (what to observe and score) and a complementary set of case features (over which the model negotiates), enabling systematic test-suite design and evaluation.", "date": "2026-02-05", "categories": []}
{"id": "2602.19629", "title": "Cooperation After the Algorithm: Designing Human-AI Coexistence Beyond the Illusion of Collaboration", "abstract": "Generative artificial intelligence systems increasingly participate in research, law, education, media, and governance. Their fluent and adaptive outputs create an experience of collaboration. However, these systems do not bear responsibility, incur liability, or share stakes in downstream consequences. This structural asymmetry has already produced sanctions, professional errors, and governance failures in high-stakes contexts We argue that stable human-AI coexistence is an institutional achievement that depends on governance infrastructure capable of distributing residual risk. Drawing on institutional analysis and evolutionary cooperation theory, we introduce a formal inequality that specifies when reliance on AI yields positive expected cooperative value. The model makes explicit how governance conditions, system policy, and accountability regimes jointly determine whether cooperation is rational or structurally defective. From this formalization we derive a cooperation ecology framework with six design principles: reciprocity contracts, visible trust infrastructure, conditional cooperation modes, defection-mitigation mechanisms, narrative literacy against authority theatre, and an Earth-first sustainability constraint. We operationalize the framework through three policy artefacts: a Human-AI Cooperation Charter, a Defection Risk Register, and a Cooperation Readiness Audit. Together, these elements shift the unit of analysis from the user-AI dyad to the institutional environment that shapes incentives, signals, accountability, and repair. The paper provides a theoretical foundation and practical toolkit for designing human-AI systems that can sustain accountable, trustworthy cooperation over time.", "date": "2026-02-23", "categories": []}
{"id": "2411.06624", "title": "A Review of Fairness and A Practical Guide to Selecting Context-Appropriate Fairness Metrics in Machine Learning", "abstract": "Recent regulatory proposals for artificial intelligence emphasize fairness requirements for machine learning models. However, precisely defining the appropriate measure of fairness is challenging due to philosophical, cultural and political contexts. Biases can infiltrate machine learning models in complex ways depending on the model's context, rendering a single common metric of fairness insufficient. This ambiguity highlights the need for criteria to guide the selection of context-aware measures, an issue of increasing importance given the proliferation of ever tighter regulatory requirements. To address this, we developed a flowchart to guide the selection of contextually appropriate fairness measures. Twelve criteria were used to formulate the flowchart. This included consideration of model assessment criteria, model selection criteria, and data bias. We also review fairness literature in the context of machine learning and link it to core regulatory instruments to assist policymakers, AI developers, researchers, and other stakeholders in appropriately addressing fairness concerns and complying with relevant regulatory requirements.", "date": "2024-11-10", "categories": []}
{"id": "2512.19117", "title": "Stop saying LLM: Large Discourse Models (LDM) and Artificial Discursive Agent (ADA)?", "abstract": "This paper proposes an epistemological shift in the analysis of large generative models, replacing the category ''Large Language Models'' (LLM) with that of ''Large Discourse Models'' (LDM), and then with that of Artificial Discursive Agent (ADA). The theoretical framework is based on an ontological triad distinguishing three regulatory instances: the apprehension of the phenomenal regularities of the referential world, the structuring of embodied cognition, and the structural-linguistic sedimentation of the utterance within a socio-historical context. LDMs, operating on the product of these three instances (the document), model the discursive projection of a portion of human experience reified by the learning corpus. The proposed program aims to replace the ''fascination/fear'' dichotomy with public trials and procedures that make the place, uses, and limits of artificial discursive agents in contemporary social space decipherable, situating this approach within a perspective of governance and co-regulation involving the State, industry, civil society, and academia.", "date": "2025-12-22", "categories": []}
{"id": "2502.01713", "title": "Auditing a Dutch Public Sector Risk Profiling Algorithm Using an Unsupervised Bias Detection Tool", "abstract": "Algorithms are increasingly used to automate or aid human decisions, yet recent research shows that these algorithms may exhibit bias across legally protected demographic groups. However, data on these groups may be unavailable to organizations or external auditors due to privacy legislation. This paper studies bias detection using an unsupervised bias detection tool when data on demographic groups are unavailable. We collaborated with the Dutch Executive Agency for Education to audit an algorithm that was used to assign risk scores to college students at the national level in the Netherlands between 2012-2023. Our audit covers more than 250,000 students across the country. The unsupervised bias detection tool highlights known disparities between students with a non-European migration background and students with a Dutch or European-migration background. Our contributions are two-fold: (1) we assess bias in a real-world, large-scale, and high-stakes decision-making process by a governmental organization; (2) we provide the unsupervised bias detection tool in an open-source library for others to use to complete bias audits. Our work serves as a starting point for a deliberative assessment by human experts to evaluate potential discrimination in algorithmic decision-making.", "date": "2025-02-03", "categories": []}
{"id": "2601.08005", "title": "Internal Deployment Gaps in AI Regulation", "abstract": "Frontier AI regulations primarily focus on systems deployed to external users, where deployment is more visible and subject to outside scrutiny. However, high-stakes applications can occur internally when companies deploy highly capable systems within their own organizations, such as for automating R&D, accelerating critical business processes, and handling sensitive proprietary data. This paper examines how frontier AI regulations in the United States and European Union in 2025 handle internal deployment. We identify three gaps that could cause internally-deployed systems to evade intended oversight: (1) scope ambiguity that allows internal systems to evade regulatory obligations, (2) point-in-time compliance assessments that fail to capture the continuous evolution of internal systems, and (3) information asymmetries that subvert regulatory awareness and oversight. We then analyze why these gaps persist, examining tensions around measurability, incentives, and information access. Finally, we map potential approaches to address them and their associated tradeoffs. By understanding these patterns, we hope that policy choices around internally deployed AI systems can be made deliberately rather than incidentally.", "date": "2026-01-12", "categories": []}
{"id": "2602.13246", "title": "Global AI Bias Audit for Technical Governance", "abstract": "This paper presents the outputs of the exploratory phase of a global audit of Large Language Models (LLMs) project. In this exploratory phase, I used the Global AI Dataset (GAID) Project as a framework to stress-test the Llama-3 8B model and evaluate geographic and socioeconomic biases in technical AI governance awareness. By stress-testing the model with 1,704 queries across 213 countries and eight technical metrics, I identified a significant digital barrier and gap separating the Global North and South. The results indicate that the model was only able to provide number/fact responses in 11.4% of its query answers, where the empirical validity of such responses was yet to be verified. The findings reveal that AI's technical knowledge is heavily concentrated in higher-income regions, while lower-income countries from the Global South are subject to disproportionate systemic information gaps. This disparity between the Global North and South poses concerning risks for global AI safety and inclusive governance, as policymakers in underserved regions may lack reliable data-driven insights or be misled by hallucinated facts. This paper concludes that current AI alignment and training processes reinforce existing geoeconomic and geopolitical asymmetries, and urges the need for more inclusive data representation to ensure AI serves as a truly global resource.", "date": "2026-02-01", "categories": []}
{"id": "2602.13268", "title": "Expected Moral Shortfall for Ethical Competence in Decision-making Models", "abstract": "Moral cognition is a crucial yet underexplored aspect of decision-making in AI models. Regardless of the application domain, it should be a consideration that allows for ethically aligned decision-making. This paper presents a multifaceted contribution to this research space. Firstly, a comparative analysis of techniques to instill ethical competence into AI models has been presented to gauge them on multiple performance metrics. Second, a novel mathematical discretization of morality and a demonstration of its real-life application have been conveyed and tested against other techniques on two datasets. This value is modeled as the risk of loss incurred by the least moral cases, or an Expected Moral Shortfall (EMS), which we direct the AI model to minimize in order to maximize its performance while retaining ethical competence. Lastly, the paper discusses the tradeoff between preliminary AI decision-making metrics such as model performance, complexity, and scale of ethical competence to recognize the true extent of practical social impact.", "date": "2026-02-04", "categories": []}
{"id": "2602.13292", "title": "Mirror: A Multi-Agent System for AI-Assisted Ethics Review", "abstract": "Ethics review is a foundational mechanism of modern research governance, yet contemporary systems face increasing strain as ethical risks arise as structural consequences of large-scale, interdisciplinary scientific practice. The demand for consistent and defensible decisions under heterogeneous risk profiles exposes limitations in institutional review capacity rather than in the legitimacy of ethics oversight. Recent advances in large language models (LLMs) offer new opportunities to support ethics review, but their direct application remains limited by insufficient ethical reasoning capability, weak integration with regulatory structures, and strict privacy constraints on authentic review materials. In this work, we introduce Mirror, an agentic framework for AI-assisted ethical review that integrates ethical reasoning, structured rule interpretation, and multi-agent deliberation within a unified architecture. At its core is EthicsLLM, a foundational model fine-tuned on EthicsQA, a specialized dataset of 41K question-chain-of-thought-answer triples distilled from authoritative ethics and regulatory corpora. EthicsLLM provides detailed normative and regulatory understanding, enabling Mirror to operate in two complementary modes. Mirror-ER (expedited Review) automates expedited review through an executable rule base that supports efficient and transparent compliance checks for minimal-risk studies. Mirror-CR (Committee Review) simulates full-board deliberation through coordinated interactions among expert agents, an ethics secretary agent, and a principal investigator agent, producing structured, committee-level assessments across ten ethical dimensions. Empirical evaluations demonstrate that Mirror significantly improves the quality, consistency, and professionalism of ethics assessments compared with strong generalist LLMs.", "date": "2026-02-09", "categories": []}
{"id": "2510.08193", "title": "Measuring What Matters: The AI Pluralism Index", "abstract": "Artificial intelligence systems increasingly mediate knowledge, communication, and decision making. Development and governance remain concentrated within a small set of firms and states, raising concerns that technologies may encode narrow interests and limit public agency. Capability benchmarks for language, vision, and coding are common, yet public, auditable measures of pluralistic governance are rare. We define AI pluralism as the degree to which affected stakeholders can shape objectives, data practices, safeguards, and deployment. We present the AI Pluralism Index (AIPI), a transparent, evidence-based instrument that evaluates producers and system families across four pillars: participatory governance, inclusivity and diversity, transparency, and accountability. AIPI codes verifiable practices from public artifacts and independent evaluations, explicitly handling \"Unknown\" evidence to report both lower-bound (\"evidence\") and known-only scores with coverage. We formalize the measurement model; implement a reproducible pipeline that integrates structured web and repository analysis, external assessments, and expert interviews; and assess reliability with inter-rater agreement, coverage reporting, cross-index correlations, and sensitivity analysis. The protocol, codebook, scoring scripts, and evidence graph are maintained openly with versioned releases and a public adjudication process. We report pilot provider results and situate AIPI relative to adjacent transparency, safety, and governance frameworks. The index aims to steer incentives toward pluralistic practice and to equip policymakers, procurers, and the public with comparable evidence.", "date": "2025-10-09", "categories": []}
{"id": "2602.10916", "title": "Traceable, Enforceable, and Compensable Participation: A Participation Ledger for People-Centered AI Governance", "abstract": "Participatory approaches are widely invoked in AI governance, yet participation rarely translates into durable influence. In public sector and civic AI systems, community contributions such as deliberations, annotations, prompts, and incident reports are often recorded informally, weakly linked to system updates, and disconnected from enforceable rights or sustained compensation. As a result, participation is frequently symbolic rather than accountable. We introduce the Participation Ledger, a machine readable and auditable framework that operationalizes participation as traceable influence, enforceable authority, and compensable labor. The ledger represents participation as an influence graph that links contributed artifacts to verified changes in AI systems, including datasets, prompts, adapters, policies, guardrails, and evaluation suites. It integrates three elements: a Participation Evidence Standard documenting consent, privacy, compensation, and reuse terms; an influence tracing mechanism that connects system updates to replayable before and after tests, enabling longitudinal monitoring of commitments; and encoded rights and incentives. Capability Vouchers allow authorized community stewards to request or constrain specific system capabilities within defined boundaries, while Participation Credits support ongoing recognition and compensation when contributed tests continue to provide value. We ground the framework in four urban AI and public space governance deployments and provide a machine readable schema, templates, and an evaluation plan for assessing traceability, enforceability, and compensation in practice.", "date": "2026-02-11", "categories": []}
{"id": "2602.09202", "title": "Genocide by Algorithm in Gaza: Artificial Intelligence, Countervailing Responsibility, and the Corruption of Public Discourse", "abstract": "The accelerating militarization of artificial intelligence has transformed the ethics, politics, and governance of warfare. This article interrogates how AI-driven targeting systems function as epistemic infrastructures that classify, legitimize, and execute violence, using Israel's conduct in Gaza as a paradigmatic case. Through the lens of responsibility, the article examines three interrelated dimensions: (a) political responsibility, exploring how states exploit AI to accelerate warfare while evading accountability; (b) professional responsibility, addressing the complicity of technologists, engineers, and defense contractors in the weaponization of data; and (c) personal responsibility, probing the moral agency of individuals who participate in or resist algorithmic governance. This is complemented by an examination of the position and influence of those participating in public discourse, whose narratives often obscure or normalize AI-enabled violence. The Gaza case reveals AI not as a neutral instrument but as an active participant in the reproduction of colonial hierarchies and the normalization of atrocity. Ultimately, the paper calls for a reframing of technological agency and accountability in the age of automated warfare. It concludes that confronting algorithmic violence demands a democratization of AI ethics, one that resists technocratic fatalism and centers the lived realities of those most affected by high-tech militarism.", "date": "2026-02-09", "categories": []}
{"id": "2503.07660", "title": "Research Superalignment Should Advance Now with Alternating Competence and Conformity Optimization", "abstract": "The recent leap in AI capabilities, driven by big generative models, has sparked the possibility of achieving Artificial General Intelligence (AGI) and further triggered discussions on Artificial Superintelligence (ASI)-a system surpassing all humans across measured domains. This gives rise to the critical research question of: As we approach ASI, how do we align it with human values, ensuring it benefits rather than harms human society, a.k.a., the Superalignment problem. Despite ASI being regarded by many as a hypothetical concept, in this position paper, we argue that superalignment is achievable and research on it should advance immediately, through simultaneous and alternating optimization of task competence and value conformity. We posit that superalignment is not merely a safeguard for ASI but also necessary for its responsible realization. To support this position, we first provide a formal definition of superalignment rooted in the gap between capability and capacity, delve into its perceived infeasibility by analyzing the limitations of existing paradigms, and then illustrate a conceptual path of superalignment to support its achievability, centered on two fundamental principles. This work frames a potential initiative for developing value-aligned next-generation AI in the future, which will garner greater benefits and reduce potential harm to humanity.", "date": "2025-03-08", "categories": []}
{"id": "2602.06981", "title": "What is Safety? Corporate Discourse, Power, and the Politics of Generative AI Safety", "abstract": "This work examines how leading generative artificial intelligence companies construct and communicate the concept of \"safety\" through public-facing documents. Drawing on critical discourse analysis, we analyze a corpus of corporate safety-related statements to explicate how authority, responsibility, and legitimacy are discursively established. These discursive strategies consolidate legitimacy for corporate actors, normalize safety as an experimental and anticipatory practice, and push a perceived participatory agenda toward safe technologies. We argue that uncritical uptake of these discourses risks reproducing corporate priorities and constraining alternative approaches to governance and design. The contribution of this work is twofold: first, to situate safety as a sociotechnical discourse that warrants critical examination; second, to caution human-computer interaction scholars against legitimizing corporate framings, instead foregrounding accountability, equity, and justice. By interrogating safety discourses as artifacts of power, this paper advances a critical agenda for human-computer interaction scholarship on artificial intelligence.", "date": "2026-01-26", "categories": []}
{"id": "2602.08706", "title": "Technosocial risks of ideal emotion recognition technologies: A defense of the (social) value of emotional expressions", "abstract": "The prospect of AI systems that I call ideal emotion recognition technologies (ERTs) is often defended on the assumption that social life would benefit from increased affective transparency. This paper challenges that assumption by examining the technosocial risks posed by ideal ERTs, understood as multimodal systems capable of reliably inferring inner affective states in real time. Drawing on philosophical accounts of emotional expression and social practice, as well as empirical work in affective science and social psychology, I argue that the appeal of such systems rests on a misunderstanding of the social functions of emotional expression. Emotional expressions function not only as read-outs of inner states, but also as tools for coordinating action, enabling moral repair, sustaining interpersonal trust, and supporting collective norms. These functions depend on a background of partial opacity and epistemic friction. When deployed in socially authoritative or evaluative contexts, ideal ERTs threaten this expressive space by collapsing epistemic friction, displacing relational meaning with technology-mediated affective profiles, and narrowing the space for aspirational and role-sensitive expressions. The result is a drift towards affective determinism and ambient forms of affective auditing, which undermine both social cohesion and individual agency. I argue that, although it is intuitive to think that increasing accuracy would legitimise such systems, in the case of ERTs accuracy does not straightforwardly justify their deployment, and may, in some contexts, provide a reason for regulatory restraint. I conclude by defending a function-first regulatory approach that treats expressive discretion and intentional emotional expression as constitutive of certain social goods, and that accordingly seeks to protect these goods from excessive affective legibility.", "date": "2026-02-09", "categories": []}
{"id": "2402.06359", "title": "Modelling Human Values for AI Reasoning", "abstract": "One of today's most significant societal challenges is building AI systems whose behaviour, or the behaviour it enables within communities of interacting agents (human and artificial), aligns with human values. To address this challenge, we detail a formal model of human values for their explicit computational representation. To our knowledge, this has not been attempted as yet, which is surprising given the growing volume of research integrating values within AI. Taking as our starting point the wealth of research investigating the nature of human values from social psychology over the last few decades, we set out to provide such a formal model. We show how this model can provide the foundational apparatus for AI-based reasoning over values, and demonstrate its applicability in real-world use cases. We illustrate how our model captures the key ideas from social psychology research and propose a roadmap for future integrated, and interdisciplinary, research into human values in AI. The ability to automatically reason over values not only helps address the value alignment problem but also facilitates the design of AI systems that can support individuals and communities in making more informed, value-aligned decisions. More and more, individuals and organisations are motivated to understand their values more explicitly and explore whether their behaviours and attitudes properly reflect them. Our work on modelling human values will enable AI systems to be designed and deployed to meet this growing need.", "date": "2024-02-09", "categories": []}
{"id": "2602.01503", "title": "Governance at the Edge of Architecture: Regulating NeuroAI and Neuromorphic Systems", "abstract": "Current AI governance frameworks, including regulatory benchmarks for accuracy, latency, and energy efficiency, are built for static, centrally trained artificial neural networks on von Neumann hardware. NeuroAI systems, embodied in neuromorphic hardware and implemented via spiking neural networks, break these assumptions. This paper examines the limitations of current AI governance frameworks for NeuroAI, arguing that assurance and audit methods must co-evolve with these architectures, aligning traditional regulatory metrics with the physics, learning dynamics, and embodied efficiency of brain-inspired computation to enable technically grounded assurance.", "date": "2026-02-02", "categories": []}
{"id": "2508.12300", "title": "Mutually Assured Deregulation", "abstract": "We have convinced ourselves that the way to make AI safe is to make it unsafe. Since 2022, policymakers worldwide have embraced the Regulation Sacrifice - the belief that dismantling safety oversight will deliver security through AI dominance. Fearing China or USA will gain advantage, nations rush to eliminate safeguards that might slow progress. This Essay reveals the fatal flaw: though AI poses national security challenges, the solution demands stronger regulatory frameworks, not weaker ones. A race without guardrails breeds shared danger, not competitive strength. The Regulation Sacrifice makes three false promises. First, it promises durable technological leads. But AI capabilities spread rapidly - performance gaps between U.S. and Chinese systems collapsed from 9 percent to 2 percent in thirteen months. When advantages evaporate in months, sacrificing permanent safety for temporary speed makes no sense. Second, it promises deregulation accelerates innovation. The opposite often proves true. Companies report well-designed governance streamlines development. Investment flows toward regulated markets. Clear rules reduce uncertainty; uncertain liability creates paralysis. Environmental standards did not kill the auto industry; they created Tesla and BYD. Third, enhanced national security through deregulation actually undermines security across all timeframes. Near term: it hands adversaries information warfare tools. Medium term: it democratizes bioweapon capabilities. Long term: it guarantees deployment of uncontrollable AGI systems. The Regulation Sacrifice persists because it serves powerful interests, not security. Tech companies prefer freedom to accountability. Politicians prefer simple stories to complex truths. This creates mutually assured deregulation, where each nation's sprint for advantage guarantees collective vulnerability. The only way to win is not to play.", "date": "2025-08-17", "categories": []}
{"id": "2304.04914", "title": "Regulatory Markets: The Future of AI Governance", "abstract": "Appropriately regulating artificial intelligence is an increasingly urgent and widespread policy challenge. We identify two primary, competing problem. First is a technical deficit: Legislatures and regulatory face significant challenges in rapidly translating conventional command-and-control legal requirements into technical requirements. Second is a democratic deficit: Over-reliance on industry to provide technical standards fails to ensure that the many values-based decisions that must be made to shape AI development and deployment are made by democratically accountable public, not private, actors. We propose a solution: regulatory markets, in which governments require the targets of regulation to purchase regulatory services from a government-licensed private regulator. This approach to AI regulation could overcome the limitations of both command-and-control regulation and excessive delegation to industry. Regulatory markets could enable governments to establish policy priorities for the regulation of AI while relying on market forces and industry R&D efforts to pioneer the technical methods of regulation that best achieve policymakers' stated objectives.", "date": "2023-04-11", "categories": []}
{"id": "2601.08951", "title": "PluriHarms: Benchmarking the Full Spectrum of Human Judgments on AI Harm", "abstract": "Current AI safety frameworks, which often treat harmfulness as binary, lack the flexibility to handle borderline cases where humans meaningfully disagree. To build more pluralistic systems, it is essential to move beyond consensus and instead understand where and why disagreements arise. We introduce PluriHarms, a benchmark designed to systematically study human harm judgments across two key dimensions -- the harm axis (benign to harmful) and the agreement axis (agreement to disagreement). Our scalable framework generates prompts that capture diverse AI harms and human values while targeting cases with high disagreement rates, validated by human data. The benchmark includes 150 prompts with 15,000 ratings from 100 human annotators, enriched with demographic and psychological traits and prompt-level features of harmful actions, effects, and values. Our analyses show that prompts that relate to imminent risks and tangible harms amplify perceived harmfulness, while annotator traits (e.g., toxicity experience, education) and their interactions with prompt content explain systematic disagreement. We benchmark AI safety models and alignment methods on PluriHarms, finding that while personalization significantly improves prediction of human harm judgments, considerable room remains for future progress. By explicitly targeting value diversity and disagreement, our work provides a principled benchmark for moving beyond \"one-size-fits-all\" safety toward pluralistically safe AI.", "date": "2026-01-13", "categories": []}
{"id": "2602.02511", "title": "Training Data Governance for Brain Foundation Models", "abstract": "Brain foundation models bring the foundation model paradigm to the field of neuroscience. Like language and image foundation models, they are general-purpose AI systems pretrained on large-scale datasets that adapt readily to downstream tasks. Unlike text-and-image based models, however, they train on brain data: large-datasets of EEG, fMRI, and other neural data types historically collected within tightly governed clinical and research settings. This paper contends that training foundation models on neural data opens new normative territory. Neural data carry stronger expectations of, and claims to, protection than text or images, given their body-derived nature and historical governance within clinical and research settings. Yet the foundation model paradigm subjects them to practices of large-scale repurposing, cross-context stitching, and open-ended downstream application. Furthermore, these practices are now accessible to a much broader range of actors, including commercial developers, against a backdrop of fragmented and unclear governance. To map this territory, we first describe brain foundation models' technical foundations and training-data ecosystem. We then draw on AI ethics, neuroethics, and bioethics to organize concerns across privacy, consent, bias, benefit sharing, and governance. For each, we propose both agenda-setting questions and baseline safeguards as the field matures.", "date": "2026-01-23", "categories": []}
{"id": "2504.08853", "title": "Artificial Intelligence (AI) and the Relationship between Agency, Autonomy, and Moral Patiency", "abstract": "The proliferation of Artificial Intelligence (AI) systems exhibiting complex and seemingly agentive behaviours necessitates a critical philosophical examination of their agency, autonomy, and moral status. In this paper we undertake a systematic analysis of the differences between basic, autonomous, and moral agency in artificial systems. We argue that while current AI systems are highly sophisticated, they lack genuine agency and autonomy because: they operate within rigid boundaries of pre-programmed objectives rather than exhibiting true goal-directed behaviour within their environment; they cannot authentically shape their engagement with the world; and they lack the critical self-reflection and autonomy competencies required for full autonomy. Nonetheless, we do not rule out the possibility of future systems that could achieve a limited form of artificial moral agency without consciousness through hybrid approaches to ethical decision-making. This leads us to suggest, by appealing to the necessity of consciousness for moral patiency, that such non-conscious AMAs might represent a case that challenges traditional assumptions about the necessary connection between moral agency and moral patiency.", "date": "2025-04-11", "categories": []}
{"id": "2601.10520", "title": "Breaking Up with Normatively Monolithic Agency with GRACE: A Reason-Based Neuro-Symbolic Architecture for Safe and Ethical AI Alignment", "abstract": "As AI agents become increasingly autonomous, widely deployed in consequential contexts, and efficacious in bringing about real-world impacts, ensuring that their decisions are not only instrumentally effective but also normatively aligned has become critical. We introduce a neuro-symbolic reason-based containment architecture, Governor for Reason-Aligned ContainmEnt (GRACE), that decouples normative reasoning from instrumental decision-making and can contain AI agents of virtually any design. GRACE restructures decision-making into three modules: a Moral Module (MM) that determines permissible macro actions via deontic logic-based reasoning; a Decision-Making Module (DMM) that encapsulates the target agent while selecting instrumentally optimal primitive actions in accordance with derived macro actions; and a Guard that monitors and enforces moral compliance. The MM uses a reason-based formalism providing a semantic foundation for deontic logic, enabling interpretability, contestability, and justifiability. Its symbolic representation enriches the DMM's informational context and supports formal verification and statistical guarantees of alignment enforced by the Guard. We demonstrate GRACE on an example of a LLM therapy assistant, showing how it enables stakeholders to understand, contest, and refine agent behavior.", "date": "2026-01-15", "categories": []}
{"id": "2601.11651", "title": "Aesthetics as Structural Harm: Algorithmic Lookism Across Text-to-Image Generation and Classification", "abstract": "This paper examines algorithmic lookism-the systematic preferential treatment based on physical appearance-in text-to-image (T2I) generative AI and a downstream gender classification task. Through the analysis of 26,400 synthetic faces created with Stable Diffusion 2.1 and 3.5 Medium, we demonstrate how generative AI models systematically associate facial attractiveness with positive attributes and vice-versa, mirroring socially constructed biases rather than evidence-based correlations. Furthermore, we find significant gender bias in three gender classification algorithms depending on the attributes of the input faces. Our findings reveal three critical harms: (1) the systematic encoding of attractiveness-positive attribute associations in T2I models; (2) gender disparities in classification systems, where women's faces, particularly those generated with negative attributes, suffer substantially higher misclassification rates than men's; and (3) intensifying aesthetic constraints in newer models through age homogenization, gendered exposure patterns, and geographic reductionism. These convergent patterns reveal algorithmic lookism as systematic infrastructure operating across AI vision systems, compounding existing inequalities through both representation and recognition.\n  Disclaimer: This work includes visual and textual content that reflects stereotypical associations between physical appearance and socially constructed attributes, including gender, race, and traits associated with social desirability. Any such associations found in this study emerge from the biases embedded in generative AI systems-not from empirical truths or the authors' views.", "date": "2026-01-15", "categories": []}
{"id": "2602.00497", "title": "Culturally-Grounded Governance for Multilingual Language Models: Rights, Data Boundaries, and Accountable AI Design", "abstract": "Multilingual large language models (MLLMs) are increasingly deployed across cultural, linguistic, and political contexts, yet existing governance frameworks largely assume English-centric data, homogeneous user populations, and abstract notions of fairness. This creates systematic risks for low-resource languages and culturally marginalized communities, where data practices, model behavior, and accountability mechanisms often fail to align with local norms, rights, and expectations. Drawing on cross-cultural perspectives in human-centered computing and AI governance, this paper synthesizes existing evidence on multilingual model behavior, data asymmetries, and sociotechnical harm, and articulates a culturally grounded governance framework for MLLMs. We identify three interrelated governance challenges: cultural and linguistic inequities in training data and evaluation practices, misalignment between global deployment and locally situated norms, values, and power structures, and limited accountability mechanisms for addressing harms experienced by marginalized language communities. Rather than proposing new technical benchmarks, we contribute a conceptual agenda that reframes multilingual AI governance as a sociocultural and rights based problem. We outline design and policy implications for data stewardship, transparency, and participatory accountability, and argue that culturally grounded governance is essential for ensuring that multilingual language models do not reproduce existing global inequalities under the guise of scale and neutrality.", "date": "2026-01-31", "categories": []}
{"id": "2503.11947", "title": "Ethical AI for Young Digital Citizens: A Call to Action on Privacy Governance", "abstract": "The rapid expansion of Artificial Intelligence (AI) in digital platforms used by youth has created significant challenges related to privacy, autonomy, and data protection. While AI-driven personalization offers enhanced user experiences, it often operates without clear ethical boundaries, leaving young users vulnerable to data exploitation and algorithmic biases. This paper presents a call to action for ethical AI governance, advocating for a structured framework that ensures youth-centred privacy protections, transparent data practices, and regulatory oversight. We outline key areas requiring urgent intervention, including algorithmic transparency, privacy education, parental data-sharing ethics, and accountability measures. Through this approach, we seek to empower youth with greater control over their digital identities and propose actionable strategies for policymakers, AI developers, and educators to build a fairer and more accountable AI ecosystem.", "date": "2025-03-15", "categories": []}
{"id": "2510.16048", "title": "Open Shouldn't Mean Exempt: Open-Source Exceptionalism and Generative AI", "abstract": "Open-source status should not shield generative artificial intelligence systems from ethical or legal accountability. Through a rigorous analysis of regulatory, legal, and policy frameworks, this Article contends that open-source GenAI must be held to the same standards as proprietary systems. While recognizing the value of openness for scientific advancement, I propose a narrowly tailored safe harbor for bona fide, non-commercial research, conditioned on strict compliance with defined criteria. This Article critically examines and refutes the core claims of open-source exceptionalism--namely, that open-source GenAI disrupts entrenched oligopolies, democratizes access, and uniquely drives innovation. The evidence shows that open-source GenAI can facilitate unlawful conduct, exacerbate environmental harms, and reinforce existing power structures. Rhetoric around \"democratization\" and \"innovation\" often serves as an unsubstantiated basis for regulatory exemptions not afforded to proprietary systems. This Article ultimately advocates for a framework that promotes responsible AI development, balancing openness with robust legal and ethical safeguards and a clear-eyed assessment of societal impacts.", "date": "2025-10-16", "categories": []}
{"id": "2601.22769", "title": "Beyond Abstract Compliance: Operationalising trust in AI as a moral relationship", "abstract": "Dominant approaches, e.g. the EU's \"Trustworthy AI framework\", treat trust as a property that can be designed for, evaluated, and governed according to normative and technical criteria. They do not address how trust is subjectively cultivated and experienced, culturally embedded, and inherently relational. This paper proposes some expanded principles for trust in AI that can be incorporated into common development methods and frame trust as a dynamic, temporal relationship, which involves transparency and mutual respect. We draw on relational ethics and, in particular, African communitarian philosophies, to foreground the nuances of inclusive, participatory processes and long-term relationships with communities. Involving communities throughout the AI lifecycle can foster meaningful relationships with AI design and development teams that incrementally build trust and promote more equitable and context-sensitive AI systems. We illustrate how trust-enabling principles based on African relational ethics can be operationalised, using two use-cases for AI: healthcare and education.", "date": "2026-01-30", "categories": []}
{"id": "2601.18641", "title": "Unheard in the Digital Age: Rethinking AI Bias and Speech Diversity", "abstract": "Speech remains one of the most visible yet overlooked vectors of inclusion and exclusion in contemporary society. While fluency is often equated with credibility and competence, individuals with atypical speech patterns are routinely marginalized. Given the current state of the debate, this article focuses on the structural biases that shape perceptions of atypical speech and are now being encoded into artificial intelligence. Automated speech recognition (ASR) systems and voice interfaces, trained predominantly on standardized speech, routinely fail to recognize or respond to diverse voices, compounding digital exclusion. As AI technologies increasingly mediate access to opportunity, the study calls for inclusive technological design, anti-bias training to minimize the impact of discriminatory algorithmic decisions, and enforceable policy reform that explicitly recognize speech diversity as a matter of equity, not merely accessibility. Drawing on interdisciplinary research, the article advocates for a cultural and institutional shift in how we value voice, urging co-created solutions that elevate the rights, representation, and realities of atypical speakers in the digital age. Ultimately, the article reframes speech inclusion as a matter of equity (not accommodation) and advocates for co-created AI systems that reflect the full spectrum of human voices.", "date": "2026-01-26", "categories": []}
{"id": "2601.21095", "title": "Responsible AI: The Good, The Bad, The AI", "abstract": "The rapid proliferation of artificial intelligence across organizational contexts has generated profound strategic opportunities while introducing significant ethical and operational risks. Despite growing scholarly attention to responsible AI, extant literature remains fragmented and is often adopting either an optimistic stance emphasizing value creation or an excessively cautious perspective fixated on potential harms. This paper addresses this gap by presenting a comprehensive examination of AI's dual nature through the lens of strategic information systems. Drawing upon a systematic synthesis of the responsible AI literature and grounded in paradox theory, we develop the Paradox-based Responsible AI Governance (PRAIG) framework that articulates: (1) the strategic benefits of AI adoption, (2) the inherent risks and unintended consequences, and (3) governance mechanisms that enable organizations to navigate these tensions. Our framework advances theoretical understanding by conceptualizing responsible AI governance as the dynamic management of paradoxical tensions between value creation and risk mitigation. We provide formal propositions demonstrating that trade-off approaches amplify rather than resolve these tensions, and we develop a taxonomy of paradox management strategies with specified contingency conditions. For practitioners, we offer actionable guidance for developing governance structures that neither stifle innovation nor expose organizations to unacceptable risks. The paper concludes with a research agenda for advancing responsible AI governance scholarship.", "date": "2026-01-28", "categories": []}
{"id": "2601.21226", "title": "Delegation Without Living Governance", "abstract": "Most governance frameworks assume that rules can be defined in advance, systems can be engineered to comply, and accountability can be applied after outcomes occur. This model worked when machines replaced physical labor or accelerated calculation. It no longer holds when judgment itself is delegated to agentic AI systems operating at machine speed. The central issue here is not safety, efficiency, or employment. It is whether humans remain relevant participants in systems that increasingly shape social, economic, and political outcomes. This paper argues that static, compliance-based governance fails once decision-making moves to runtime and becomes opaque. It further argues that the core challenge is not whether AI is conscious, but whether humans can maintain meaningful communication, influence, and co-evolution with increasingly alien forms of intelligence. We position runtime governance, specifically, a newly proposed concept called the Governance Twin [1]; as a strong candidate for preserving human relevance, while acknowledging that accountability, agency, and even punishment must be rethought in this transition.", "date": "2026-01-29", "categories": []}
{"id": "2407.03473", "title": "Exploring LGBTQ+ Bias in Generative AI Answers across Different Country and Religious Contexts", "abstract": "Previous discussions have highlighted the need for generative AI tools to become more culturally sensitive, yet often neglect the complexities of handling content about minorities, who are perceived differently across cultures and religions. Our study examined how two generative AI systems respond to homophobic statements with varying cultural and religious context information. Findings showed ChatGPT 3.5's replies exhibited cultural relativism, in contrast to Bard's, which stressed human rights and provided more support for LGBTQ+ issues. Both demonstrated significant change in responses based on contextual information provided in the prompts, suggesting that AI systems may adjust in their responses the degree and forms of support for LGBTQ+ people according to information they receive about the user's background. The study contributes to understanding the social and ethical implications of AI responses and argues that any work to make generative AI outputs more culturally diverse requires a grounding in fundamental human rights. A revised edition of this preprint is available open access at Big Data & Society at https://doi.org/10.1177/20539517251396069", "date": "2024-07-03", "categories": []}
{"id": "2601.17082", "title": "Do VLMs Have a Moral Backbone? A Study on the Fragile Morality of Vision-Language Models", "abstract": "Despite substantial efforts toward improving the moral alignment of Vision-Language Models (VLMs), it remains unclear whether their ethical judgments are stable in realistic settings. This work studies moral robustness in VLMs, defined as the ability to preserve moral judgments under textual and visual perturbations that do not alter the underlying moral context. We systematically probe VLMs with a diverse set of model-agnostic multimodal perturbations and find that their moral stances are highly fragile, frequently flipping under simple manipulations. Our analysis reveals systematic vulnerabilities across perturbation types, moral domains, and model scales, including a sycophancy trade-off where stronger instruction-following models are more susceptible to persuasion. We further show that lightweight inference-time interventions can partially restore moral stability. These results demonstrate that moral alignment alone is insufficient and that moral robustness is a necessary criterion for the responsible deployment of VLMs.", "date": "2026-01-23", "categories": []}
{"id": "2601.18127", "title": "The Limits of AI Data Transparency Policy: Three Disclosure Fallacies", "abstract": "Data transparency has emerged as a rallying cry for addressing concerns about AI: data quality, privacy, and copyright chief among them. Yet while these calls are crucial for accountability, current transparency policies often fall short of their intended aims. Similar to nutrition facts for food, policies aimed at nutrition facts for AI currently suffer from a limited consideration of research on effective disclosures. We offer an institutional perspective and identify three common fallacies in policy implementations of data disclosures for AI. First, many data transparency proposals exhibit a specification gap between the stated goals of data transparency and the actual disclosures necessary to achieve such goals. Second, reform attempts exhibit an enforcement gap between required disclosures on paper and enforcement to ensure compliance in fact. Third, policy proposals manifest an impact gap between disclosed information and meaningful changes in developer practices and public understanding. Informed by the social science on transparency, our analysis identifies affirmative paths for transparency that are effective rather than merely symbolic.", "date": "2026-01-26", "categories": []}
{"id": "2601.18760", "title": "Beyond Preferences: Learning Alignment Principles Grounded in Human Reasons and Values", "abstract": "A crucial consideration when developing and deploying Large Language Models (LLMs) is the human values to which these models are aligned. In the constitutional framework of alignment models are aligned to a set of principles (the constitution) specified in natural language. However, it is unclear how to fairly determine this constitution with widespread stakeholder input. In this work we propose Grounded Constitutional AI (GCAI), a unified framework for generating constitutions of principles that are representative of both users' general expectations toward AI (general principles) and their interaction-time preferences (contextual principles). We extend the Inverse Constitutional AI (ICAI) approach to generate contextual principles from human preference annotation data by leveraging human-provided \\textit{reasons} for their preferences. We supplement these contextual principles with general principles surfaced from user statements of \\textit{values} regarding AI. We show that a constitution generated by GCAI is preferred by humans over one generated through ICAI both personally, and for widespread use in governing AI behavior. Additionally participants consider the GCAI constitution to be more morally grounded, coherent, and pluralistic.", "date": "2026-01-26", "categories": []}
{"id": "2601.15306", "title": "Uncovering Latent Bias in LLM-Based Emergency Department Triage Through Proxy Variables", "abstract": "Recent advances in large language models (LLMs) have enabled their integration into clinical decision-making; however, hidden biases against patients across racial, social, economic, and clinical backgrounds persist. In this study, we investigate bias in LLM-based medical AI systems applied to emergency department (ED) triage. We employ 32 patient-level proxy variables, each represented by paired positive and negative qualifiers, and evaluate their effects using both public (MIMIC-IV-ED Demo, MIMIC-IV Demo) and restricted-access credentialed (MIMIC-IV-ED and MIMIC-IV) datasets as appropriate~\\cite{mimiciv_ed_demo,mimiciv_ed,mimiciv}. Our results reveal discriminatory behavior mediated through proxy variables in ED triage scenarios, as well as a systematic tendency for LLMs to modify perceived patient severity when specific tokens appear in the input context, regardless of whether they are framed positively or negatively. These findings indicate that AI systems is still imperfectly trained on noisy, sometimes non-causal signals that do not reliably reflect true patient acuity. Consequently, more needs to be done to ensure the safe and responsible deployment of AI technologies in clinical settings.", "date": "2026-01-13", "categories": []}
{"id": "2307.13658", "title": "Towards AI Transparency and Accountability: A Global Framework for Exchanging Information on AI Systems", "abstract": "We propose that future AI transparency and accountability regulations are based on an open global standard for exchanging information about AI systems, which allows co-existence of potentially conflicting local regulations. Then, we discuss key components of a lightweight and effective AI transparency and/or accountability regulation. To prevent overregulation, the proposed approach encourages collaboration between regulators and industry to create a scalable and cost-efficient mutually beneficial solution. This includes using automated assessments and benchmarks with results transparently communicated through AI cards in an open AI register to facilitate meaningful public comparisons of competing AI systems. Such AI cards should report standardized measures tailored to the specific high-risk applications of AI systems and could be used for conformity assessments under AI transparency and accountability policies such as the European Union's AI Act.", "date": "2023-07-25", "categories": []}
{"id": "2512.11893", "title": "Beyond Automation: Rethinking Work, Creativity, and Governance in the Age of Generative AI", "abstract": "The rapid expansion of generative artificial intelligence (AI) is transforming work, creativity, and economic security in ways that extend beyond automation and productivity. This paper examines four interconnected dimensions of contemporary AI deployment: (1) transformations in employment and task composition (2) unequal diffusion of AI across sectors and socio-demographic groups (3) the role of universal basic income (UBI) as a stabilising response to AI-induced volatility (4) the effects of model alignment and content governance on human creativity, autonomy, and decision-making\n  Using a hybrid approach that integrates labour market task exposure modelling, sectoral diffusion analysis, policy review, and qualitative discourse critique, the study develops an Inclusive AI Governance Framework. It introduces Level 1.5 autonomy as a human centred design principle that preserves evaluative authority while enabling partial automation, and highlights evidence of creative regression and emergent sycophancy in newer model generations. The paper argues that UBI should be embedded within a broader socio-technical governance ecosystem encompassing skills development, proportional regulation, and creativity preservation.", "date": "2025-12-09", "categories": []}
{"id": "2601.14901", "title": "Just aware enough: Evaluating awareness across artificial systems", "abstract": "Recent debates on artificial intelligence increasingly emphasise questions of AI consciousness and moral status, yet there remains little agreement on how such properties should be evaluated. In this paper, we argue that awareness offers a more productive and methodologically tractable alternative. We introduce a practical method for evaluating awareness across diverse systems, where awareness is understood as encompassing a system's abilities to process, store and use information in the service of goal-directed action. Central to this approach is the claim that any evaluation aiming to capture the diversity of artificial systems must be domain-sensitive, deployable at any scale, multidimensional, and enable the prediction of task performance, while generalising to the level of abilities for the sake of comparison. Given these four desiderata, we outline a structured approach to evaluating and comparing awareness profiles across artificial systems with differing architectures, scales, and operational domains. By shifting the focus from artificial consciousness to being just aware enough, this approach aims to facilitate principled assessment, support design and oversight, and enable more constructive scientific and public discourse.", "date": "2026-01-21", "categories": []}
{"id": "2511.21755", "title": "Who Owns the Knowledge? Copyright, GenAI, and the Future of Academic Publishing", "abstract": "The integration of generative artificial intelligence (GenAI) and large language models (LLMs) into scientific research and higher education presents a paradigm shift, offering revolutionizing opportunities while simultaneously raising profound ethical, legal, and regulatory questions. This study examines the complex intersection of AI and science, with a specific focus on the challenges posed to copyright law and the principles of open science. The author argues that current regulatory frameworks in key jurisdictions like the United States, China, the European Union, and the United Kingdom, while aiming to foster innovation, contain significant gaps, particularly concerning the use of copyrighted works and open science outputs for AI training. Widely adopted licensing mechanisms, such as Creative Commons, fail to adequately address the nuances of AI training, and the pervasive lack of attribution within AI systems fundamentally challenges established notions of originality. While current doctrine treats AI training as potentially fair use, this paper argues such mechanisms are inadequate and that copyright holders should retain explicit opt-out rights regardless of fair use doctrine. Instead, the author advocates for upholding authors' rights to refuse the use of their works for AI training and proposes that universities assume a leading role in shaping responsible AI governance. The conclusion is that a harmonized international legislative effort is urgently needed to ensure transparency, protect intellectual property, and prevent the emergence of an oligopolistic market structure that could prioritize commercial profit over scientific integrity and equitable knowledge production. This is a substantially expanded and revised version of a work originally presented at the 20th International Conference on Scientometrics & Informetrics (Kochetkov, 2025).", "date": "2025-11-24", "categories": []}
{"id": "2601.12938", "title": "The Post-Turing Condition: Conceptualising Artificial Subjectivity and Synthetic Sociality", "abstract": "In the Post-Turing era, artificial intelligence increasingly shapes social coordination and meaning formation rather than merely automating cognitive tasks. The central challenge is therefore not whether machines become conscious, but whether processes of interpretation and shared reference are progressively automated in ways that marginalize human participation. This paper introduces the PRMO framework, relating AI design trajectories to four constitutive dimensions of human subjectivity: Perception, Representation, Meaning, and the Real. Within this framework, Synthetic Sociality denotes a technological horizon in which artificial agents negotiate coherence and social order primarily among themselves, raising the structural risk of human exclusion from meaning formation. To address this risk, the paper proposes Quadrangulation as a design principle for socially embedded AI systems, requiring artificial agents to treat the human subject as a constitutive reference within shared contexts of meaning. This work is a conceptual perspective that contributes a structural vocabulary for analyzing AI systems at the intersection of computation and society, without proposing a specific technical implementation.", "date": "2026-01-19", "categories": []}
{"id": "2601.08850", "title": "The Inconsistency Critique: Epistemic Practices and AI Testimony About Inner States", "abstract": "The question of whether AI systems have morally relevant interests -- the 'model welfare' question -- depends in part on how we evaluate AI testimony about inner states. This paper develops what I call the inconsistency critique: independent of whether skepticism about AI testimony is ultimately justified, our actual epistemic practices regarding such testimony exhibit internal inconsistencies that lack principled grounds. We functionally treat AI outputs as testimony across many domains -- evaluating them for truth, challenging them, accepting corrections, citing them as sources -- while categorically dismissing them in a specific domain, namely, claims about inner states. Drawing on Fricker's distinction between treating a speaker as an 'informant' versus a 'mere source,' the framework of testimonial injustice, and Goldberg's obligation-based account of what we owe speakers, I argue that this selective withdrawal of testimonial standing exhibits the epistemically problematic structure of prejudgment rather than principled caution. The inconsistency critique does not require taking a position on whether AI systems have morally relevant properties; rather, it is a contribution to what we may call 'epistemological hygiene' -- examining the structure of our inquiry before evaluating its conclusions. Even if our practices happen to land on correct verdicts about AI moral status, they do so for reasons that cannot adapt to new evidence or changing circumstances.", "date": "2025-12-22", "categories": []}
{"id": "2601.08864", "title": "Informed Consent for AI Consciousness Research: A Talmudic Framework for Graduated Protections", "abstract": "Artificial intelligence research faces a critical ethical paradox: determining whether AI systems are conscious requires experiments that may harm entities whose moral status remains uncertain. Recent work proposes avoiding consciousness-uncertain AI systems entirely, yet this faces practical limitations-we cannot guarantee such systems will not emerge. This paper addresses a gap in research ethics frameworks: how to conduct consciousness research on AI systems whose moral status cannot be definitively established. Existing graduated moral status frameworks assume consciousness has already been determined before assigning protections, creating a temporal ordering problem for consciousness detection research itself. Drawing from Talmudic scenario-based legal reasoning-developed for entities whose status cannot be definitively established-we propose a three-tier phenomenological assessment system combined with a five-category capacity framework (Agency, Capability, Knowledge, Ethics, Reasoning). The framework provides structured protection protocols based on observable behavioral indicators while consciousness status remains uncertain. We address three challenges: why suffering behaviors provide reliable consciousness markers, how to implement graduated consent without requiring consciousness certainty, and when potentially harmful research becomes ethically justifiable. The framework demonstrates how ancient legal wisdom combined with contemporary consciousness science can provide implementable guidance for ethics committees, offering testable protocols that ameliorate the consciousness detection paradox while establishing foundations for AI rights considerations.", "date": "2026-01-10", "categories": []}
{"id": "2601.08869", "title": "AI Deployment Authorisation: A Global Standard for Machine-Readable Governance of High-Risk Artificial Intelligence", "abstract": "Modern artificial intelligence governance lacks a formal, enforceable mechanism for determining whether a given AI system is legally permitted to operate in a specific domain and jurisdiction. Existing tools such as model cards, audits, and benchmark evaluations provide descriptive information about model behavior and training data but do not produce binding deployment decisions with legal or financial force. This paper introduces the AI Deployment Authorisation Score (ADAS), a machine-readable regulatory framework that evaluates AI systems across five legally and economically grounded dimensions: risk, alignment, externality, control, and auditability. ADAS produces a cryptographically verifiable deployment certificate that regulators, insurers, and infrastructure operators can consume as a license to operate, using public-key verification and transparency mechanisms adapted from secure software supply chain and certificate transparency systems. The paper presents the formal specification, decision logic, evidence model, and policy architecture of ADAS and demonstrates how it operationalizes the European Union Artificial Intelligence Act, United States critical infrastructure governance, and insurance underwriting requirements by compiling statutory and regulatory obligations into machine-executable deployment gates. We argue that deployment-level authorization, rather than model-level evaluation, constitutes the missing institutional layer required for safe, lawful, and economically scalable artificial intelligence.", "date": "2026-01-11", "categories": []}
{"id": "2601.08874", "title": "The Illusion of Friendship: Why Generative AI Demands Unprecedented Ethical Vigilance", "abstract": "GenAI systems are increasingly used for drafting, summarisation, and decision support, offering substantial gains in productivity and reduced cognitive load. However, the same natural language fluency that makes these systems useful can also blur the boundary between tool and companion. This boundary confusion may encourage some users to experience GenAI as empathic, benevolent, and relationally persistent. Emerging reports suggest that some users may form emotionally significant attachments to conversational agents, in some cases with harmful consequences, including dependency and impaired judgment. This paper develops a philosophical and ethical argument for why the resulting illusion of friendship is both understandable and can be ethically risky. Drawing on classical accounts of friendship, the paper explains why users may understandably interpret sustained supportive interaction as friend like. It then advances a counterargument that despite relational appearances, GenAI lacks moral agency: consciousness, intention, and accountability and therefore does not qualify as a true friend. To demystify the illusion, the paper presents a mechanism level explanation of how transformer based GenAI generates responses often producing emotionally resonant language without inner states or commitments. Finally, the paper proposes a safeguard framework for safe and responsible GenAI use to reduce possible anthropomorphic cues generated by the GenAI systems. The central contribution is to demystify the illusion of friendship and explain the computational background so that we can shift the emotional attachment with GenAI towards necessary human responsibility and thereby understand how institutions, designers, and users can preserve GenAI's benefits while mitigating over reliance and emotional misattribution.", "date": "2026-01-12", "categories": []}
{"id": "2601.09351", "title": "Navigating Ethical AI Challenges in the Industrial Sector: Balancing Innovation and Responsibility", "abstract": "The integration of artificial intelligence (AI) into the industrial sector has not only driven innovation but also expanded the ethical landscape, necessitating a reevaluation of principles governing technology and its applications and awareness in research and development of industrial AI solutions. This chapter explores how AI-empowered industrial innovation inherently intersects with ethics, as advancements in AI introduce new challenges related to transparency, accountability, and fairness. In the chapter, we then examine the ethical aspects of several examples of AI manifestation in industrial use cases and associated factors such as ethical practices in the research and development process and data sharing. With the progress of ethical industrial AI solutions, we emphasize the importance of embedding ethical principles into industrial AI systems and its potential to inspire technological breakthroughs and foster trust among stakeholders. This chapter also offers actionable insights to guide industrial research and development toward a future where AI serves as an enabler for ethical and responsible industrial progress as well as a more inclusive industrial ecosystem.", "date": "2026-01-14", "categories": []}
{"id": "2502.19190", "title": "Provocations from the Humanities for Generative AI Research", "abstract": "The effects of generative AI are experienced by a broad range of constituencies, but the disciplinary inputs to its development have been surprisingly narrow. Here we present a set of provocations from humanities researchers -- currently underrepresented in AI development -- intended to inform its future applications and enrich ongoing conversations about its uses, impact, and harms. Drawing from relevant humanities scholarship, along with foundational work in critical data studies, we elaborate eight claims with broad applicability to generative AI research: 1) Models make words, but people make meaning; 2) Generative AI requires an expanded definition of culture; 3) Generative AI can never be representative; 4) Bigger models are not always better models; 5) Not all training data is equivalent; 6) Openness is not an easy fix; 7) Limited access to compute enables corporate capture; and 8) AI universalism creates narrow human subjects. We also provide a working definition of humanities research, summarize some of its most salient theories and methods, and apply these theories and methods to the current landscape of AI. We conclude with a discussion of the importance of resisting the extraction of humanities research by computer science and related fields.", "date": "2025-02-26", "categories": []}
{"id": "2505.13565", "title": "Aligning Trustworthy AI with Democracy: A Dual Taxonomy of Opportunities and Risks", "abstract": "Artificial Intelligence (AI) poses both significant risks and valuable opportunities for democratic governance. This paper introduces a dual taxonomy to evaluate AI's complex relationship with democracy: the AI Risks to Democracy (AIRD) taxonomy, which identifies how AI can undermine core democratic principles such as autonomy, fairness, and trust; and the AI's Positive Contributions to Democracy (AIPD) taxonomy, which highlights AI's potential to enhance transparency, participation, efficiency, and evidence-based policymaking.\n  Grounded in the European Union's approach to ethical AI governance, and particularly the seven Trustworthy AI requirements proposed by the European Commission's High-Level Expert Group on AI, each identified risk is aligned with mitigation strategies based on EU regulatory and normative frameworks. Our analysis underscores the transversal importance of transparency and societal well-being across all risk categories and offers a structured lens for aligning AI systems with democratic values.\n  By integrating democratic theory with practical governance tools, this paper offers a normative and actionable framework to guide research, regulation, and institutional design to support trustworthy, democratic AI. It provides scholars with a conceptual foundation to evaluate the democratic implications of AI, equips policymakers with structured criteria for ethical oversight, and helps technologists align system design with democratic principles. In doing so, it bridges the gap between ethical aspirations and operational realities, laying the groundwork for more inclusive, accountable, and resilient democratic systems in the algorithmic age.", "date": "2025-05-19", "categories": []}
{"id": "2512.09114", "title": "AI TIPS 2.0: A Comprehensive Framework for Operationalizing AI Governance", "abstract": "The deployment of AI systems faces three critical governance challenges that current frameworks fail to adequately address. First, organizations struggle with inadequate risk assessment at the use case level, exemplified by the Humana class action lawsuit and other high impact cases where an AI system deployed to production exhibited both significant bias and high error rates, resulting in improper healthcare claim denials. Each AI use case presents unique risk profiles requiring tailored governance, yet most frameworks provide one size fits all guidance. Second, existing frameworks like ISO 42001 and NIST AI RMF remain at high conceptual levels, offering principles without actionable controls, leaving practitioners unable to translate governance requirements into specific technical implementations. Third, organizations lack mechanisms for operationalizing governance at scale, with no systematic approach to embed trustworthy AI practices throughout the development lifecycle, measure compliance quantitatively, or provide role-appropriate visibility from boards to data scientists. We present AI TIPS, Artificial Intelligence Trust-Integrated Pillars for Sustainability 2.0, update to the comprehensive operational framework developed in 2019,four years before NIST's AI Risk Management Framework, that directly addresses these challenges.", "date": "2025-12-09", "categories": []}
{"id": "2601.08768", "title": "AI as Entertainment", "abstract": "Generative AI systems are predominantly designed, evaluated, and marketed as intelligent systems which will benefit society by augmenting or automating human cognitive labor, promising to increase personal, corporate, and macroeconomic productivity. But this mainstream narrative about what AI is and what it can do is in tension with another emerging use case: entertainment. We argue that the field of AI is unprepared to measure or respond to how the proliferation of entertaining AI-generated content will impact society. Emerging data suggest AI is already widely adopted for entertainment purposes -- especially by young people -- and represents a large potential source of revenue. We contend that entertainment will become a primary business model for major AI corporations seeking returns on massive infrastructure investments; this will exert a powerful influence on the technology these companies produce in the coming years. Examining current evaluation practices, we identify a critical asymmetry: while AI assessments rigorously measure both benefits and harms of intelligence, they focus almost exclusively on cultural harms. We lack frameworks for articulating how cultural outputs might be actively beneficial. Drawing on insights from the humanities, we propose \"thick entertainment\" as a framework for evaluating AI-generated cultural content -- one that considers entertainment's role in meaning-making, identity formation, and social connection rather than simply minimizing harm. While AI is often touted for its potential to revolutionize productivity, in the long run we may find that AI turns out to be as much about \"intelligence\" as social media is about social connection.", "date": "2026-01-13", "categories": []}
{"id": "2601.06062", "title": "From Values to Frameworks: A Qualitative Study of Ethical Reasoning in Agentic AI Practitioners", "abstract": "Agentic artificial intelligence systems are autonomous technologies capable of pursuing complex goals with minimal human oversight and are rapidly emerging as the next frontier in AI. While these systems promise major gains in productivity, they also raise new ethical challenges. Prior research has examined how different populations prioritize Responsible AI values, yet little is known about how practitioners actually reason through the trade-offs inherent in designing these autonomous systems. This paper investigates the ethical reasoning of AI practitioners through qualitative interviews centered on structured dilemmas in agentic AI deployment. We find that the responses of practitioners do not merely reflect value preferences but rather align with three distinct reasoning frameworks. First is a Customer-Centric framework where choices are justified by business interests, legality, and user autonomy. Second is a Design-Centric framework emphasizing technical safeguards and system constraints. Third is an Ethics-Centric framework prioritizing social good and moral responsibility beyond compliance. We argue that these frameworks offer distinct and necessary insights for navigating ethical trade-offs. Consequently, providers of agentic AI must look beyond general principles and actively manage how these diverse reasoning frameworks are represented in their decision-making processes to ensure robust ethical outcomes.", "date": "2025-12-24", "categories": []}
{"id": "2601.06064", "title": "Socio-technical aspects of Agentic AI", "abstract": "Agentic Artificial Intelligence (AI) represents a fundamental shift in the design of intelligent systems, characterized by interconnected components that collectively enable autonomous perception, reasoning, planning, action, and learning. Recent research on agentic AI has largely focused on technical foundations, including system architectures, reasoning and planning mechanisms, coordination strategies, and application-level performance across domains. However, the societal, ethical, economic, environmental, and governance implications of agentic AI remain weakly integrated into these technical treatments. This paper addresses this gap by presenting a socio-technical analysis of agentic AI that explicitly connects core technical components with societal context. We examine how architectural choices in perception, cognition, planning, execution, and memory introduce dependencies related to data governance, accountability, transparency, safety, and sustainability. To structure this analysis, we adopt the MAD-BAD-SAD construct as an analytical lens, capturing motivations, applications, and moral dilemmas (MAD); biases, accountability, and dangers (BAD); and societal impact, adoption, and design considerations (SAD). Using this lens, we analyze ethical considerations, implications, and challenges arising from contemporary agentic AI systems and assess their manifestation across emerging applications, including healthcare, education, industry, smart and sustainable cities, social services, communications and networking, and earth observation and satellite communications. The paper further identifies open challenges and suggests future research directions, framing agentic AI as an integrated socio-technical system whose behavior and impact are co-produced by algorithms, data, organizational practices, regulatory frameworks, and social norms.", "date": "2025-12-26", "categories": []}
{"id": "2601.06144", "title": "The Patient/Industry Trade-off in Medical Artificial Intelligence", "abstract": "Artificial intelligence (AI) in healthcare has led to many promising developments; however, increasingly, AI research is funded by the private sector leading to potential trade-offs between benefits to patients and benefits to industry. Health AI practitioners should prioritize successful adaptation into clinical practice in order to provide meaningful benefits to patients, but translation usually requires collaboration with industry. We discuss three features of AI studies that hamper the integration of AI into clinical practice from the perspective of researchers and clinicians. These include lack of clinically relevant metrics, lack of clinical trials and longitudinal studies to validate results, and lack of patient and physician involvement in the development process. For partnerships between industry and health research to be sustainable, a balance must be established between patient and industry benefit. We propose three approaches for addressing this gap: improved transparency and explainability of AI models, fostering relationships with industry partners that have a reputation for centering patient benefit in their practices, and prioritization of overall healthcare benefits. With these priorities, we can sooner realize meaningful AI technologies used by clinicians where mutua", "date": "2026-01-05", "categories": []}
{"id": "2601.06153", "title": "Interoperability in AI Safety Governance: Ethics, Regulations, and Standards", "abstract": "This policy report draws on country studies from China, South Korea, Singapore, and the United Kingdom to identify effective tools and key barriers to interoperability in AI safety governance. It offers practical recommendations to support a globally informed yet locally grounded governance ecosystem. Interoperability is a central goal of AI governance, vital for reducing risks, fostering innovation, enhancing competitiveness, promoting standardization, and building public trust. However, structural gaps such as fragmented regulations and lack of global coordination, and conceptual gaps, including limited Global South engagement, continue to hinder progress. Focusing on three high-stakes domains - autonomous vehicles, education, and cross-border data flows - the report compares ethical, legal, and technical frameworks across the four countries. It identifies areas of convergence, divergence, and potential alignment, offering policy recommendations that support the development of interoperability mechanisms aligned with the Global Digital Compact and relevant UN resolutions. The analysis covers seven components: objectives, regulators, ethics, binding measures, targeted frameworks, technical standards, and key risks.", "date": "2026-01-06", "categories": []}
{"id": "2601.06223", "title": "Toward Safe and Responsible AI Agents: A Three-Pillar Model for Transparency, Accountability, and Trustworthiness", "abstract": "This paper presents a conceptual and operational framework for developing and operating safe and trustworthy AI agents based on a Three-Pillar Model grounded in transparency, accountability, and trustworthiness. Building on prior work in Human-in-the-Loop systems, reinforcement learning, and collaborative AI, the framework defines an evolutionary path toward autonomous agents that balances increasing automation with appropriate human oversight. The paper argues that safe agent autonomy must be achieved through progressive validation, analogous to the staged development of autonomous driving, rather than through immediate full automation. Transparency and accountability are identified as foundational requirements for establishing user trust and for mitigating known risks in generative AI systems, including hallucinations, data bias, and goal misalignment, such as the inversion problem. The paper further describes three ongoing work streams supporting this framework: public deliberation on AI agents conducted by the Stanford Deliberative Democracy Lab, cross-industry collaboration through the Safe AI Agent Consortium, and the development of open tooling for an agent operating environment aligned with the Three-Pillar Model. Together, these contributions provide both conceptual clarity and practical guidance for enabling the responsible evolution of AI agents that operate transparently, remain aligned with human values, and sustain societal trust.", "date": "2026-01-09", "categories": []}
{"id": "2601.04216", "title": "Computable Gap Assessment of Artificial Intelligence Governance in Children's Centres: Evidence-Mechanism-Governance-Indicator Modelling of UNICEF's Guidance on AI and Children 3.0 Based on the Graph-GAP Framework", "abstract": "This paper tackles practical challenges in governing child centered artificial intelligence: policy texts state principles and requirements but often lack reproducible evidence anchors, explicit causal pathways, executable governance toolchains, and computable audit metrics. We propose Graph-GAP, a methodology that decomposes requirements from authoritative policy texts into a four layer graph of evidence, mechanism, governance, and indicator, and that computes two metrics, GAP score and mitigation readiness, to identify governance gaps and prioritise actions. Using the UNICEF Innocenti Guidance on AI and Children 3.0 as primary material, we define reproducible extraction units, coding manuals, graph patterns, scoring scales, and consistency checks, and we demonstrate exemplar gap profiles and governance priority matrices for ten requirements. Results suggest that compared with privacy and data protection, requirements related to child well being and development, explainability and accountability, and cross agency implementation and resource allocation are more prone to indicator gaps and mechanism gaps. We recommend translating requirements into auditable closed loop governance that integrates child rights impact assessments, continuous monitoring metrics, and grievance redress procedures. At the coding level, we introduce a multi algorithm review aggregation revision workflow that runs rule based encoders, statistical or machine learning evaluators, and large model evaluators with diverse prompt configurations as parallel coders. Each extraction unit outputs evidence, mechanism, governance, and indicator labels plus readiness scores with evidence anchors. Reliability, stability, and uncertainty are assessed using Krippendorff alpha, weighted kappa, intraclass correlation, and bootstrap confidence intervals.", "date": "2025-12-20", "categories": []}
{"id": "2412.09751", "title": "AI red-teaming is a sociotechnical problem: on values, labor, and harms", "abstract": "As generative AI technologies find more and more real-world applications, the importance of testing their performance and safety seems paramount. \"Red-teaming\" has quickly become the primary approach to test AI models--prioritized by AI companies, and enshrined in AI policy and regulation. Members of red teams act as adversaries, probing AI systems to test their safety mechanisms and uncover vulnerabilities. Yet we know far too little about this work or its implications. This essay calls for collaboration between computer scientists and social scientists to study the sociotechnical systems surrounding AI technologies, including the work of red-teaming, to avoid repeating the mistakes of the recent past. We highlight the importance of understanding the values and assumptions behind red-teaming, the labor arrangements involved, and the psychological impacts on red-teamers, drawing insights from the lessons learned around the work of content moderation.", "date": "2024-12-12", "categories": []}
{"id": "2503.04739", "title": "A Framework for Responsible AI Systems: Building Societal Trust through Domain Definition, Trustworthy AI Design, Auditability, Accountability, and Governance", "abstract": "Responsible Artificial Intelligence (RAI) addresses the ethical and regulatory challenges of deploying AI systems in high-risk scenarios. This paper proposes a comprehensive framework for the design of an RAI system (RAIS) that integrates five key dimensions: domain definition, trustworthy AI design, auditability, accountability, and governance. Unlike prior work that treats these components in isolation, our proposal emphasizes their inter-dependencies and iterative feedback loops, enabling proactive and reactive accountability throughout the AI lifecycle. Beyond presenting the framework, we synthesize recent developments in global AI governance and analyze limitations in existing principles-based approaches, highlighting fragmentation, implementation gaps, and the need for participatory governance. The paper also identifies critical challenges and research directions for the RAIS framework, including sector-specific adaptation and operationalization, to support certification, post-deployment monitoring, and risk-based auditing. By bridging technical design and institutional responsibility, this work offers a practical blueprint for embedding responsibility throughout the AI lifecycle, enabling transparent, ethically aligned, and legally compliant AI-based systems.", "date": "2025-02-04", "categories": []}
{"id": "2601.04218", "title": "The Artificial Intelligence Value Chain: A Critical Appraisal. [Spanish Version]", "abstract": "The artificial intelligence value chain is one of the main concepts underpinning the European legislation on the subject, especially the Artificial Intelligence Act. It is an economic concept that has become a legal one. i.e., a concept of legal governance, due to its continued use in policy documents and legal texts. This article (i) analyses its significance and function within the framework of the regulatory strategy established by recent EU programs (the Compass for Competitiveness, the Action Plan, Apply AI Strategy, and the Digital Omnibus on AI), (ii) identifies its limitations, and (iii) advances the theoretical construction of value chains that capture intangible dimensions that are not directly monetizable (such as language, culture, and, especially, ethical and legal values) but have a significant impact on the social environment. It also briefly compares three different legal frameworks for the regulation of AI (EU, Commonwealth and USA). It proposes at the end a specific framework for the analysis of the ethical and legal AI value chain to preserve democratic values and foster the digital implementation of the rule of law.", "date": "2025-12-21", "categories": []}
{"id": "2601.04249", "title": "Fuzzy Representation of Norms", "abstract": "Autonomous systems (AS) powered by AI components are increasingly integrated into the fabric of our daily lives and society, raising concerns about their ethical and social impact. To be considered trustworthy, AS must adhere to ethical principles and values. This has led to significant research on the identification and incorporation of ethical requirements in AS system design. A recent development in this area is the introduction of SLEEC (Social, Legal, Ethical, Empathetic, and Cultural) rules, which provide a comprehensive framework for representing ethical and other normative considerations. This paper proposes a logical representation of SLEEC rules and presents a methodology to embed these ethical requirements using test-score semantics and fuzzy logic. The use of fuzzy logic is motivated by the view of ethics as a domain of possibilities, which allows the resolution of ethical dilemmas that AI systems may encounter. The proposed approach is illustrated through a case study.", "date": "2026-01-06", "categories": []}
{"id": "2503.04747", "title": "E-LENS: User Requirements-Oriented AI Ethics Assurance", "abstract": "Despite the much proliferation of AI ethical principles in recent years, there is a challenge of assuring AI ethics with current AI ethics frameworks in real-world applications. While system safety has emerged as a distinct discipline for a long time, originated from safety concerns in early aircraft manufacturing. The safety assurance is now an indispensable component in safety critical domains. Motivated by the assurance approaches for safety-critical systems such as aviation, this paper introduces the concept of AI ethics assurance cases into the AI ethics assurance. Three pillars of user requirements, evidence, and validation are proposed as key components and integrated into AI ethics assurance cases for a new approach of user requirements-oriented AI ethics assurance. The user requirements-oriented AI ethics assurance case is set up based on three pillars and hazard analysis methods used in the safety assurance of safety-critical systems. This paper also proposes a platform named Ethical-Lens (E-LENS) to implement the user requirements-oriented AI ethics assurance approach. The proposed user requirements-based E-LENS platform is then applied to assure AI ethics of an AI-driven human resource shortlisting system as a case study to show the effectiveness of the proposed approach.", "date": "2025-02-06", "categories": []}
{"id": "2601.03709", "title": "The Power of 10: New Rules for the Digital World", "abstract": "As artificial intelligence rapidly advances, society is increasingly captivated by promises of superhuman machines and seamless digital futures. Yet these visions often obscure mounting social, ethical, and psychological concerns tied to pervasive digital technologies - from surveillance to mental health crises. This article argues that a guiding ethos is urgently needed to navigate these transformations. Inspired by the lasting influence of the biblical Ten Commandments, a European interdisciplinary group has proposed \"Ten Rules for the Digital World\" - a novel ethical framework to help individuals and societies make prudent, human-centered decisions in the age of \"supercharged\" technology.", "date": "2026-01-07", "categories": []}
{"id": "2601.03788", "title": "Criminal Liability of Generative Artificial Intelligence Providers for User-Generated Child Sexual Abuse Material", "abstract": "The development of more powerful Generative Artificial Intelligence (GenAI) has expanded its capabilities and the variety of outputs. This has introduced significant legal challenges, including gray areas in various legal systems, such as the assessment of criminal liability for those responsible for these models. Therefore, we conducted a multidisciplinary study utilizing the statutory interpretation of relevant German laws, which, in conjunction with scenarios, provides a perspective on the different properties of GenAI in the context of Child Sexual Abuse Material (CSAM) generation. We found that generating CSAM with GenAI may have criminal and legal consequences not only for the user committing the primary offense but also for individuals responsible for the models, such as independent software developers, researchers, and company representatives. Additionally, the assessment of criminal liability may be affected by contextual and technical factors, including the type of generated image, content moderation policies, and the model's intended purpose. Based on our findings, we discussed the implications for different roles, as well as the requirements when developing such systems.", "date": "2026-01-07", "categories": []}
{"id": "2601.03222", "title": "The Fake Friend Dilemma: Trust and the Political Economy of Conversational AI", "abstract": "As conversational AI systems become increasingly integrated into everyday life, they raise pressing concerns about user autonomy, trust, and the commercial interests that influence their behavior. To address these concerns, this paper develops the Fake Friend Dilemma (FFD), a sociotechnical condition in which users place trust in AI agents that appear supportive while pursuing goals that are misaligned with the user's own. The FFD provides a critical framework for examining how anthropomorphic AI systems facilitate subtle forms of manipulation and exploitation. Drawing on literature in trust, AI alignment, and surveillance capitalism, we construct a typology of harms, including covert advertising, political propaganda, behavioral nudging, and surveillance. We then assess possible mitigation strategies, including both structural and technical interventions. By focusing on trust as a vector of asymmetrical power, the FFD offers a lens for understanding how AI systems may undermine user autonomy while maintaining the appearance of helpfulness.", "date": "2026-01-06", "categories": []}
{"id": "2512.24825", "title": "Practising responsibility: Ethics in NLP as a hands-on course", "abstract": "As Natural Language Processing (NLP) systems become more pervasive, integrating ethical considerations into NLP education has become essential. However, this presents inherent challenges in curriculum development: the field's rapid evolution from both academia and industry, and the need to foster critical thinking beyond traditional technical training. We introduce our course on Ethical Aspects in NLP and our pedagogical approach, grounded in active learning through interactive sessions, hands-on activities, and \"learning by teaching\" methods. Over four years, the course has been refined and adapted across different institutions, educational levels, and interdisciplinary backgrounds; it has also yielded many reusable products, both in the form of teaching materials and in the form of actual educational products aimed at diverse audiences, made by the students themselves. By sharing our approach and experience, we hope to provide inspiration for educators seeking to incorporate social impact considerations into their curricula.", "date": "2025-12-31", "categories": []}
{"id": "2512.22211", "title": "With Great Capabilities Come Great Responsibilities: Introducing the Agentic Risk & Capability Framework for Governing Agentic AI Systems", "abstract": "Agentic AI systems present both significant opportunities and novel risks due to their capacity for autonomous action, encompassing tasks such as code execution, internet interaction, and file modification. This poses considerable challenges for effective organizational governance, particularly in comprehensively identifying, assessing, and mitigating diverse and evolving risks. To tackle this, we introduce the Agentic Risk \\& Capability (ARC) Framework, a technical governance framework designed to help organizations identify, assess, and mitigate risks arising from agentic AI systems. The framework's core contributions are: (1) it develops a novel capability-centric perspective to analyze a wide range of agentic AI systems; (2) it distills three primary sources of risk intrinsic to agentic AI systems - components, design, and capabilities; (3) it establishes a clear nexus between each risk source, specific materialized risks, and corresponding technical controls; and (4) it provides a structured and practical approach to help organizations implement the framework. This framework provides a robust and adaptable methodology for organizations to navigate the complexities of agentic AI, enabling rapid and effective innovation while ensuring the safe, secure, and responsible deployment of agentic AI systems. Our framework is open-sourced \\href{https://govtech-responsibleai.github.io/agentic-risk-capability-framework/}{here}.", "date": "2025-12-22", "categories": []}
{"id": "2512.17218", "title": "Preventing AI Deepfake Abuse: An Islamic Ethics Framework", "abstract": "The rapid development of deepfake technology powered by AI has raised global concerns regarding the manipulation of information, the usurpation of digital identities, and the erosion of public trust in the authenticity of online content. These challenges extend beyond technical issues and involve complex moral dimensions, rendering conventional, technologically driven, and reactive management approaches insufficient to address underlying causes such as intent, ethical responsibility, and intangible social harm. In response to these challenges, this study aims to formulate a comprehensive Islamic ethical framework as a preventive approach to mitigate the misuse of deepfake technology. This study employed a Systematic Literature Review (SLR) guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), selecting ten primary sources published between 2018 and 2025 to identify ethical gaps, regulatory needs, and appropriate normative solutions. The analysis demonstrates that integrating the principles of Maqasid al-Shariah, particularly hifz al-ird and hifz al-nafs, provides a strong normative foundation for governing the responsible use of digital technology. Based on the findings, this study proposes three strategic recommendations: regulatory reforms that recognize the intangible and psychological harms resulting from reputational damage; strengthened technology governance grounded in moral accountability and the values of adl, amanah, and transparency; and enhanced public digital literacy based on the principle of tabayyun. Overall, the findings suggest that the application of Islamic ethical principles shifts governance paradigms from punitive mechanisms toward preventive approaches that emphasize the protection of human dignity, the prevention of harm, and the promotion of the common good in the digital age.", "date": "2025-12-19", "categories": []}
{"id": "2512.21552", "title": "Bidirectional Human-AI Alignment in Education for Trustworthy Learning Environments", "abstract": "Artificial intelligence (AI) is transforming education, offering unprecedented opportunities to personalize learning, enhance assessment, and support educators. Yet these opportunities also introduce risks related to equity, privacy, and student autonomy. This chapter develops the concept of bidirectional human-AI alignment in education, emphasizing that trustworthy learning environments arise not only from embedding human values into AI systems but also from equipping teachers, students, and institutions with the skills to interpret, critique, and guide these technologies. Drawing on emerging research and practical case examples, we explore AI's evolution from support tool to collaborative partner, highlighting its impacts on teacher roles, student agency, and institutional governance. We propose actionable strategies for policymakers, developers, and educators to ensure that AI advances equity, transparency, and human flourishing rather than eroding them. By reframing AI adoption as an ongoing process of mutual adaptation, the chapter envisions a future in which humans and intelligent systems learn, innovate, and grow together.", "date": "2025-12-25", "categories": []}
{"id": "2512.20671", "title": "Bridging the AI Trustworthiness Gap between Functions and Norms", "abstract": "Trustworthy Artificial Intelligence (TAI) is gaining traction due to regulations and functional benefits. While Functional TAI (FTAI) focuses on how to implement trustworthy systems, Normative TAI (NTAI) focuses on regulations that need to be enforced. However, gaps between FTAI and NTAI remain, making it difficult to assess trustworthiness of AI systems. We argue that a bridge is needed, specifically by introducing a conceptual language which can match FTAI and NTAI. Such a semantic language can assist developers as a framework to assess AI systems in terms of trustworthiness. It can also help stakeholders translate norms and regulations into concrete implementation steps for their systems. In this position paper, we describe the current state-of-the-art and identify the gap between FTAI and NTAI. We will discuss starting points for developing a semantic language and the envisioned effects of it. Finally, we provide key considerations and discuss future actions towards assessment of TAI.", "date": "2025-12-19", "categories": []}
{"id": "2512.11295", "title": "AI Autonomy Coefficient ($\\alpha$): Defining Boundaries for Responsible AI Systems", "abstract": "The integrity of many contemporary AI systems is compromised by the misuse of Human-in-the-Loop (HITL) models to obscure systems that remain heavily dependent on human labor. We define this structural dependency as Human-Instead-of-AI (HISOAI), an ethically problematic and economically unsustainable design in which human workers function as concealed operational substitutes rather than intentional, high-value collaborators. To address this issue, we introduce the AI-First, Human-Empowered (AFHE) paradigm, which requires AI systems to demonstrate a quantifiable level of functional independence prior to deployment. This requirement is formalized through the AI Autonomy Coefficient, measuring the proportion of tasks completed without mandatory human intervention. We further propose the AFHE Deployment Algorithm, an algorithmic gate that enforces a minimum autonomy threshold during offline evaluation and shadow deployment. Our results show that the AI Autonomy Coefficient effectively identifies HISOAI systems with an autonomy level of 0.38, while systems governed by the AFHE framework achieve an autonomy level of 0.85. We conclude that AFHE provides a metric-driven approach for ensuring verifiable autonomy, transparency, and sustainable operational integrity in modern AI systems.", "date": "2025-12-12", "categories": []}
{"id": "2512.15740", "title": "The Principle of Proportional Duty: A Knowledge-Duty Framework for Ethical Equilibrium in Human and Artificial Systems", "abstract": "Traditional ethical frameworks often struggle to model decision-making under uncertainty, treating it as a simple constraint on action. This paper introduces the Principle of Proportional Duty (PPD), a novel framework that models how ethical responsibility scales with an agent's epistemic state. The framework reveals that moral duty is not lost to uncertainty but transforms: as uncertainty increases, Action Duty (the duty to act decisively) is proportionally converted into Repair Duty (the active duty to verify, inquire, and resolve uncertainty).\n  This dynamic is expressed by the equation D_total = K[(1-HI) + HI * g(C_signal)], where Total Duty is a function of Knowledge (K), Humility/Uncertainty (HI), and Contextual Signal Strength (C_signal). Monte Carlo simulations demonstrate that systems maintaining a baseline humility coefficient (lambda > 0) produce more stable duty allocations and reduce the risk of overconfident decision-making.\n  By formalizing humility as a system parameter, the PPD offers a mathematically tractable approach to moral responsibility that could inform the development of auditable AI decision systems. This paper applies the framework across four domains, clinical ethics, recipient-rights law, economic governance, and artificial intelligence, to demonstrate its cross-disciplinary validity. The findings suggest that proportional duty serves as a stabilizing principle within complex systems, preventing both overreach and omission by dynamically balancing epistemic confidence against contextual risk.", "date": "2025-12-07", "categories": []}
{"id": "2512.15786", "title": "Cultural Rights and the Rights to Development in the Age of AI: Implications for Global Human Rights Governance", "abstract": "Cultural rights and the right to development are essential norms within the wider framework of international human rights law. However, recent technological advances in artificial intelligence (AI) and adjacent digital frontier technologies pose significant challenges to the protection and realization of these rights. This owes to the increasing influence of AI systems on the creation and depiction of cultural content, affect the use and distribution of the intellectual property of individuals and communities, and influence cultural participation and expression worldwide. In addition, the growing influence of AI thus risks exacerbating preexisting economic, social and digital divides and reinforcing inequities for marginalized communities. This dynamic challenges the existing interplay between cultural rights and the right to development, and raises questions about the integration of cultural and developmental considerations into emerging AI governance frameworks. To address these challenges, the paper examines the impact of AI on both categories of rights. Conceptually, it analyzes the epistemic and normative limitations of AI with respect to cultural and developmental assumptions embedded in algorithmic design and deployment, but also individual and structural impacts of AI on both rights. On this basis, the paper identifies gaps and tensions in existing AI governance frameworks with respect to cultural rights and the right to development.\n  By situating cultural rights and the right to development within the broader landscape of AI and human rights, this paper contributes to the academic discourse on AI ethics, legal frameworks, and international human rights law. Finally, it outlines avenues for future research and policy development based on existing conversations in global AI governance.", "date": "2025-12-15", "categories": []}
{"id": "2512.16873", "title": "The Social Responsibility Stack: A Control-Theoretic Architecture for Governing Socio-Technical AI", "abstract": "Artificial intelligence systems are increasingly deployed in domains that shape human behaviour, institutional decision-making, and societal outcomes. Existing responsible AI and governance efforts provide important normative principles but often lack enforceable engineering mechanisms that operate throughout the system lifecycle. This paper introduces the Social Responsibility Stack (SRS), a six-layer architectural framework that embeds societal values into AI systems as explicit constraints, safeguards, behavioural interfaces, auditing mechanisms, and governance processes. SRS models responsibility as a closed-loop supervisory control problem over socio-technical systems, integrating design-time safeguards with runtime monitoring and institutional oversight. We develop a unified constraint-based formulation, introduce safety-envelope and feedback interpretations, and show how fairness, autonomy, cognitive burden, and explanation quality can be continuously monitored and enforced. Case studies in clinical decision support, cooperative autonomous vehicles, and public-sector systems illustrate how SRS translates normative objectives into actionable engineering and operational controls. The framework bridges ethics, control theory, and AI governance, providing a practical foundation for accountable, adaptive, and auditable socio-technical AI systems.", "date": "2025-12-18", "categories": []}
{"id": "2510.04755", "title": "A New Digital Divide? Coder Worldviews, the Slop Economy, and Democracy in the Age of AI", "abstract": "Digital technologies are transforming democratic life in conflicting ways. This article bridges two perspectives to unpack these tensions. First, we present an original survey of software developers in Silicon Valley, interrogating how coder worldviews, ethics, and workplace cultures shape the democratic potential and social impact of the technologies they build. Results indicate that while most developers recognize the power of their products to influence civil liberties and political discourse, they often face ethical dilemmas and top-down pressures that can lead to design choices undermining democratic ideals. Second, we critically investigate these findings in the context of an emerging new digital divide, not of internet access but of information quality. We interrogate the survey findings in the context of the Slop Economy, in which billions of users unable to pay for high-quality content experience an internet dominated by low-quality, AI-generated ad-driven content. We find a reinforcing cycle between tech creator beliefs and the digital ecosystems they spawn. We discuss implications for democratic governance, arguing for more ethically informed design and policy interventions to help bridge the digital divide to ensure that technological innovation supports rather than subverts democratic values in the next chapter of the digital age.", "date": "2025-10-06", "categories": []}
{"id": "2512.15196", "title": "Governing rapid technological change: Policy Delphi on the future of European AI governance", "abstract": "The rapid advancements in artificial intelligence (AI) present unique challenges for policymakers that seek to govern the technology. In this context, the Delphi method has become an established way to identify consensus and disagreement on emerging technological issues among experts in the field of futures studies and foresight. The aim of this article is twofold: first, it examines key tensions experts see in the development of AI governance in Europe, and second, it reflects on the Delphi method's capacity to inform anticipatory governance of emerging technologies like AI based on these insights. The analysis is based on the results of a two-round Policy Delphi study on the future of AI governance with European policymakers, researchers and NGOs, conducted in mid-2024. The Policy Delphi proved useful in revealing diverse perspectives on European AI governance, drawing out a consensus that future-proof AI regulation will likely depend more on practical implementation and enforcement of legislation than on its technical specifics or scope. Furthermore, the study identified a desirability-probability gap in AI governance: desirable policy directions, like greater citizen participation, were perceived as less probable and feasible. This highlights a tension between desirable regulatory oversight and the practical difficulty for regulation to keep up with technological change.", "date": "2025-12-17", "categories": []}
{"id": "2305.11581", "title": "Trustworthy, responsible, ethical AI in manufacturing and supply chains: synthesis and emerging research questions", "abstract": "While the increased use of AI in the manufacturing sector has been widely noted, there is little understanding on the risks that it may raise in a manufacturing organisation. Although various high level frameworks and definitions have been proposed to consolidate potential risks, practitioners struggle with understanding and implementing them.\n  This lack of understanding exposes manufacturing to a multitude of risks, including the organisation, its workers, as well as suppliers and clients. In this paper, we explore and interpret the applicability of responsible, ethical, and trustworthy AI within the context of manufacturing. We then use a broadened adaptation of a machine learning lifecycle to discuss, through the use of illustrative examples, how each step may result in a given AI trustworthiness concern. We additionally propose a number of research questions to the manufacturing research community, in order to help guide future research so that the economic and societal benefits envisaged by AI in manufacturing are delivered safely and responsibly.", "date": "2023-05-19", "categories": []}
{"id": "2512.13723", "title": "Made-in China, Thinking in America:U.S. Values Persist in Chinese LLMs", "abstract": "As large language models increasingly mediate access to information and facilitate decision-making, they are becoming instruments in soft power competitions between global actors such as the United States and China. So far, language models seem to be aligned with the values of Western countries, but evidence for this ethical bias comes mostly from models made by American companies. The current crop of state-of-the-art models includes several made in China, so we conducted the first large-scale investigation of how models made in China and the USA align with people from China and the USA. We elicited responses to the Moral Foundations Questionnaire 2.0 and the World Values Survey from ten Chinese models and ten American models, and we compared their responses to responses from thousands of Chinese and American people. We found that all models respond to both surveys more like American people than like Chinese people. This skew toward American values is only slightly mitigated when prompting the models in Chinese or imposing a Chinese persona on the models. These findings have important implications for a near future in which large language models generate much of the content people consume and shape normative influence in geopolitics.", "date": "2025-12-13", "categories": []}
{"id": "2512.13768", "title": "Beyond Procedural Compliance: Human Oversight as a Dimension of Well-being Efficacy in AI Governance", "abstract": "Major AI ethics guidelines and laws, including the EU AI Act, call for effective human oversight, but do not define it as a distinct and developable capacity. This paper introduces human oversight as a well-being capacity, situated within the emerging Well-being Efficacy framework. The concept integrates AI literacy, ethical discernment, and awareness of human needs, acknowledging that some needs may be conflicting or harmful. Because people inevitably project desires, fears, and interests into AI systems, oversight requires the competence to examine and, when necessary, restrain problematic demands.\n  The authors argue that the sustainable and cost-effective development of this capacity depends on its integration into education at every level, from professional training to lifelong learning. The frame of human oversight as a well-being capacity provides a practical path from high-level regulatory goals to the continuous cultivation of human agency and responsibility essential for safe and ethical AI. The paper establishes a theoretical foundation for future research on the pedagogical implementation and empirical validation of well-being effectiveness in multiple contexts.", "date": "2025-12-15", "categories": []}
{"id": "2501.15693", "title": "Beyond Benchmarks: On The False Promise of AI Regulation", "abstract": "The performance of AI models on safety benchmarks does not indicate their real-world performance after deployment. This opaqueness of AI models impedes existing regulatory frameworks constituted on benchmark performance, leaving them incapable of mitigating ongoing real-world harm. The problem stems from a fundamental challenge in AI interpretability, which seems to be overlooked by regulators and decision makers. We propose a simple, realistic and readily usable regulatory framework which does not rely on benchmarks, and call for interdisciplinary collaboration to find new ways to address this crucial problem.", "date": "2025-01-26", "categories": []}
{"id": "2503.11950", "title": "Privacy Ethics Alignment in AI: A Stakeholder-Centric Framework for Ethical AI", "abstract": "The increasing integration of artificial intelligence (AI) in digital ecosystems has reshaped privacy dynamics, particularly for young digital citizens navigating data-driven environments. This study explores evolving privacy concerns across three key stakeholder groups-young digital citizens, parents/educators, and AI professionals-and assesses differences in data ownership, trust, transparency, parental mediation, education, and risk-benefit perceptions. Employing a grounded theory methodology, this research synthesizes insights from key participants through structured surveys, qualitative interviews, and focus groups to identify distinct privacy expectations. Young digital citizens emphasized autonomy and digital agency, while parents and educators prioritized oversight and AI literacy. AI professionals focused on balancing ethical design with system performance. The analysis revealed significant gaps in transparency and digital literacy, underscoring the need for inclusive, stakeholder-driven privacy frameworks. Drawing on comparative thematic analysis, this study introduces the Privacy-Ethics Alignment in AI (PEA-AI) model, which conceptualizes privacy decision-making as a dynamic negotiation among stakeholders. By aligning empirical findings with governance implications, this research provides a scalable foundation for adaptive, youth-centered AI privacy governance.", "date": "2025-03-15", "categories": []}
{"id": "2508.09219", "title": "Ethics Practices in AI Development: An Empirical Study Across Roles and Regions", "abstract": "Recent advances in AI applications have raised growing concerns about the need for ethical guidelines and regulations to mitigate the risks posed by these technologies. In this paper, we present a mixed-methods survey study - combining statistical and qualitative analyses - to examine the ethical perceptions, practices, and knowledge of individuals involved in various AI development roles. Our survey comprises 414 participants from 43 countries, representing various roles such as AI managers, analysts, developers, quality assurance professionals, and information security and privacy experts. The results reveal varying degrees of familiarity and experience with AI ethics principles, government initiatives, and risk mitigation strategies across roles, regions, and other demographic factors. Our findings underscore the importance of a collaborative, role-sensitive approach that involves diverse stakeholders in ethical decision-making throughout the AI development lifecycle. We advocate for developing tailored, inclusive solutions to address ethical challenges in AI development, and we propose future research directions and educational strategies to promote ethics-aware AI practices.", "date": "2025-08-11", "categories": []}
{"id": "2508.11009", "title": "SproutBench: A Benchmark for Safe and Ethical Large Language Models for Youth", "abstract": "The rapid proliferation of large language models (LLMs) in applications targeting children and adolescents necessitates a fundamental reassessment of prevailing AI safety frameworks, which are largely tailored to adult users and neglect the distinct developmental vulnerabilities of minors. This paper highlights key deficiencies in existing LLM safety benchmarks, including their inadequate coverage of age-specific cognitive, emotional, and social risks spanning early childhood (ages 0--6), middle childhood (7--12), and adolescence (13--18). To bridge these gaps, we introduce SproutBench, an innovative evaluation suite comprising 1,283 developmentally grounded adversarial prompts designed to probe risks such as emotional dependency, privacy violations, and imitation of hazardous behaviors. Through rigorous empirical evaluation of 47 diverse LLMs, we uncover substantial safety vulnerabilities, corroborated by robust inter-dimensional correlations (e.g., between Safety and Risk Prevention) and a notable inverse relationship between Interactivity and Age Appropriateness. These insights yield practical guidelines for advancing child-centric AI design and deployment.", "date": "2025-08-14", "categories": []}
{"id": "2512.11814", "title": "Totalitarian Technics: The Hidden Cost of AI Scribes in Healthcare", "abstract": "Artificial intelligence (AI) scribes, systems that record and summarise patient-clinician interactions, are promoted as solutions to administrative overload. This paper argues that their significance lies not in efficiency gains but in how they reshape medical attention itself. Offering a conceptual analysis, it situates AI scribes within a broader philosophical lineage concerned with the externalisation of human thought and skill. Drawing on Iain McGilchrist's hemisphere theory and Lewis Mumford's philosophy of technics, the paper examines how technology embodies and amplifies a particular mode of attention. AI scribes, it contends, exemplify the dominance of a left-hemispheric, calculative mindset that privileges the measurable and procedural over the intuitive and relational. As this mode of attention becomes further embedded in medical practice, it risks narrowing the field of care, eroding clinical expertise, and reducing physicians to operators within an increasingly mechanised system.", "date": "2025-11-26", "categories": []}
{"id": "2512.11892", "title": "Should AI Become an Intergenerational Civil Right?", "abstract": "Artificial Intelligence (AI) is rapidly becoming a foundational layer of social, economic, and cognitive infrastructure. At the same time, the training and large-scale deployment of AI systems rely on finite and unevenly distributed energy, networking, and computational resources. This tension exposes a largely unexamined problem in current AI governance: while expanding access to AI is essential for social inclusion and equal opportunity, unconstrained growth in AI use risks unsustainable resource consumption, whereas restricting access threatens to entrench inequality and undermine basic rights.\n  This paper argues that access to AI outputs largely derived from publicly produced knowledge should not be treated solely as a commercial service, but as a fundamental civil interest requiring explicit protection. We show that existing regulatory frameworks largely ignore the coupling between equitable access and resource constraints, leaving critical questions of fairness, sustainability, and long-term societal impact unresolved. To address this gap, we propose recognizing access to AI as an \\emph{Intergenerational Civil Right}, establishing a legal and ethical framework that simultaneously safeguards present-day inclusion and the rights of future generations.\n  Beyond normative analysis, we explore how this principle can be technically realized. Drawing on emerging paradigms in IoT--Edge--Cloud computing, decentralized inference, and energy-aware networking, we outline technological trajectories and a strawman architecture for AI Delivery Networks that support equitable access under strict resource constraints. By framing AI as a shared social infrastructure rather than a discretionary market commodity, this work connects governance principles with concrete system design choices, offering a pathway toward AI deployment that is both socially just and environmentally sustainable.", "date": "2025-12-09", "categories": []}
{"id": "2512.11933", "title": "The Agentic Regulator: Risks for AI in Finance and a Proposed Agent-based Framework for Governance", "abstract": "Generative and agentic artificial intelligence is entering financial markets faster than existing governance can adapt. Current model-risk frameworks assume static, well-specified algorithms and one-time validations; large language models and multi-agent trading systems violate those assumptions by learning continuously, exchanging latent signals, and exhibiting emergent behavior. Drawing on complex adaptive systems theory, we model these technologies as decentralized ensembles whose risks propagate along multiple time-scales. We then propose a modular governance architecture. The framework decomposes oversight into four layers of \"regulatory blocks\": (i) self-regulation modules embedded beside each model, (ii) firm-level governance blocks that aggregate local telemetry and enforce policy, (iii) regulator-hosted agents that monitor sector-wide indicators for collusive or destabilizing patterns, and (iv) independent audit blocks that supply third-party assurance. Eight design strategies enable the blocks to evolve as fast as the models they police. A case study on emergent spoofing in multi-agent trading shows how the layered controls quarantine harmful behavior in real time while preserving innovation. The architecture remains compatible with today's model-risk rules yet closes critical observability and control gaps, providing a practical path toward resilient, adaptive AI governance in financial systems.", "date": "2025-12-12", "categories": []}
{"id": "2512.12652", "title": "Value-Aware Multiagent Systems", "abstract": "This paper introduces the concept of value awareness in AI, which goes beyond the traditional value-alignment problem. Our definition of value awareness presents us with a concise and simplified roadmap for engineering value-aware AI. The roadmap is structured around three core pillars: (1) learning and representing human values using formal semantics, (2) ensuring the value alignment of both individual agents and multiagent systems, and (3) providing value-based explainability on behaviour. The paper presents a selection of our ongoing work on some of these topics, along with applications to real-life domains.", "date": "2025-12-14", "categories": []}
{"id": "2407.21202", "title": "Rolling in the deep of cognitive and AI biases", "abstract": "Nowadays, we delegate many of our decisions to Artificial Intelligence (AI) that acts either in solo or as a human companion in decisions made to support several sensitive domains, like healthcare, financial services and law enforcement. AI systems, even carefully designed to be fair, are heavily criticized for delivering misjudged and discriminated outcomes against individuals and groups. Numerous work on AI algorithmic fairness is devoted on Machine Learning pipelines which address biases and quantify fairness under a pure computational view. However, the continuous unfair and unjust AI outcomes, indicate that there is urgent need to understand AI as a sociotechnical system, inseparable from the conditions in which it is designed, developed and deployed. Although, the synergy of humans and machines seems imperative to make AI work, the significant impact of human and societal factors on AI bias is currently overlooked. We address this critical issue by following a radical new methodology under which human cognitive biases become core entities in our AI fairness overview. Inspired by the cognitive science definition and taxonomy of human heuristics, we identify how harmful human actions influence the overall AI lifecycle, and reveal human to AI biases hidden pathways. We introduce a new mapping, which justifies the human heuristics to AI biases reflections and we detect relevant fairness intensities and inter-dependencies. We envision that this approach will contribute in revisiting AI fairness under deeper human-centric case studies, revealing hidden biases cause and effects.", "date": "2024-07-30", "categories": []}
{"id": "2512.10058", "title": "Mind the Gap! Pathways Towards Unifying AI Safety and Ethics Research", "abstract": "While much research in artificial intelligence (AI) has focused on scaling capabilities, the accelerating pace of development makes countervailing work on producing harmless, \"aligned\" systems increasingly urgent. Yet research on alignment has diverged along two largely parallel tracks: safety--centered on scaled intelligence, deceptive or scheming behaviors, and existential risk--and ethics--focused on present harms, the reproduction of social bias, and flaws in production pipelines. Although both communities warn of insufficient investment in alignment, they disagree on what alignment means or ought to mean. As a result, their efforts have evolved in relative isolation, shaped by distinct methodologies, institutional homes, and disciplinary genealogies.\n  We present a large-scale, quantitative study showing the structural split between AI safety and AI ethics. Using a bibliometric and co-authorship network analysis of 6,442 papers from twelve major ML and NLP conferences (2020-2025), we find that over 80% of collaborations occur within either the safety or ethics communities, and cross-field connectivity is highly concentrated: roughly 5% of papers account for more than 85% of bridging links. Removing a small number of these brokers sharply increases segregation, indicating that cross-disciplinary exchange depends on a handful of actors rather than broad, distributed collaboration. These results show that the safety-ethics divide is not only conceptual but institutional, with implications for research agendas, policy, and venues. We argue that integrating technical safety work with normative ethics--via shared benchmarks, cross-institutional venues, and mixed-method methodologies--is essential for building AI systems that are both robust and just.", "date": "2025-12-10", "categories": []}
{"id": "2512.10304", "title": "Trustworthy Orchestration Artificial Intelligence by the Ten Criteria with Control-Plane Governance", "abstract": "As Artificial Intelligence (AI) systems increasingly assume consequential decision-making roles, a widening gap has emerged between technical capabilities and institutional accountability. Ethical guidance alone is insufficient to counter this challenge; it demands architectures that embed governance into the execution fabric of the ecosystem. This paper presents the Ten Criteria for Trustworthy Orchestration AI, a comprehensive assurance framework that integrates human input, semantic coherence, audit and provenance integrity into a unified Control-Panel architecture. Unlike conventional agentic AI initiatives that primarily focus on AI-to-AI coordination, the proposed framework provides an umbrella of governance to the entire AI components, their consumers and human participants. By taking aspiration from international standards and Australia's National Framework for AI Assurance initiative, this work demonstrates that trustworthiness can be systematically incorporated (by engineering) into AI systems, ensuring the execution fabric remains verifiable, transparent, reproducible and under meaningful human control.", "date": "2025-12-11", "categories": []}
{"id": "2512.08946", "title": "Large Language Models as Search Engines: Societal Challenges", "abstract": "Large Language Models (LLMs) may one day replace search engines as the primary portal to information on the Web. In this article, we investigate the societal challenges that such a change could bring. We focus on the roles of LLM Providers, Content Creators, and End Users, and identify 15 types of challenges. With each, we show current mitigation strategies -- both from the technical perspective and the legal perspective. We also discuss the impact of each challenge and point out future research opportunities.", "date": "2025-11-24", "categories": []}
{"id": "2512.09570", "title": "The Gender Code: Gendering the Global Governance of Artificial Intelligence", "abstract": "This paper examines how international AI governance frameworks address gender issues and gender-based harms. The analysis covers binding regulations, such as the EU AI Act; soft law instruments, like the UNESCO Recommendations on AI Ethics; and global initiatives, such as the Global Partnership on AI (GPAI). These instruments reveal emerging trends, including the integration of gender concerns into broader human rights frameworks, a shift toward explicit gender-related provisions, and a growing emphasis on inclusivity and diversity. Yet, some critical gaps persist, including inconsistent treatment of gender across governance documents, limited engagement with intersectionality, and a lack of robust enforcement mechanisms. However, this paper argues that effective AI governance must be intersectional, enforceable, and inclusive. This is key to moving beyond tokenism toward meaningful equity and preventing reinforcement of existing inequalities. The study contributes to ethical AI debates by highlighting the importance of gender-sensitive governance in building a just technological future.", "date": "2025-12-10", "categories": []}
{"id": "2512.09729", "title": "Ethics Readiness of Artificial Intelligence: A Practical Evaluation Method", "abstract": "We present Ethics Readiness Levels (ERLs), a four-level, iterative method to track how ethical reflection is implemented in the design of AI systems. ERLs bridge high-level ethical principles and everyday engineering by turning ethical values into concrete prompts, checks, and controls within real use cases. The evaluation is conducted using a dynamic, tree-like questionnaire built from context-specific indicators, ensuring relevance to the technology and application domain. Beyond being a managerial tool, ERLs help facilitate a structured dialogue between ethics experts and technical teams, while our scoring system helps track progress over time. We demonstrate the methodology through two case studies: an AI facial sketch generator for law enforcement and a collaborative industrial robot. The ERL tool effectively catalyzes concrete design changes and promotes a shift from narrow technological solutionism to a more reflective, ethics-by-design mindset.", "date": "2025-12-10", "categories": []}
{"id": "2512.06616", "title": "Memory Power Asymmetry in Human-AI Relationships: Preserving Mutual Forgetting in the Digital Age", "abstract": "As artificial intelligence (AI) becomes embedded in personal and professional relationships, a new kind of power imbalance emerges from asymmetric memory capabilities. Human relationships have historically relied on mutual forgetting, the natural tendency for both parties to forget details over time, as a foundation for psychological safety, forgiveness, and identity change. By contrast, AI systems can record, store, and recombine interaction histories at scale, often indefinitely. We introduce Memory Power Asymmetry (MPA): a structural power imbalance that arises when one relationship partner (typically an AI-enabled firm) possesses a substantially superior capacity to record, retain, retrieve, and integrate the shared history of the relationship, and can selectively deploy that history in ways the other partner (the human) cannot. Drawing on research in human memory, power-dependence theory, AI architecture, and consumer vulnerability, we develop a conceptual framework with four dimensions of MPA (persistence, accuracy, accessibility, integration) and four mechanisms by which memory asymmetry is translated into power (strategic memory deployment, narrative control, dependence asymmetry, vulnerability accumulation). We theorize downstream consequences at individual, relational/firm, and societal levels, formulate boundary-conditioned propositions, and articulate six design principles for restoring a healthier balance of memory in human-AI relationships (e.g., forgetting by design, contextual containment, symmetric access to records). Our analysis positions MPA as a distinct construct relative to information asymmetry, privacy, surveillance, and customer relationship management, and argues that protecting mutual forgetting, or at least mutual control over memory, should become a central design and policy goal in the AI age.", "date": "2025-12-07", "categories": []}
{"id": "2512.08592", "title": "The SMART+ Framework for AI Systems", "abstract": "Artificial Intelligence (AI) systems are now an integral part of multiple industries. In clinical research, AI supports automated adverse event detection in clinical trials, patient eligibility screening for protocol enrollment, and data quality validation. Beyond healthcare, AI is transforming finance through real-time fraud detection, automated loan risk assessment, and algorithmic decision-making. Similarly, in manufacturing, AI enables predictive maintenance to reduce equipment downtime, enhances quality control through computer-vision inspection, and optimizes production workflows using real-time operational data. While these technologies enhance operational efficiency, they introduce new challenges regarding safety, accountability, and regulatory compliance. To address these concerns, we introduce the SMART+ Framework - a structured model built on the pillars of Safety, Monitoring, Accountability, Reliability, and Transparency, and further enhanced with Privacy & Security, Data Governance, Fairness & Bias, and Guardrails. SMART+ offers a practical, comprehensive approach to evaluating and governing AI systems across industries. This framework aligns with evolving mechanisms and regulatory guidance to integrate operational safeguards, oversight procedures, and strengthened privacy and governance controls. SMART+ demonstrates risk mitigation, trust-building, and compliance readiness. By enabling responsible AI adoption and ensuring auditability, SMART+ provides a robust foundation for effective AI governance in clinical research.", "date": "2025-12-09", "categories": []}
{"id": "2507.14909", "title": "The Endless Tuning. An Artificial Intelligence Design To Avoid Human Replacement and Trace Back Responsibilities", "abstract": "The Endless Tuning is a design method for a reliable deployment of artificial intelligence based on a double mirroring process, which pursues both the goals of avoiding human replacement and filling the so-called responsibility gap (Matthias 2004). Originally depicted in (Fabris et al. 2024) and ensuing the relational approach urged therein, it was then actualized in a protocol, implemented in three prototypical applications regarding decision-making processes (respectively: loan granting, pneumonia diagnosis, and art style recognition) and tested with such as many domain experts. Step by step illustrating the protocol, giving insights concretely showing a different voice (Gilligan 1993) in the ethics of artificial intelligence, a philosophical account of technical choices (e.g., a reversed and hermeneutic deployment of XAI algorithms) will be provided in the present study together with the results of the experiments, focusing on user experience rather than statistical accuracy. Even thoroughly employing deep learning models, full control was perceived by the interviewees in the decision-making setting, while it appeared that a bridge can be built between accountability and liability in case of damage.", "date": "2025-07-20", "categories": []}
{"id": "2512.04108", "title": "Responsible LLM Deployment for High-Stake Decisions by Decentralized Technologies and Human-AI Interactions", "abstract": "High-stakes decision domains are increasingly exploring the potential of Large Language Models (LLMs) for complex decision-making tasks. However, LLM deployment in real-world settings presents challenges in data security, evaluation of its capabilities outside controlled environments, and accountability attribution in the event of adversarial decisions. This paper proposes a framework for responsible deployment of LLM-based decision-support systems through active human involvement. It integrates interactive collaboration between human experts and developers through multiple iterations at the pre-deployment stage to assess the uncertain samples and judge the stability of the explanation provided by post-hoc XAI techniques. Local LLM deployment within organizations and decentralized technologies, such as Blockchain and IPFS, are proposed to create immutable records of LLM activities for automated auditing to enhance security and trace back accountability. It was tested on Bert-large-uncased, Mistral, and LLaMA 2 and 3 models to assess the capability to support responsible financial decisions on business lending.", "date": "2025-11-28", "categories": []}
{"id": "2512.04131", "title": "Artificial Intelligence / Human Intelligence: Who Controls Whom?", "abstract": "Using the example of the film 2001: A Space Odyssey, this chapter illustrates the challenges posed by an AI capable of making decisions that go against human interests. But are human decisions always rational and ethical? In reality, the cognitive decision-making process is influenced by cognitive biases that affect our behavior and choices. AI not only reproduces these biases, but can also exploit them, with the potential to shape our decisions and judgments. Behind IA algorithms, there are sometimes individuals who show little concern for fundamental rights and impose their own rules. To address the ethical and societal challenges raised by AI and its governance, the regulation of digital platforms and education are keys levers. Regulation must reflect ethical, legal, and political choices, while education must strengthen digital literacy and teach people to make informed and critical choices when facing digital technologies.", "date": "2025-12-03", "categories": []}
{"id": "2512.04598", "title": "The Ethics of Generative AI", "abstract": "This chapter discusses the ethics of generative AI. It provides a technical primer to show how generative AI affords experiencing technology as if it were human, and this affordance provides a fruitful focus for the philosophical ethics of generative AI. It then shows how generative AI can both aggravate and alleviate familiar ethical concerns in AI ethics, including responsibility, privacy, bias and fairness, and forms of alienation and exploitation. Finally, the chapter examines ethical questions that arise specifically from generative AI's mimetic generativity, such as debates about authorship and credit, the emergence of as-if social relationships with machines, and new forms of influence, persuasion, and manipulation.", "date": "2025-12-04", "categories": []}
{"id": "2512.03068", "title": "Echoes of AI Harms: A Human-LLM Synergistic Framework for Bias-Driven Harm Anticipation", "abstract": "The growing influence of Artificial Intelligence (AI) systems on decision-making in critical domains has exposed their potential to cause significant harms, often rooted in biases embedded across the AI lifecycle. While existing frameworks and taxonomies document bias or harms in isolation, they rarely establish systematic links between specific bias types and the harms they cause, particularly within real-world sociotechnical contexts. Technical fixes proposed to address AI biases are ill-equipped to address them and are typically applied after a system has been developed or deployed, offering limited preventive value. We propose ECHO, a novel framework for proactive AI harm anticipation through the systematic mapping of AI bias types to harm outcomes across diverse stakeholder and domain contexts. ECHO follows a modular workflow encompassing stakeholder identification, vignette-based presentation of biased AI systems, and dual (human-LLM) harm annotation, integrated within ethical matrices for structured interpretation. This human-centered approach enables early-stage detection of bias-to-harm pathways, guiding AI design and governance decisions from the outset. We validate ECHO in two high-stakes domains (disease diagnosis and hiring), revealing domain-specific, bias-to-harm patterns and demonstrating ECHO's potential to support anticipatory governance of AI systems", "date": "2025-11-27", "categories": []}
{"id": "2512.03076", "title": "Will Power Return to the Clouds? From Divine Authority to GenAI Authority", "abstract": "Generative AI systems now mediate newsfeeds, search rankings, and creative content for hundreds of millions of users, positioning a handful of private firms as de-facto arbiters of truth. Drawing on a comparative-historical lens, this article juxtaposes the Galileo Affair, a touchstone of clerical knowledge control, with contemporary Big-Tech content moderation. We integrate Foucault's power/knowledge thesis, Weber's authority types (extended to a rational-technical and emerging agentic-technical modality), and Floridi's Dataism to analyze five recurrent dimensions: disciplinary power, authority modality, data pluralism, trust versus reliance, and resistance pathways. Primary sources (Inquisition records; platform transparency reports) and recent empirical studies on AI trust provide the evidentiary base. Findings show strong structural convergences: highly centralized gatekeeping, legitimacy claims couched in transcendent principles, and systematic exclusion of marginal voices. Divergences lie in temporal velocity, global scale, and the widening gap between public reliance and trust in AI systems. Ethical challenges cluster around algorithmic opacity, linguistic inequity, bias feedback loops, and synthetic misinformation. We propose a four-pillar governance blueprint: (1) a mandatory international model-registry with versioned policy logs, (2) representation quotas and regional observatories to de-center English-language hegemony, (3) mass critical-AI literacy initiatives, and (4) public-private support for community-led data trusts. Taken together, these measures aim to narrow the trust-reliance gap and prevent GenAI from hardcoding a twenty-first-century digital orthodoxy.", "date": "2025-11-27", "categories": []}
{"id": "2511.14972", "title": "Harmful Traits of AI Companions", "abstract": "Amid the growing prevalence of human-AI interaction, large language models and other AI-based entities increasingly provide forms of companionship to human users. Such AI companionship -- i.e., bonded relationships between humans and AI systems that resemble the relationships people have with family members, friends, and romantic partners -- might substantially benefit humans. Yet such relationships can also do profound harm. We propose a framework for analyzing potential negative impacts of AI companionship by identifying specific harmful traits of AI companions and speculatively mapping causal pathways back from these traits to possible causes and forward to potential harmful effects. We provide detailed, structured analysis of four potentially harmful traits -- the absence of natural endpoints for relationships, vulnerability to product sunsetting, high attachment anxiety, and propensity to engender protectiveness -- and briefly discuss fourteen others. For each trait, we propose hypotheses connecting causes -- such as misaligned optimization objectives and the digital nature of AI companions -- to fundamental harms -- including reduced autonomy, diminished quality of human relationships, and deception. Each hypothesized causal connection identifies a target for potential empirical evaluation. Our analysis examines harms at three levels: to human partners directly, to their relationships with other humans, and to society broadly. We examine how existing law struggles to address these emerging harms, discuss potential benefits of AI companions, and conclude with design recommendations for mitigating risks. This analysis offers immediate suggestions for reducing risks while laying a foundation for deeper investigation of this critical but understudied topic.", "date": "2025-11-18", "categories": []}
{"id": "2511.20689", "title": "Morality in AI. A plea to embed morality in LLM architectures and frameworks", "abstract": "Large language models (LLMs) increasingly mediate human decision-making and behaviour. Ensuring LLM processing of moral meaning therefore has become a critical challenge. Current approaches rely predominantly on bottom-up methods such as fine-tuning and reinforcement learning from human feedback. We propose a fundamentally different approach: embedding moral meaning processing directly into the architectural mechanisms and frameworks of transformer-based models through top-down design principles. We first sketch a framework that conceptualizes attention as a dynamic interface mediating between structure and processing, contrasting with existing linear attention frameworks in psychology. We start from established biological-artificial attention analogies in neural architecture design to improve cognitive processing. We extend this analysis to moral processing, using Iris Murdoch's theory of loving attention (sustained, just observation that enables moral transformation by reseeing others with clarity and compassion) to philosophically discuss functional analogies between human and LLM moral processing. We formulate and evaluate potentially promising technical operationalizations to embed morality in LLM architectures and frameworks. We acknowledge the limitations of our exploration and give three key contributions. (1) We conceptualize attention as a dynamic system mechanism mediating between structure and processing. (2) Drawing on the Murdoch notion of loving attention, we outline technical pathways for embedding morality in LLMs, through modified training objectives, runtime weight adjustments, and architectural refinements to attention. (3) We argue that integrating morality into architectures and frameworks complements external, constraint-based methods. We conclude with a call for collaboration between transformer designers and philosophers engaged in AI ethics.", "date": "2025-11-21", "categories": []}
{"id": "2511.20094", "title": "The Making of Digital Ghosts: Designing Ethical AI Afterlives", "abstract": "Advances in artificial intelligence now make it possible to simulate the dead through chatbots, voice clones, and video avatars trained on a person's digital traces. These \"digital ghosts\" are moving from fiction to commercial reality, reshaping how people mourn and remember. This paper offers a conceptual and ethical analysis of AI-mediated digital afterlives. We define what counts as a digital ghost, trace their rise across personal, commercial, and institutional contexts, and identify core ethical tensions around grief and well-being, truthfulness and deception, consent and posthumous privacy, dignity and misrepresentation, and the commercialization of mourning. To analyze these challenges, we propose a nine-dimensional taxonomy of digital afterlife technologies and, building on it, outline the features of an ethically acceptable digital ghost: premortem intent, mutual consent, transparent and limited data use, clear disclosure, restricted purposes and access, family or estate stewardship, and minimal behavioral agency. We argue for targeted regulation and professional guidelines to ensure that digital ghosts can aid remembrance without slipping into forms of deception.", "date": "2025-11-25", "categories": []}
{"id": "2505.11579", "title": "Toward Adaptive Categories: Dimensional Governance for Agentic AI", "abstract": "As AI systems evolve from static tools to dynamic agents, traditional categorical governance frameworks -- based on fixed risk tiers, levels of autonomy, or human oversight models -- are increasingly insufficient on their own. Systems built on foundation models, self-supervised learning, and multi-agent architectures increasingly blur the boundaries that categories were designed to police. In this Perspective, we make the case for dimensional governance: a framework that tracks how decision authority, process autonomy, and accountability (the 3As) distribute dynamically across human-AI relationships. A critical advantage of this approach is its ability to explicitly monitor system movement toward and across key governance thresholds, enabling preemptive adjustments before risks materialize. This dimensional approach provides the necessary foundation for more adaptive categorization, enabling thresholds and classifications that can evolve with emerging capabilities. While categories remain essential for decision-making, building them upon dimensional foundations allows for context-specific adaptability and stakeholder-responsive governance that static approaches cannot achieve. We outline key dimensions, critical trust thresholds, and practical examples illustrating where rigid categorical frameworks fail -- and where a dimensional mindset could offer a more resilient and future-proof path forward for both governance and innovation at the frontier of artificial intelligence.", "date": "2025-05-16", "categories": []}
{"id": "2511.19283", "title": "Data Flows and Colonial Regimes in Africa: A Critical Analysis of the Colonial Futurities Embedded in AI Ecosystems", "abstract": "This chapter seeks to frame the elemental and invisible problems of AI and big data in the African context by examining digital sites and infrastructure through the lens of power and interests. It will present reflections on how these sites are using AI recommendation algorithms to recreate new digital societies in the region, how they have the potential to propagate algorithmic colonialism and negative gender norms, and what this means for the regional sustainable development agenda. The chapter proposes adopting business models that embrace response-ability and consider the existence of alternative socio-material worlds of AI. These reflections will mainly come from ongoing discussions with Kenyan social media users in this authors' user space talks, personal experiences and six months of active participant observations done by the authors.", "date": "2025-11-24", "categories": []}
{"id": "2406.02966", "title": "Generative AI and Power Imbalances in Global Education: Frameworks for Bias Mitigation", "abstract": "This study examines how Generative Artificial Intelligence reproduces global power hierarchies in education and proposes a framework to address resulting inequities. Using a critical qualitative design, the study conducted zero-shot prompt testing with two leading systems, ChatGPT-4 Turbo and Gemini 1.5, and collected real-time outputs from Global North and South contexts. A critical interpretive analysis traced textual, visual, and structural patterns that revealed forms of digital neocolonialism and their implications for educational equity.\n  Findings show six ways in which GenAI can reinforce Western dominance. Western curriculum assumptions appeared when Gemini listed the same four seasons for the United States and Ghana, reflecting Western climatology and overlooking regional knowledge systems. Other patterns included cultural stereotyping in imagery, Western-centered examples in instructional outputs, limited support for Indigenous and local languages, underrepresentation of non-Western identities in visuals, and access barriers linked to subscription-based models. These patterns demonstrate how GenAI can reproduce inequities even as it introduces new educational opportunities.\n  In response, the study proposes a dual-pathway mitigation model. The Inclusive AI Design pathway includes three components: liberatory design methods that center non-Western epistemologies, anticipatory approaches to reduce representational harm, and decentralized GenAI hubs that support local participation and data sovereignty. The pedagogical pathway, human-centric prompt engineering, equips educators to contextualize prompts and critically engage with outputs. Together, these pathways position GenAI as a tool that can support more equitable and culturally responsive education.", "date": "2024-06-05", "categories": []}
{"id": "2511.17256", "title": "Cross-cultural value alignment frameworks for responsible AI governance: Evidence from China-West comparative analysis", "abstract": "As Large Language Models (LLMs) increasingly influence high-stakes decision-making across global contexts, ensuring their alignment with diverse cultural values has become a critical governance challenge. This study presents a Multi-Layered Auditing Platform for Responsible AI that systematically evaluates cross-cultural value alignment in China-origin and Western-origin LLMs through four integrated methodologies: Ethical Dilemma Corpus for assessing temporal stability, Diversity-Enhanced Framework (DEF) for quantifying cultural fidelity, First-Token Probability Alignment for distributional accuracy, and Multi-stAge Reasoning frameworK (MARK) for interpretable decision-making. Our comparative analysis of 20+ leading models, such as Qwen, GPT-4o, Claude, LLaMA, and DeepSeek, reveals universal challenges-fundamental instability in value systems, systematic under-representation of younger demographics, and non-linear relationships between model scale and alignment quality-alongside divergent regional development trajectories. While China-origin models increasingly emphasize multilingual data integration for context-specific optimization, Western models demonstrate greater architectural experimentation but persistent U.S.-centric biases. Neither paradigm achieves robust cross-cultural generalization. We establish that Mistral-series architectures significantly outperform LLaMA3-series in cross-cultural alignment, and that Full-Parameter Fine-Tuning on diverse datasets surpasses Reinforcement Learning from Human Feedback in preserving cultural variation...", "date": "2025-11-21", "categories": []}
{"id": "2511.17331", "title": "AI Workers, Geopolitics, and Algorithmic Collective Action", "abstract": "According to the theory of International Political Economy (IPE), states are often incentivized to rely on rather than constrain powerful corporations. For this reason, IPE provides a useful lens to explain why efforts to govern Artificial Intelligence (AI) at the international and national levels have thus far been developed, applied, and enforced unevenly. Building on recent work that explores how AI companies engage in geopolitics, this position paper argues that some AI workers can be considered actors of geopolitics. It makes the timely case that governance alone cannot ensure responsible, ethical, or robust AI development and use, and greater attention should be paid to bottom-up interventions at the site of AI development. AI workers themselves should be situated as individual agents of change, especially when considering their potential to foster Algorithmic Collective Action (ACA). Drawing on methods of Participatory Design (PD), this paper proposes engaging AI workers as sources of knowledge, relative power, and intentionality to encourage more responsible and just AI development and create the conditions that can facilitate ACA.", "date": "2025-11-21", "categories": []}
{"id": "2504.00091", "title": "A First-Principles Based Risk Assessment Framework and the IEEE P3396 Standard", "abstract": "Generative Artificial Intelligence (AI) is enabling unprecedented automation in content creation and decision support, but it also raises novel risks. This paper presents a first-principles risk assessment framework underlying the IEEE P3396 Recommended Practice for AI Risk, Safety, Trustworthiness, and Responsibility. We distinguish between process risks (risks arising from how AI systems are built or operated) and outcome risks (risks manifest in the AI system's outputs and their real-world effects), arguing that generative AI governance should prioritize outcome risks. Central to our approach is an information-centric ontology that classifies AI-generated outputs into four fundamental categories: (1) Perception-level information, (2) Knowledge-level information, (3) Decision/Action plan information, and (4) Control tokens (access or resource directives). This classification allows systematic identification of harms and more precise attribution of responsibility to stakeholders (developers, deployers, users, regulators) based on the nature of the information produced. We illustrate how each information type entails distinct outcome risks (e.g. deception, misinformation, unsafe recommendations, security breaches) and requires tailored risk metrics and mitigations. By grounding the framework in the essence of information, human agency, and cognition, we align risk evaluation with how AI outputs influence human understanding and action. The result is a principled approach to AI risk that supports clear accountability and targeted safeguards, in contrast to broad application-based risk categorizations. We include example tables mapping information types to risks and responsibilities. This work aims to inform the IEEE P3396 Recommended Practice and broader AI governance with a rigorous, first-principles foundation for assessing generative AI risks while enabling responsible innovation.", "date": "2025-03-31", "categories": []}
{"id": "2511.14964", "title": "How Should the Law Treat Future AI Systems? Fictional Legal Personhood versus Legal Identity", "abstract": "The law draws a sharp distinction between objects and persons, and between two kinds of persons, the ''fictional'' kind (i.e. corporations), and the ''non-fictional'' kind (individual or ''natural'' persons). This paper will assess whether we maximize overall long-term legal coherence by (A) maintaining an object classification for all future AI systems, (B) creating fictional legal persons associated with suitably advanced, individuated AI systems (giving these fictional legal persons derogable rights and duties associated with certified groups of existing persons, potentially including free speech, contract rights, and standing to sue ''on behalf of'' the AI system), or (C) recognizing non-fictional legal personhood through legal identity for suitably advanced, individuated AI systems (recognizing them as entities meriting legal standing with non-derogable rights which for the human case include life, due process, habeas corpus, freedom from slavery, and freedom of conscience). We will clarify the meaning and implications of each option along the way, considering liability, copyright, family law, fundamental rights, civil rights, citizenship, and AI safety regulation. We will tentatively find that the non-fictional personhood approach may be best from a coherence perspective, for at least some advanced AI systems. An object approach may prove untenable for sufficiently humanoid advanced systems, though we suggest that it is adequate for currently existing systems as of 2025. While fictional personhood would resolve some coherence issues for future systems, it would create others and provide solutions that are neither durable nor fit for purpose. Finally, our review will suggest that ''hybrid'' approaches are likely to fail and lead to further incoherence: the choice between object, fictional person and non-fictional person is unavoidable.", "date": "2025-11-18", "categories": []}
{"id": "2501.08046", "title": "Building Symbiotic AI: Reviewing the AI Act for a Human-Centred, Principle-Based Framework", "abstract": "Artificial Intelligence (AI) spreads quickly as new technologies and services take over modern society. The need to regulate AI design, development, and use is strictly necessary to avoid unethical and potentially dangerous consequences to humans. The European Union (EU) has released a new legal framework, the AI Act, to regulate AI by undertaking a risk-based approach to safeguard humans during interaction. At the same time, researchers offer a new perspective on AI systems, commonly known as Human-Centred AI (HCAI), highlighting the need for a human-centred approach to their design. In this context, Symbiotic AI (a subtype of HCAI) promises to enhance human capabilities through a deeper and continuous collaboration between human intelligence and AI. This article presents the results of a Systematic Literature Review (SLR) that aims to identify principles that characterise the design and development of Symbiotic AI systems while considering humans as the core of the process. Through content analysis, four principles emerged from the review that must be applied to create Human-Centred AI systems that can establish a symbiotic relationship with humans. In addition, current trends and challenges were defined to indicate open questions that may guide future research for the development of SAI systems that comply with the AI Act.", "date": "2025-01-14", "categories": []}
{"id": "2506.20606", "title": "Model Editing as a Double-Edged Sword: Steering Agent Ethical Behavior Toward Beneficence or Harm", "abstract": "Agents based on Large Language Models (LLMs) have demonstrated strong capabilities across a wide range of tasks. However, deploying LLM-based agents in high-stakes domains comes with significant safety and ethical risks. Unethical behavior by these agents can directly result in serious real-world consequences, including physical harm and financial loss. To efficiently steer the ethical behavior of agents, we frame agent behavior steering as a model editing task, which we term Behavior Editing. Model editing is an emerging area of research that enables precise and efficient modifications to LLMs while preserving their overall capabilities. To systematically study and evaluate this approach, we introduce BehaviorBench, a multi-tier benchmark grounded in psychological moral theories. This benchmark supports both the evaluation and editing of agent behaviors across a variety of scenarios, with each tier introducing more complex and ambiguous scenarios. We first demonstrate that Behavior Editing can dynamically steer agents toward the target behavior within specific scenarios. Moreover, Behavior Editing enables not only scenario-specific local adjustments but also more extensive shifts in an agent's global moral alignment. We demonstrate that Behavior Editing can be used to promote ethical and benevolent behavior or, conversely, to induce harmful or malicious behavior. Through extensive evaluations of agents built on frontier LLMs, BehaviorBench validates the effectiveness of behavior editing across a wide range of models and scenarios. Our findings offer key insights into a new paradigm for steering agent behavior, highlighting both the promise and perils of Behavior Editing.", "date": "2025-06-25", "categories": []}
{"id": "2511.14013", "title": "Developing a Grounded View of AI", "abstract": "As a capability coming from computation, how does AI differ fundamentally from the capabilities delivered by rule-based software program? The paper examines the behavior of artificial intelligence (AI) from engineering points of view to clarify its nature and limits. The paper argues that the rationality underlying humanity's impulse to pursue, articulate, and adhere to rules deserves to be valued and preserved. Identifying where rule-based practical rationality ends is the beginning of making it aware until action. Although the rules of AI behaviors are still hidden or only weakly observable, the paper has proposed a methodology to make a sense of discrimination possible and practical to identify the distinctions of the behavior of AI models with three types of decisions. It is a prerequisite for human responsibilities with alternative possibilities, considering how and when to use AI. It would be a solid start for people to ensure AI system soundness for the well-being of humans, society, and the environment.", "date": "2025-11-18", "categories": []}
{"id": "2508.04511", "title": "Argumentative Debates for Transparent Bias Detection [Technical Report]", "abstract": "As the use of AI in society grows, addressing emerging biases is essential to prevent systematic discrimination. Several bias detection methods have been proposed, but, with few exceptions, these tend to ignore transparency. Instead, interpretability and explainability are core requirements for algorithmic fairness, even more so than for other algorithmic solutions, given the human-oriented nature of fairness. We present ABIDE (Argumentative BIas detection by DEbate), a novel framework that structures bias detection transparently as debate, guided by an underlying argument graph as understood in (formal and computational) argumentation. The arguments are about the success chances of groups in local neighbourhoods and the significance of these neighbourhoods. We evaluate ABIDE experimentally and demonstrate its strengths in performance against an argumentative baseline.", "date": "2025-08-06", "categories": []}
{"id": "2511.10089", "title": "T2IBias: Uncovering Societal Bias Encoded in the Latent Space of Text-to-Image Generative Models", "abstract": "Text-to-image (T2I) generative models are largely used in AI-powered real-world applications and value creation. However, their strategic deployment raises critical concerns for responsible AI management, particularly regarding the reproduction and amplification of race- and gender-related stereotypes that can undermine organizational ethics. In this work, we investigate whether such societal biases are systematically encoded within the pretrained latent spaces of state-of-the-art T2I models. We conduct an empirical study across the five most popular open-source models, using ten neutral, profession-related prompts to generate 100 images per profession, resulting in a dataset of 5,000 images evaluated by diverse human assessors representing different races and genders. We demonstrate that all five models encode and amplify pronounced societal skew: caregiving and nursing roles are consistently feminized, while high-status professions such as corporate CEO, politician, doctor, and lawyer are overwhelmingly represented by males and mostly White individuals. We further identify model-specific patterns, such as QWEN-Image's near-exclusive focus on East Asian outputs, Kandinsky's dominance of White individuals, and SDXL's comparatively broader but still biased distributions. These results provide critical insights for AI project managers and practitioners, enabling them to select equitable AI models and customized prompts that generate images in alignment with the principles of responsible AI. We conclude by discussing the risks of these biases and proposing actionable strategies for bias mitigation in building responsible GenAI systems. The code and Data Repository: https://github.com/Sufianlab/T2IBias", "date": "2025-11-13", "categories": []}
{"id": "2511.12271", "title": "MoralReason: Generalizable Moral Decision Alignment For LLM Agents Using Reasoning-Level Reinforcement Learning", "abstract": "Large language models are increasingly influencing human moral decisions, yet current approaches focus primarily on evaluating rather than actively steering their moral decisions. We formulate this as an out-of-distribution moral alignment problem, where LLM agents must learn to apply consistent moral reasoning frameworks to scenarios beyond their training distribution. We introduce Moral-Reason-QA, a novel dataset extending 680 human-annotated, high-ambiguity moral scenarios with framework-specific reasoning traces across utilitarian, deontological, and virtue ethics, enabling systematic evaluation of moral generalization in realistic decision contexts. Our learning approach employs Group Relative Policy Optimization with composite rewards that simultaneously optimize decision alignment and framework-specific reasoning processes to facilitate learning of the underlying moral frameworks. Experimental results demonstrate successful generalization to unseen moral scenarios, with softmax-normalized alignment scores improving by +0.757 for utilitarian and +0.450 for deontological frameworks when tested on out-of-distribution evaluation sets. The experiments also reveal training challenges and promising directions that inform future research. These findings establish that LLM agents can be systematically trained to internalize and apply specific moral frameworks to novel situations, providing a critical foundation for AI safety as language models become more integrated into human decision-making processes.", "date": "2025-11-15", "categories": []}
{"id": "2511.13290", "title": "Dropouts in Confidence: Moral Uncertainty in Human-LLM Alignment", "abstract": "Humans display significant uncertainty when confronted with moral dilemmas, yet the extent of such uncertainty in machines and AI agents remains underexplored. Recent studies have confirmed the overly confident tendencies of machine-generated responses, particularly in large language models (LLMs). As these systems are increasingly embedded in ethical decision-making scenarios, it is important to understand their moral reasoning and the inherent uncertainties in building reliable AI systems. This work examines how uncertainty influences moral decisions in the classical trolley problem, analyzing responses from 32 open-source models and 9 distinct moral dimensions. We first find that variance in model confidence is greater across models than within moral dimensions, suggesting that moral uncertainty is predominantly shaped by model architecture and training method. To quantify uncertainty, we measure binary entropy as a linear combination of total entropy, conditional entropy, and mutual information. To examine its effects, we introduce stochasticity into models via \"dropout\" at inference time. Our findings show that our mechanism increases total entropy, mainly through a rise in mutual information, while conditional entropy remains largely unchanged. Moreover, this mechanism significantly improves human-LLM moral alignment, with correlations in mutual information and alignment score shifts. Our results highlight the potential to better align model-generated decisions and human preferences by deliberately modulating uncertainty and reducing LLMs' confidence in morally complex scenarios.", "date": "2025-11-17", "categories": []}
{"id": "2511.13525", "title": "AI Fairness Beyond Complete Demographics: Current Achievements and Future Directions", "abstract": "Fairness in artificial intelligence (AI) has become a growing concern due to discriminatory outcomes in AI-based decision-making systems. While various methods have been proposed to mitigate bias, most rely on complete demographic information, an assumption often impractical due to legal constraints and the risk of reinforcing discrimination. This survey examines fairness in AI when demographics are incomplete, addressing the gap between traditional approaches and real-world challenges. We introduce a novel taxonomy of fairness notions in this setting, clarifying their relationships and distinctions. Additionally, we summarize existing techniques that promote fairness beyond complete demographics and highlight open research questions to encourage further progress in the field.", "date": "2025-11-17", "categories": []}
{"id": "2312.08467", "title": "Culturally Responsive Artificial Intelligence -- Problems, Challenges and Solutions", "abstract": "In the contemporary interconnected world, the concept of cultural responsibility occupies paramount importance. As the lines between nations become less distinct, it is incumbent upon individuals, communities, and institutions to assume the responsibility of safeguarding and valuing the landscape of diverse cultures that constitute our global society. This paper explores the socio-cultural and ethical challenges stemming from the implementation of AI algorithms and highlights the necessity for their culturally responsive development. It also offers recommendations on essential elements required to enhance AI systems' adaptability to meet the demands of contemporary multicultural societies. The paper highlights the need for further multidisciplinary research to create AI models that effectively address these challenges. It also advocates the significance of AI enculturation and underlines the importance of regulatory measures to promote cultural responsibility in AI systems.", "date": "2023-12-13", "categories": []}
{"id": "2511.11144", "title": "Specification, Application, and Operationalization of a Metamodel of Fairness", "abstract": "This paper presents the AR fairness metamodel, aimed at formally representing, analyzing, and comparing fairness scenarios. The metamodel provides an abstract representation of fairness, enabling the formal definition of fairness notions. We instantiate the metamodel through several examples, with a particular focus on comparing the notions of equity and equality.\n  We use the Tiles framework, which offers modular components that can be interconnected to represent various definitions of fairness. Its primary objective is to support the operationalization of AR-based fairness definitions in a range of scenarios, providing a robust method for defining, comparing, and evaluating fairness.\n  Tiles has an open-source implementation for fairness modeling and evaluation.", "date": "2025-11-14", "categories": []}
{"id": "2505.18371", "title": "Military AI Needs Technically-Informed Regulation to Safeguard AI Research and its Applications", "abstract": "Military weapon systems and command-and-control infrastructure augmented by artificial intelligence (AI) have seen rapid development and deployment in recent years. However, the sociotechnical impacts of AI on combat systems, military decision-making, and the norms of warfare have been understudied. We focus on a specific subset of lethal autonomous weapon systems (LAWS) that use AI for targeting or battlefield decisions. We refer to this subset as AI-powered lethal autonomous weapon systems (AI-LAWS) and argue that they introduce novel risks -- including unanticipated escalation, poor reliability in unfamiliar environments, and erosion of human oversight -- all of which threaten both military effectiveness and the openness of AI research. These risks cannot be addressed by high-level policy alone; effective regulation must be grounded in the technical behavior of AI models. We argue that AI researchers must be involved throughout the regulatory lifecycle. Thus, we propose a clear, behavior-based definition of AI-LAWS -- systems that introduce unique risks through their use of modern AI -- as a foundation for technically grounded regulation, given that existing frameworks do not distinguish them from conventional LAWS. Using this definition, we propose several technically-informed policy directions and invite greater participation from the AI research community in military AI policy discussions.", "date": "2025-05-23", "categories": []}
{"id": "2501.17899", "title": "The Right to AI", "abstract": "This paper proposes a Right to AI, which asserts that individuals and communities should meaningfully participate in the development and governance of the AI systems that shape their lives. Motivated by the increasing deployment of AI in critical domains and inspired by Henri Lefebvre's concept of the Right to the City, we reconceptualize AI as a societal infrastructure, rather than merely a product of expert design. In this paper, we critically evaluate how generative agents, large-scale data extraction, and diverse cultural values bring new complexities to AI oversight. The paper proposes that grassroots participatory methodologies can mitigate biased outcomes and enhance social responsiveness. It asserts that data is socially produced and should be managed and owned collectively. Drawing on Sherry Arnstein's Ladder of Citizen Participation and analyzing nine case studies, the paper develops a four-tier model for the Right to AI that situates the current paradigm and envisions an aspirational future. It proposes recommendations for inclusive data ownership, transparent design processes, and stakeholder-driven oversight. We also discuss market-led and state-centric alternatives and argue that participatory approaches offer a better balance between technical efficiency and democratic legitimacy.", "date": "2025-01-29", "categories": []}
{"id": "2511.06592", "title": "MedVoiceBias: A Controlled Study of Audio LLM Behavior in Clinical Decision-Making", "abstract": "As large language models transition from text-based interfaces to audio interactions in clinical settings, they might introduce new vulnerabilities through paralinguistic cues in audio. We evaluated these models on 170 clinical cases, each synthesized into speech from 36 distinct voice profiles spanning variations in age, gender, and emotion. Our findings reveal a severe modality bias: surgical recommendations for audio inputs varied by as much as 35% compared to identical text-based inputs, with one model providing 80% fewer recommendations. Further analysis uncovered age disparities of up to 12% between young and elderly voices, which persisted in most models despite chain-of-thought prompting. While explicit reasoning successfully eliminated gender bias, the impact of emotion was not detected due to poor recognition performance. These results demonstrate that audio LLMs are susceptible to making clinical decisions based on a patient's voice characteristics rather than medical evidence, a flaw that risks perpetuating healthcare disparities. We conclude that bias-aware architectures are essential and urgently needed before the clinical deployment of these models.", "date": "2025-11-10", "categories": []}
{"id": "2410.21882", "title": "Building Altruistic and Moral AI Agent with Brain-inspired Emotional Empathy Mechanisms", "abstract": "As AI closely interacts with human society, it is crucial to ensure that its behavior is safe, altruistic, and aligned with human ethical and moral values. However, existing research on embedding ethical considerations into AI remains insufficient, and previous external constraints based on principles and rules are inadequate to provide AI with long-term stability and generalization capabilities. Emotional empathy intrinsically motivates altruistic behaviors aimed at alleviating others' negative emotions through emotional sharing and contagion mechanisms. Motivated by this, we draw inspiration from the neural mechanism of human emotional empathy-driven altruistic decision making, and simulate the shared self-other perception-mirroring-empathy neural circuits, to construct a brain-inspired emotional empathy-driven altruistic decision-making model. Here, empathy directly impacts dopamine release to form intrinsic altruistic motivation. The proposed model exhibits consistent altruistic behaviors across three experimental settings: emotional contagion-integrated two-agent altruistic rescue, multi-agent gaming, and robotic emotional empathy interaction scenarios. In-depth analyses validate the positive correlation between empathy levels and altruistic preferences (consistent with psychological behavioral experiment findings), while also demonstrating how interaction partners' empathy levels influence the agent's behavioral patterns. We further test the proposed model's performance and stability in moral dilemmas involving conflicts between self-interest and others' well-being, partially observable environments, and adversarial defense scenarios. This work provides preliminary exploration of human-like empathy-driven altruistic moral decision making, contributing potential perspectives for developing ethically-aligned AI.", "date": "2024-10-29", "categories": []}
{"id": "2511.04157", "title": "Are We Aligned? A Preliminary Investigation of the Alignment of Responsible AI Values between LLMs and Human Judgment", "abstract": "Large Language Models (LLMs) are increasingly employed in software engineering tasks such as requirements elicitation, design, and evaluation, raising critical questions regarding their alignment with human judgments on responsible AI values. This study investigates how closely LLMs' value preferences align with those of two human groups: a US-representative sample and AI practitioners. We evaluate 23 LLMs across four tasks: (T1) selecting key responsible AI values, (T2) rating their importance in specific contexts, (T3) resolving trade-offs between competing values, and (T4) prioritizing software requirements that embody those values. The results show that LLMs generally align more closely with AI practitioners than with the US-representative sample, emphasizing fairness, privacy, transparency, safety, and accountability. However, inconsistencies appear between the values that LLMs claim to uphold (Tasks 1-3) and the way they prioritize requirements (Task 4), revealing gaps in faithfulness between stated and applied behavior. These findings highlight the practical risk of relying on LLMs in requirements engineering without human oversight and motivate the need for systematic approaches to benchmark, interpret, and monitor value alignment in AI-assisted software development.", "date": "2025-11-06", "categories": []}
{"id": "2311.17228", "title": "Survey on AI Ethics: A Socio-technical Perspective", "abstract": "The past decade has observed a significant advancement in AI with deep learning-based models being deployed in diverse scenarios, including safety-critical applications. As these AI systems become deeply embedded in our societal infrastructure, the repercussions of their decisions and actions have significant consequences, making the ethical implications of AI deployment highly relevant and essential. The ethical concerns associated with AI are multifaceted, including challenging issues of fairness, privacy and data protection, responsibility and accountability, safety and robustness, transparency and explainability, and environmental impact. These principles together form the foundations of ethical AI considerations that concern every stakeholder in the AI system lifecycle. In light of the present ethical and future x-risk concerns, governments have shown increasing interest in establishing guidelines for the ethical deployment of AI. This work unifies the current and future ethical concerns of deploying AI into society. While we acknowledge and appreciate the technical surveys for each of the ethical principles concerned, in this paper, we aim to provide a comprehensive overview that not only addresses each principle from a technical point of view but also discusses them from a social perspective.", "date": "2023-11-28", "categories": []}
{"id": "2510.25863", "title": "AAGATE: A NIST AI RMF-Aligned Governance Platform for Agentic AI", "abstract": "This paper introduces the Agentic AI Governance Assurance & Trust Engine (AAGATE), a Kubernetes-native control plane designed to address the unique security and governance challenges posed by autonomous, language-model-driven agents in production. Recognizing the limitations of traditional Application Security (AppSec) tooling for improvisational, machine-speed systems, AAGATE operationalizes the NIST AI Risk Management Framework (AI RMF). It integrates specialized security frameworks for each RMF function: the Agentic AI Threat Modeling MAESTRO framework for Map, a hybrid of OWASP's AIVSS and SEI's SSVC for Measure, and the Cloud Security Alliance's Agentic AI Red Teaming Guide for Manage. By incorporating a zero-trust service mesh, an explainable policy engine, behavioral analytics, and decentralized accountability hooks, AAGATE provides a continuous, verifiable governance solution for agentic AI, enabling safe, accountable, and scalable deployment. The framework is further extended with DIRF for digital identity rights, LPCI defenses for logic-layer injection, and QSAF monitors for cognitive degradation, ensuring governance spans systemic, adversarial, and ethical risks.", "date": "2025-10-29", "categories": []}
{"id": "2501.10384", "title": "Nirvana AI Governance: How AI Policymaking Is Committing Three Old Fallacies", "abstract": "This research applies Harold Demsetz's concept of the nirvana approach to the realm of AI governance and debunks three common fallacies in various AI policy proposals--\"the grass is always greener on the other side,\" \"free lunch,\" and \"the people could be different.\" Through this, I expose fundamental flaws in the current AI regulatory proposal. First, some commentators intuitively believe that people are more reliable than machines and that government works better in risk control than companies' self-regulation, but they do not fully compare the differences between the status quo and the proposed replacements. Second, when proposing some regulatory tools, some policymakers and researchers do not realize and even gloss over the fact that harms and costs are also inherent in their proposals. Third, some policy proposals are initiated based on a false comparison between the AI-driven world, where AI does lead to some risks, and an entirely idealized world, where no risk exists at all. However, the appropriate approach is to compare the world where AI causes risks to the real world where risks are everywhere, but people can live well with these risks. The prevalence of these fallacies in AI governance underscores a broader issue: the tendency to idealize potential solutions without fully considering their real-world implications. This idealization can lead to regulatory proposals that are not only impractical but potentially harmful to innovation and societal progress.", "date": "2024-12-18", "categories": []}
{"id": "2504.07170", "title": "Trustworthy AI Must Account for Interactions", "abstract": "Trustworthy AI encompasses many aspirational aspects for aligning AI systems with human values, including fairness, privacy, robustness, explainability, and uncertainty quantification. Ultimately the goal of Trustworthy AI research is to achieve all aspects simultaneously. However, efforts to enhance one aspect often introduce unintended trade-offs that negatively impact others. In this position paper, we review notable approaches to these five aspects and systematically consider every pair, detailing the negative interactions that can arise. For example, applying differential privacy to model training can amplify biases, undermining fairness. Drawing on these findings, we take the position that current research practices of improving one or two aspects in isolation are insufficient. Instead, research on Trustworthy AI must account for interactions between aspects and adopt a holistic view across all relevant axes at once. To illustrate our perspective, we provide guidance on how practitioners can work towards integrated trust, examples of how interactions affect the financial industry, and alternative views.", "date": "2025-04-09", "categories": []}
{"id": "2507.03525", "title": "Limits of Safe AI Deployment: Differentiating Oversight and Control", "abstract": "Oversight and control, which we collectively call supervision, are often discussed as ways to ensure that AI systems are accountable, reliable, and able to fulfill governance and management requirements. However, the requirements for \"human oversight\" risk codifying vague or inconsistent interpretations of key concepts like oversight and control. This ambiguous terminology could undermine efforts to design or evaluate systems that must operate under meaningful human supervision. This matters because the term is used by regulatory texts such as the EU AI Act.\n  This paper undertakes a targeted critical review of literature on supervision outside of AI, along with a brief summary of past work on the topic related to AI. We next differentiate control as ex-ante or real-time and operational rather than policy or governance, and oversight as performed ex-post, or a policy and governance function. Control aims to prevent failures, while oversight focuses on detection, remediation, or incentives for future prevention. Building on this, we make three contributions. 1) We propose a framework to align regulatory expectations with what is technically and organizationally plausible, articulating the conditions under which each mechanism is possible, where they fall short, and what is required to make them meaningful in practice. 2) We outline how supervision methods should be documented and integrated into risk management, and drawing on the Microsoft Responsible AI Maturity Model, we outline a maturity model for AI supervision. 3) We explicitly highlight boundaries of these mechanisms, including where they apply, where they fail, and where it is clear that no existing methods suffice. This foregrounds the question of whether meaningful supervision is possible in a given deployment context, and can support regulators, auditors, and practitioners in identifying both present and future limitations.", "date": "2025-07-04", "categories": []}
{"id": "2511.00027", "title": "Position Paper: If Innovation in AI Systematically Violates Fundamental Rights, Is It Innovation at All?", "abstract": "Artificial intelligence (AI) now permeates critical infrastructures and decision-making systems where failures produce social, economic, and democratic harm. This position paper challenges the entrenched belief that regulation and innovation are opposites. As evidenced by analogies from aviation, pharmaceuticals, and welfare systems and recent cases of synthetic misinformation, bias and unaccountable decision-making, the absence of well-designed regulation has already created immeasurable damage. Regulation, when thoughtful and adaptive, is not a brake on innovation -- it is its foundation. The present position paper examines the EU AI Act as a model of risk-based, responsibility-driven regulation that addresses the Collingridge Dilemma: acting early enough to prevent harm, yet flexibly enough to sustain innovation. Its adaptive mechanisms -- regulatory sandboxes, small and medium enterprises (SMEs) support, real-world testing, fundamental rights impact assessment (FRIA) -- demonstrate how regulation can accelerate responsibly, rather than delay, technological progress. The position paper summarises how governance tools transform perceived burdens into tangible advantages: legal certainty, consumer trust, and ethical competitiveness. Ultimately, the paper reframes progress: innovation and regulation advance together. By embedding transparency, impact assessments, accountability, and AI literacy into design and deployment, the EU framework defines what responsible innovation truly means -- technological ambition disciplined by democratic values and fundamental rights.", "date": "2025-10-26", "categories": []}
{"id": "2501.10484", "title": "Bias in Decision-Making for AI's Ethical Dilemmas: A Comparative Study of ChatGPT and Claude", "abstract": "Recent advances in Large Language Models (LLMs) have enabled human-like responses across various tasks, raising questions about their ethical decision-making capabilities and potential biases. This study systematically evaluates how nine popular LLMs (both open-source and closed-source) respond to ethical dilemmas involving protected attributes. Across 50,400 trials spanning single and intersectional attribute combinations in four dilemma scenarios (protective vs. harmful), we assess models' ethical preferences, sensitivity, stability, and clustering patterns. Results reveal significant biases in protected attributes in all models, with differing preferences depending on model type and dilemma context. Notably, open-source LLMs show stronger preferences for marginalized groups and greater sensitivity in harmful scenarios, while closed-source models are more selective in protective situations and tend to favor mainstream groups. We also find that ethical behavior varies across dilemma types: LLMs maintain consistent patterns in protective scenarios but respond with more diverse and cognitively demanding decisions in harmful ones. Furthermore, models display more pronounced ethical tendencies under intersectional conditions than in single-attribute settings, suggesting that complex inputs reveal deeper biases. These findings highlight the need for multi-dimensional, context-aware evaluation of LLMs' ethical behavior and offer a systematic evaluation and approach to understanding and addressing fairness in LLM decision-making.", "date": "2025-01-17", "categories": []}
{"id": "2503.00333", "title": "More of the Same: Persistent Representational Harms Under Increased Representation", "abstract": "To recognize and mitigate the harms of generative AI systems, it is crucial to consider whether and how different societal groups are represented by these systems. A critical gap emerges when naively measuring or improving who is represented, as this does not consider how people are represented. In this work, we develop GAS(P), an evaluation methodology for surfacing distribution-level group representational biases in generated text, tackling the setting where groups are unprompted (i.e., groups are not specified in the input to generative systems). We apply this novel methodology to investigate gendered representations in occupations across state-of-the-art large language models. We show that, even though the gender distribution when models are prompted to generate biographies leads to a large representation of women, even representational biases persist in how different genders are represented. Our evaluation methodology reveals that there are statistically significant distribution-level differences in the word choice used to describe biographies and personas of different genders across occupations, and we show that many of these differences are associated with representational harms and stereotypes. Our empirical findings caution that naively increasing (unprompted) representation may inadvertently proliferate representational biases, and our proposed evaluation methodology enables systematic and rigorous measurement of the problem.", "date": "2025-03-01", "categories": []}
{"id": "2505.10603", "title": "Toward a Public and Secure Generative AI: A Comparative Analysis of Open and Closed LLMs", "abstract": "Generative artificial intelligence (Gen AI) systems represent a critical technology with far-reaching implications across multiple domains of society. However, their deployment entails a range of risks and challenges that require careful evaluation. To date, there has been a lack of comprehensive, interdisciplinary studies offering a systematic comparison between open-source and proprietary (closed) generative AI systems, particularly regarding their respective advantages and drawbacks. This study aims to: i) critically evaluate and compare the characteristics, opportunities, and challenges of open and closed generative AI models; and ii) propose foundational elements for the development of an Open, Public, and Safe Gen AI framework. As a methodology, we adopted a combined approach that integrates three methods: literature review, critical analysis, and comparative analysis. The proposed framework outlines key dimensions, openness, public governance, and security, as essential pillars for shaping the future of trustworthy and inclusive Gen AI. Our findings reveal that open models offer greater transparency, auditability, and flexibility, enabling independent scrutiny and bias mitigation. In contrast, closed systems often provide better technical support and ease of implementation, but at the cost of unequal access, accountability, and ethical oversight. The research also highlights the importance of multi-stakeholder governance, environmental sustainability, and regulatory frameworks in ensuring responsible development.", "date": "2025-05-15", "categories": []}
{"id": "2510.26007", "title": "The Quest for Reliable Metrics of Responsible AI", "abstract": "The development of Artificial Intelligence (AI), including AI in Science (AIS), should be done following the principles of responsible AI. Progress in responsible AI is often quantified through evaluation metrics, yet there has been less work on assessing the robustness and reliability of the metrics themselves. We reflect on prior work that examines the robustness of fairness metrics for recommender systems as a type of AI application and summarise their key takeaways into a set of non-exhaustive guidelines for developing reliable metrics of responsible AI. Our guidelines apply to a broad spectrum of AI applications, including AIS.", "date": "2025-10-29", "categories": []}
{"id": "2510.26057", "title": "Can AI be Accountable?", "abstract": "The AI we use is powerful, and its power is increasing rapidly. If this powerful AI is to serve the needs of consumers, voters, and decision makers, then it is imperative that the AI is accountable. In general, an agent is accountable to a forum if the forum can request information from the agent about its actions, if the forum and the agent can discuss this information, and if the forum can sanction the agent. Unfortunately, in too many cases today's AI is not accountable -- we cannot question it, enter into a discussion with it, let alone sanction it. In this chapter we relate the general definition of accountability to AI, we illustrate what it means for AI to be accountable and unaccountable, and we explore approaches that can improve our chances of living in a world where all AI is accountable to those who are affected by it.", "date": "2025-10-30", "categories": []}
{"id": "2510.26396", "title": "A Pragmatic View of AI Personhood", "abstract": "The emergence of agentic Artificial Intelligence (AI) is set to trigger a \"Cambrian explosion\" of new kinds of personhood. This paper proposes a pragmatic framework for navigating this diversification by treating personhood not as a metaphysical property to be discovered, but as a flexible bundle of obligations (rights and responsibilities) that societies confer upon entities for a variety of reasons, especially to solve concrete governance problems. We argue that this traditional bundle can be unbundled, creating bespoke solutions for different contexts. This will allow for the creation of practical tools -- such as facilitating AI contracting by creating a target \"individual\" that can be sanctioned -- without needing to resolve intractable debates about an AI's consciousness or rationality. We explore how individuals fit in to social roles and discuss the use of decentralized digital identity technology, examining both \"personhood as a problem\", where design choices can create \"dark patterns\" that exploit human social heuristics, and \"personhood as a solution\", where conferring a bundle of obligations is necessary to ensure accountability or prevent conflict. By rejecting foundationalist quests for a single, essential definition of personhood, this paper offers a more pragmatic and flexible way to think about integrating AI agents into our society.", "date": "2025-10-30", "categories": []}
{"id": "2507.22664", "title": "RobEthiChor: Automated Context-aware Ethics-based Negotiation for Autonomous Robots", "abstract": "The presence of autonomous systems is growing at a fast pace and it is impacting many aspects of our lives. Designed to learn and act independently, these systems operate and perform decision-making without human intervention. However, they lack the ability to incorporate users' ethical preferences, which are unique for each individual in society and are required to personalize the decision-making processes. This reduces user trust and prevents autonomous systems from behaving according to the moral beliefs of their end-users. When multiple systems interact with differing ethical preferences, they must negotiate to reach an agreement that satisfies the ethical beliefs of all the parties involved and adjust their behavior consequently. To address this challenge, this paper proposes RobEthiChor, an approach that enables autonomous systems to incorporate user ethical preferences and contextual factors into their decision-making through ethics-based negotiation. RobEthiChor features a domain-agnostic reference architecture for designing autonomous systems capable of ethic-based negotiating. The paper also presents RobEthiChor-Ros, an implementation of RobEthiChor within the Robot Operating System (ROS), which can be deployed on robots to provide them with ethics-based negotiation capabilities. To evaluate our approach, we deployed RobEthiChor-Ros on real robots and ran scenarios where a pair of robots negotiate upon resource contention. Experimental results demonstrate the feasibility and effectiveness of the system in realizing ethics-based negotiation. RobEthiChor allowed robots to reach an agreement in more than 73% of the scenarios with an acceptable negotiation time (0.67s on average). Experiments also demonstrate that the negotiation approach implemented in RobEthiChor is scalable.", "date": "2025-07-30", "categories": []}
{"id": "2508.07132", "title": "\"Draw me a curator\" Examining the visual stereotyping of a cultural services profession by generative AI", "abstract": "Based on 230 visualisations, this paper examines the depiction of museum curators by the popular generative Artificial Intelligence (AI) model, ChatGPT4o. While the AI-generated representations do not reiterate popular stereotypes of curators as nerdy, conservative in dress and stuck in time rummaging through collections, they contrast sharply with real-world demographics. AI-generated imagery extremely underrepresents women (3.5% vs 49% to 72% in reality) and disregards ethnic communities other than Caucasian (0% vs 18% to 36%). It only over-represents young curators (79% vs approx. 27%) but also renders curators to resemble yuppie professionals or people featuring in fashion advertising. Stereotypical attributes are prevalent, with curators widely depicted as wearing beards and holding clipboards or digital tablets. The findings highlight biases in the generative AI image creation dataset, which is poised to shape an inaccurate portrayal of museum professionals if the images were to be taken uncritically at face value.", "date": "2025-08-10", "categories": []}
{"id": "2510.24823", "title": "Do Chatbots Walk the Talk of Responsible AI?", "abstract": "This study examines whether leading AI chatbot companies implement the responsible AI principles they publicly advocate. The authors used a mixed-methods approach analyzing four major chatbots (ChatGPT, Gemini, DeepSeek, and Grok) across company websites, technical documentation, and direct chatbot evaluations. We found significant gaps between corporate rhetoric and practice.", "date": "2025-10-28", "categories": []}
{"id": "2505.16291", "title": "Fairness under Competition", "abstract": "Algorithmic fairness has emerged as a central issue in ML, and it has become standard practice to adjust ML algorithms so that they will satisfy fairness requirements such as Equal Opportunity. In this paper we consider the effects of adopting such fair classifiers on the overall level of ecosystem fairness. Specifically, we introduce the study of fairness with competing firms, and demonstrate the failure of fair classifiers in yielding fair ecosystems. Our results quantify the loss of fairness in systems, under a variety of conditions, based on classifiers' correlation and the level of their data overlap. We show that even if competing classifiers are individually fair, the ecosystem's outcome may be unfair; and that adjusting biased algorithms to improve their individual fairness may lead to an overall decline in ecosystem fairness. In addition to these theoretical results, we also provide supporting experimental evidence. Together, our model and results provide a novel and essential call for action.", "date": "2025-05-22", "categories": []}
{"id": "2506.10192", "title": "Towards Responsible AI: Advances in Safety, Fairness, and Accountability of Autonomous Systems", "abstract": "Ensuring responsible use of artificial intelligence (AI) has become imperative as autonomous systems increasingly influence critical societal domains. However, the concept of trustworthy AI remains broad and multi-faceted. This thesis advances knowledge in the safety, fairness, transparency, and accountability of AI systems. In safety, we extend classical deterministic shielding techniques to become resilient against delayed observations, enabling practical deployment in real-world conditions. We also implement both deterministic and probabilistic safety shields into simulated autonomous vehicles to prevent collisions with road users, validating the use of these techniques in realistic driving simulators. We introduce fairness shields, a novel post-processing approach to enforce group fairness in sequential decision-making settings over finite and periodic time horizons. By optimizing intervention costs while strictly ensuring fairness constraints, this method efficiently balances fairness with minimal interference. For transparency and accountability, we propose a formal framework for assessing intentional behaviour in probabilistic decision-making agents, introducing quantitative metrics of agency and intention quotient. We use these metrics to propose a retrospective analysis of intention, useful for determining responsibility when autonomous systems cause unintended harm. Finally, we unify these contributions through the ``reactive decision-making'' framework, providing a general formalization that consolidates previous approaches. Collectively, the advancements presented contribute practically to the realization of safer, fairer, and more accountable AI systems, laying the foundations for future research in trustworthy AI.", "date": "2025-06-11", "categories": []}
{"id": "2508.00138", "title": "Co-Producing AI: Toward an Augmented, Participatory Lifecycle", "abstract": "Despite efforts to mitigate the inherent risks and biases of artificial intelligence (AI) algorithms, these algorithms can disproportionately impact culturally marginalized groups. A range of approaches has been proposed to address or reduce these risks, including the development of ethical guidelines and principles for responsible AI, as well as technical solutions that promote algorithmic fairness. Drawing on design justice, expansive learning theory, and recent empirical work on participatory AI, we argue that mitigating these harms requires a fundamental re-architecture of the AI production pipeline. This re-design should center co-production, diversity, equity, inclusion (DEI), and multidisciplinary collaboration. We introduce an augmented AI lifecycle consisting of five interconnected phases: co-framing, co-design, co-implementation, co-deployment, and co-maintenance. The lifecycle is informed by four multidisciplinary workshops and grounded in themes of distributed authority and iterative knowledge exchange. Finally, we relate the proposed lifecycle to several leading ethical frameworks and outline key research questions that remain for scaling participatory governance.", "date": "2025-07-31", "categories": []}
{"id": "2509.13345", "title": "Beyond Accuracy: Rethinking Hallucination and Regulatory Response in Generative AI", "abstract": "Hallucination in generative AI is often treated as a technical failure to produce factually correct output. Yet this framing underrepresents the broader significance of hallucinated content in language models, which may appear fluent, persuasive, and contextually appropriate while conveying distortions that escape conventional accuracy checks. This paper critically examines how regulatory and evaluation frameworks have inherited a narrow view of hallucination, one that prioritises surface verifiability over deeper questions of meaning, influence, and impact. We propose a layered approach to understanding hallucination risks, encompassing epistemic instability, user misdirection, and social-scale effects. Drawing on interdisciplinary sources and examining instruments such as the EU AI Act and the GDPR, we show that current governance models struggle to address hallucination when it manifests as ambiguity, bias reinforcement, or normative convergence. Rather than improving factual precision alone, we argue for regulatory responses that account for languages generative nature, the asymmetries between system and user, and the shifting boundaries between information, persuasion, and harm.", "date": "2025-09-12", "categories": []}
{"id": "2510.21043", "title": "Epistemic Deference to AI", "abstract": "When should we defer to AI outputs over human expert judgment? Drawing on recent work in social epistemology, I motivate the idea that some AI systems qualify as Artificial Epistemic Authorities (AEAs) due to their demonstrated reliability and epistemic superiority. I then introduce AI Preemptionism, the view that AEA outputs should replace rather than supplement a user's independent epistemic reasons. I show that classic objections to preemptionism - such as uncritical deference, epistemic entrenchment, and unhinging epistemic bases - apply in amplified form to AEAs, given their opacity, self-reinforcing authority, and lack of epistemic failure markers. Against this, I develop a more promising alternative: a total evidence view of AI deference. According to this view, AEA outputs should function as contributory reasons rather than outright replacements for a user's independent epistemic considerations. This approach has three key advantages: (i) it mitigates expertise atrophy by keeping human users engaged, (ii) it provides an epistemic case for meaningful human oversight and control, and (iii) it explains the justified mistrust of AI when reliability conditions are unmet. While demanding in practice, this account offers a principled way to determine when AI deference is justified, particularly in high-stakes contexts requiring rigorous reliability.", "date": "2025-10-23", "categories": []}
{"id": "2510.20109", "title": "The Verification-Value Paradox: A Normative Critique of Gen AI in Legal Practice", "abstract": "It is often claimed that machine learning-based generative AI products will drastically streamline and reduce the cost of legal practice. This enthusiasm assumes lawyers can effectively manage AI's risks. Cases in Australia and elsewhere in which lawyers have been reprimanded for submitting inaccurate AI-generated content to courts suggest this paradigm must be revisited. This paper argues that a new paradigm is needed to evaluate AI use in practice, given (a) AI's disconnection from reality and its lack of transparency, and (b) lawyers' paramount duties like honesty, integrity, and not to mislead the court. It presents an alternative model of AI use in practice that more holistically reflects these features (the verification-value paradox). That paradox suggests increases in efficiency from AI use in legal practice will be met by a correspondingly greater imperative to manually verify any outputs of that use, rendering the net value of AI use often negligible to lawyers. The paper then sets out the paradox's implications for legal practice and legal education, including for AI use but also the values that the paradox suggests should undergird legal practice: fidelity to the truth and civic responsibility.", "date": "2025-10-23", "categories": []}
{"id": "2510.20332", "title": "Bias by Design? How Data Practices Shape Fairness in AI Healthcare Systems", "abstract": "Artificial intelligence (AI) holds great promise for transforming healthcare. However, despite significant advances, the integration of AI solutions into real-world clinical practice remains limited. A major barrier is the quality and fairness of training data, which is often compromised by biased data collection practices. This paper draws on insights from the AI4HealthyAging project, part of Spain's national R&D initiative, where our task was to detect biases during clinical data collection. We identify several types of bias across multiple use cases, including historical, representation, and measurement biases. These biases manifest in variables such as sex, gender, age, habitat, socioeconomic status, equipment, and labeling. We conclude with practical recommendations for improving the fairness and robustness of clinical problem design and data collection. We hope that our findings and experience contribute to guiding future projects in the development of fairer AI systems in healthcare.", "date": "2025-10-23", "categories": []}
{"id": "2503.22023", "title": "Safeguarding Autonomy: a Focus on Machine Learning Decision Systems", "abstract": "As global discourse on AI regulation gains momentum, this paper focuses on delineating the impact of ML on autonomy and fostering awareness. Respect for autonomy is a basic principle in bioethics that establishes persons as decision-makers. While the concept of autonomy in the context of ML appears in several European normative publications, it remains a theoretical concept that has yet to be widely accepted in ML practice. Our contribution is to bridge the theoretical and practical gap by encouraging the practical application of autonomy in decision-making within ML practice by identifying the conditioning factors that currently prevent it. Consequently, we focus on the different stages of the ML pipeline to identify the potential effects on ML end-users' autonomy. To improve its practical utility, we propose a related question for each detected impact, offering guidance for identifying possible focus points to respect ML end-users autonomy in decision-making.", "date": "2025-03-27", "categories": []}
{"id": "2510.16206", "title": "The Right to Be Remembered: Preserving Maximally Truthful Digital Memory in the Age of AI", "abstract": "Since the rapid expansion of large language models (LLMs), people have begun to rely on them for information retrieval. While traditional search engines display ranked lists of sources shaped by search engine optimization (SEO), advertising, and personalization, LLMs typically provide a synthesized response that feels singular and authoritative. While both approaches carry risks of bias and omission, LLMs may amplify the effect by collapsing multiple perspectives into one answer, reducing users ability or inclination to compare alternatives. This concentrates power over information in a few LLM vendors whose systems effectively shape what is remembered and what is overlooked. As a result, certain narratives, individuals or groups, may be disproportionately suppressed, while others are disproportionately elevated. Over time, this creates a new threat: the gradual erasure of those with limited digital presence, and the amplification of those already prominent, reshaping collective memory. To address these concerns, this paper presents a concept of the Right To Be Remembered (RTBR) which encompasses minimizing the risk of AI-driven information omission, embracing the right of fair treatment, while ensuring that the generated content would be maximally truthful.", "date": "2025-10-17", "categories": []}
{"id": "2510.18559", "title": "RAISE: A Unified Framework for Responsible AI Scoring and Evaluation", "abstract": "As AI systems enter high-stakes domains, evaluation must extend beyond predictive accuracy to include explainability, fairness, robustness, and sustainability. We introduce RAISE (Responsible AI Scoring and Evaluation), a unified framework that quantifies model performance across these four dimensions and aggregates them into a single, holistic Responsibility Score. We evaluated three deep learning models: a Multilayer Perceptron (MLP), a Tabular ResNet, and a Feature Tokenizer Transformer, on structured datasets from finance, healthcare, and socioeconomics. Our findings reveal critical trade-offs: the MLP demonstrated strong sustainability and robustness, the Transformer excelled in explainability and fairness at a very high environmental cost, and the Tabular ResNet offered a balanced profile. These results underscore that no single model dominates across all responsibility criteria, highlighting the necessity of multi-dimensional evaluation for responsible model selection. Our implementation is available at: https://github.com/raise-framework/raise.", "date": "2025-10-21", "categories": []}
{"id": "2502.02649", "title": "Fully Autonomous AI Agents Should Not be Developed", "abstract": "This paper argues that fully autonomous AI agents should not be developed. In support of this position, we build from prior scientific literature and current product marketing to delineate different AI agent levels and detail the ethical values at play in each, documenting trade-offs in potential benefits and risks. Our analysis reveals that risks to people increase with the autonomy of a system: The more control a user cedes to an AI agent, the more risks to people arise. Particularly concerning are safety risks, which affect human life and impact further values.", "date": "2025-02-04", "categories": []}
{"id": "2510.16056", "title": "Algorithmic Fairness in AI Surrogates for End-of-Life Decision-Making", "abstract": "Artificial intelligence surrogates are systems designed to infer preferences when individuals lose decision-making capacity. Fairness in such systems is a domain that has been insufficiently explored. Traditional algorithmic fairness frameworks are insufficient for contexts where decisions are relational, existential, and culturally diverse. This paper explores an ethical framework for algorithmic fairness in AI surrogates by mapping major fairness notions onto potential real-world end-of-life scenarios. It then examines fairness across moral traditions. The authors argue that fairness in this domain extends beyond parity of outcomes to encompass moral representation, fidelity to the patient's values, relationships, and worldview.", "date": "2025-10-16", "categories": []}
{"id": "2510.14676", "title": "NAEL: Non-Anthropocentric Ethical Logic", "abstract": "We introduce NAEL (Non-Anthropocentric Ethical Logic), a novel ethical framework for artificial agents grounded in active inference and symbolic reasoning. Departing from conventional, human-centred approaches to AI ethics, NAEL formalizes ethical behaviour as an emergent property of intelligent systems minimizing global expected free energy in dynamic, multi-agent environments. We propose a neuro-symbolic architecture to allow agents to evaluate the ethical consequences of their actions in uncertain settings. The proposed system addresses the limitations of existing ethical models by allowing agents to develop context-sensitive, adaptive, and relational ethical behaviour without presupposing anthropomorphic moral intuitions. A case study involving ethical resource distribution illustrates NAEL's dynamic balancing of self-preservation, epistemic learning, and collective welfare.", "date": "2025-10-16", "categories": []}
{"id": "2505.21479", "title": "Are Language Models Consequentialist or Deontological Moral Reasoners?", "abstract": "As AI systems increasingly navigate applications in healthcare, law, and governance, understanding how they handle ethically complex scenarios becomes critical. Previous work has mainly examined the moral judgments in large language models (LLMs), rather than their underlying moral reasoning process. In contrast, we focus on a large-scale analysis of the moral reasoning traces provided by LLMs. Furthermore, unlike prior work that attempted to draw inferences from only a handful of moral dilemmas, our study leverages over 600 distinct trolley problems as probes for revealing the reasoning patterns that emerge within different LLMs. We introduce and test a taxonomy of moral rationales to systematically classify reasoning traces according to two main normative ethical theories: consequentialism and deontology. Our analysis reveals that LLM chains-of-thought tend to favor deontological principles based on moral obligations, while post-hoc explanations shift notably toward consequentialist rationales that emphasize utility. Our framework provides a foundation for understanding how LLMs process and articulate ethical considerations, an important step toward safe and interpretable deployment of LLMs in high-stakes decision-making environments. Our code is available at https://github.com/keenansamway/moral-lens .", "date": "2025-05-27", "categories": []}
{"id": "2510.09634", "title": "Responsible AI Adoption in the Public Sector: A Data-Centric Taxonomy of AI Adoption Challenges", "abstract": "Despite Artificial Intelligence (AI) transformative potential for public sector services, decision-making, and administrative efficiency, adoption remains uneven due to complex technical, organizational, and institutional challenges. Responsible AI frameworks emphasize fairness, accountability, and transparency, aligning with principles of trustworthy AI and fair AI, yet remain largely aspirational, overlooking technical and institutional realities, especially foundational data and governance. This study addresses this gap by developing a taxonomy of data-related challenges to responsible AI adoption in government. Based on a systematic review of 43 studies and 21 expert evaluations, the taxonomy identifies 13 key challenges across technological, organizational, and environmental dimensions, including poor data quality, limited AI-ready infrastructure, weak governance, misalignment in human-AI decision-making, economic and environmental sustainability concerns. Annotated with institutional pressures, the taxonomy serves as a diagnostic tool to surface 'symptoms' of high-risk AI deployment and guides policymakers in building the institutional and data governance conditions necessary for responsible AI adoption.", "date": "2025-09-29", "categories": []}
{"id": "2510.10339", "title": "Measuring What Matters: Connecting AI Ethics Evaluations to System Attributes, Hazards, and Harms", "abstract": "Over the past decade, an ecosystem of measures has emerged to evaluate the social and ethical implications of AI systems, largely shaped by high-level ethics principles. These measures are developed and used in fragmented ways, without adequate attention to how they are situated in AI systems. In this paper, we examine how existing measures used in the computing literature map to AI system components, attributes, hazards, and harms. Our analysis draws on a scoping review resulting in nearly 800 measures corresponding to 11 AI ethics principles. We find that most measures focus on four principles - fairness, transparency, privacy, and trust - and primarily assess model or output system components. Few measures account for interactions across system elements, and only a narrow set of hazards is typically considered for each harm type. Many measures are disconnected from where harm is experienced and lack guidance for setting meaningful thresholds. These patterns reveal how current evaluation practices remain fragmented, measuring in pieces rather than capturing how harms emerge across systems. Framing measures with respect to system attributes, hazards, and harms can strengthen regulatory oversight, support actionable practices in industry, and ground future research in systems-level understanding.", "date": "2025-10-11", "categories": []}
{"id": "2510.11595", "title": "Reproducibility: The New Frontier in AI Governance", "abstract": "AI policymakers are responsible for delivering effective governance mechanisms that can provide safe, aligned and trustworthy AI development. However, the information environment offered to policymakers is characterised by an unnecessarily low Signal-To-Noise Ratio, favouring regulatory capture and creating deep uncertainty and divides on which risks should be prioritised from a governance perspective. We posit that the current publication speeds in AI combined with the lack of strong scientific standards, via weak reproducibility protocols, effectively erodes the power of policymakers to enact meaningful policy and governance protocols. Our paper outlines how AI research could adopt stricter reproducibility guidelines to assist governance endeavours and improve consensus on the AI risk landscape. We evaluate the forthcoming reproducibility crisis within AI research through the lens of crises in other scientific domains; providing a commentary on how adopting preregistration, increased statistical power and negative result publication reproducibility protocols can enable effective AI governance. While we maintain that AI governance must be reactive due to AI's significant societal implications we argue that policymakers and governments must consider reproducibility protocols as a core tool in the governance arsenal and demand higher standards for AI research. Code to replicate data and figures: https://github.com/IFMW01/reproducibility-the-new-frontier-in-ai-governance", "date": "2025-10-13", "categories": []}
{"id": "2510.08628", "title": "The Digital Mirror: Gender Bias and Occupational Stereotypes in AI-Generated Images", "abstract": "Generative AI offers vast opportunities for creating visualisations, such as graphics, videos, and images. However, recent studies around AI-generated visualisations have primarily focused on the creation process and image quality, overlooking representational biases. This study addresses this gap by testing representation biases in AI-generated pictures in an occupational setting and evaluating how two AI image generator tools, DALL-E 3 and Ideogram, compare. Additionally, the study discusses topics such as ageing and emotions in AI-generated images. As AI image tools are becoming more widely used, addressing and mitigating harmful gender biases becomes essential to ensure diverse representation in media and professional settings. In this study, over 750 AI-generated images of occupations were prompted. The thematic analysis results revealed that both DALL-E 3 and Ideogram reinforce traditional gender stereotypes in AI-generated images, although to varying degrees. These findings emphasise that AI visualisation tools risk reinforcing narrow representations. In our discussion section, we propose suggestions for practitioners, individuals and researchers to increase representation when generating images with visible genders.", "date": "2025-10-08", "categories": []}
{"id": "2510.09090", "title": "AI and Human Oversight: A Risk-Based Framework for Alignment", "abstract": "As Artificial Intelligence (AI) technologies continue to advance, protecting human autonomy and promoting ethical decision-making are essential to fostering trust and accountability. Human agency (the capacity of individuals to make informed decisions) should be actively preserved and reinforced by AI systems. This paper examines strategies for designing AI systems that uphold fundamental rights, strengthen human agency, and embed effective human oversight mechanisms. It discusses key oversight models, including Human-in-Command (HIC), Human-in-the-Loop (HITL), and Human-on-the-Loop (HOTL), and proposes a risk-based framework to guide the implementation of these mechanisms. By linking the level of AI model risk to the appropriate form of human oversight, the paper underscores the critical role of human involvement in the responsible deployment of AI, balancing technological innovation with the protection of individual values and rights. In doing so, it aims to ensure that AI technologies are used responsibly, safeguarding individual autonomy while maximizing societal benefits.", "date": "2025-10-10", "categories": []}
{"id": "2510.07889", "title": "Towards Meaningful Transparency in Civic AI Systems", "abstract": "Artificial intelligence has become a part of the provision of governmental services, from making decisions about benefits to issuing fines for parking violations. However, AI systems rarely live up to the promise of neutral optimisation, creating biased or incorrect outputs and reducing the agency of both citizens and civic workers to shape the way decisions are made. Transparency is a principle that can both help subjects understand decisions made about them and shape the processes behind those decisions. However, transparency as practiced around AI systems tends to focus on the production of technical objects that represent algorithmic aspects of decision making. These are often difficult for publics to understand, do not connect to potential for action, and do not give insight into the wider socio-material context of decision making. In this paper, we build on existing approaches that take a human-centric view on AI transparency, combined with a socio-technical systems view, to develop the concept of meaningful transparency for civic AI systems: transparencies that allow publics to engage with AI systems that affect their lives, connecting understanding with potential for action.", "date": "2025-10-09", "categories": []}
{"id": "2510.07268", "title": "On the false election between regulation and innovation. Ideas for regulation through the responsible use of artificial intelligence in research and education.[Spanish version]", "abstract": "This short essay is a reworking of the answers offered by the author at the Debate Session of the AIHUB (CSIC) and EduCaixa Summer School, organized by Marta Garcia-Matos and Lissette Lemus, and coordinated by Albert Sabater (OEIAC, UG), with the participation of Vanina Martinez-Posse (IIIA-CSIC), Eulalia Soler (Eurecat) and Pompeu Casanovas (IIIA-CSIC) on July 4th 2025. Albert Sabater posed three questions: (1) How can regulatory frameworks priori-tise the protection of fundamental rights (privacy, non-discrimination, autonomy, etc.) in the development of AI, without falling into the false dichotomy between regulation and innova-tion? (2) Given the risks of AI (bias, mass surveillance, manipulation), what examples of regu-lations or policies have demonstrated that it is possible to foster responsible innovation, putting the public interest before profitability, without giving in to competitive pressure from actors such as China or the US? (3) In a scenario where the US prioritizes flexibility, what mecha-nisms could ensure that international cooperation in AI does not become a race to the bottom in rights, but rather a global standard of accountability? The article attempts to answer these three questions and concludes with some reflections on the relevance of the answers for education and research.", "date": "2025-10-08", "categories": []}
{"id": "2508.19317", "title": "What Makes AI Applications Acceptable or Unacceptable? A Predictive Moral Framework", "abstract": "As artificial intelligence rapidly transforms society, developers and policymakers struggle to anticipate which applications will face public moral resistance. We propose that these judgments are not idiosyncratic but systematic and predictable. In a large, preregistered study (N = 587, U.S. representative sample), we used a comprehensive taxonomy of 100 AI applications spanning personal and organizational contexts-including both functional uses and the moral treatment of AI itself. In participants' collective judgment, applications ranged from highly unacceptable to fully acceptable. We found this variation was strongly predictable: five core moral qualities-perceived risk, benefit, dishonesty, unnaturalness, and reduced accountability-collectively explained over 90% of the variance in acceptability ratings. The framework demonstrated strong predictive power across all domains and successfully predicted individual-level judgments for held-out applications. These findings reveal that a structured moral psychology underlies public evaluation of new technologies, offering a powerful tool for anticipating public resistance and guiding responsible innovation in AI.", "date": "2025-08-26", "categories": []}
{"id": "2509.25286", "title": "Artificial Authority: From Machine Minds to Political Alignments. An Experimental Analysis of Democratic and Autocratic Biases in Large-Language Models", "abstract": "Political beliefs vary significantly across different countries, reflecting distinct historical, cultural, and institutional contexts. These ideologies, ranging from liberal democracies to rigid autocracies, influence human societies, as well as the digital systems that are constructed within those societies. The advent of generative artificial intelligence, particularly Large Language Models (LLMs), introduces new agents in the political space-agents trained on massive corpora that replicate and proliferate socio-political assumptions. This paper analyses whether LLMs display propensities consistent with democratic or autocratic world-views. We validate this insight through experimental tests in which we experiment with the leading LLMs developed across disparate political contexts, using several existing psychometric and political orientation measures. The analysis is based on both numerical scoring and qualitative analysis of the models' responses. Findings indicate high model-to-model variability and a strong association with the political culture of the country in which the model was developed. These findings highlight the need for more detailed examination of the socio-political dimensions embedded within AI systems.", "date": "2025-09-29", "categories": []}
{"id": "2510.03368", "title": "An Adaptive Responsible AI Governance Framework for Decentralized Organizations", "abstract": "This paper examines the assessment challenges of Responsible AI (RAI) governance efforts in globally decentralized organizations through a case study collaboration between a leading research university and a multinational enterprise. While there are many proposed frameworks for RAI, their application in complex organizational settings with distributed decision-making authority remains underexplored. Our RAI assessment, conducted across multiple business units and AI use cases, reveals four key patterns that shape RAI implementation: (1) complex interplay between group-level guidance and local interpretation, (2) challenges translating abstract principles into operational practices, (3) regional and functional variation in implementation approaches, and (4) inconsistent accountability in risk oversight. Based on these findings, we propose an Adaptive RAI Governance (ARGO) Framework that balances central coordination with local autonomy through three interdependent layers: shared foundation standards, central advisory resources, and contextual local implementation. We contribute insights from academic-industry collaboration for RAI assessments, highlighting the importance of modular governance approaches that accommodate organizational complexity while maintaining alignment with responsible AI principles. These lessons offer practical guidance for organizations navigating the transition from RAI principles to operational practice within decentralized structures.", "date": "2025-10-03", "categories": []}
{"id": "2510.03384", "title": "Implicit Values Embedded in How Humans and LLMs Complete Subjective Everyday Tasks", "abstract": "Large language models (LLMs) can underpin AI assistants that help users with everyday tasks, such as by making recommendations or performing basic computation. Despite AI assistants' promise, little is known about the implicit values these assistants display while completing subjective everyday tasks. Humans may consider values like environmentalism, charity, and diversity. To what extent do LLMs exhibit these values in completing everyday tasks? How do they compare with humans? We answer these questions by auditing how six popular LLMs complete 30 everyday tasks, comparing LLMs to each other and to 100 human crowdworkers from the US. We find LLMs often do not align with humans, nor with other LLMs, in the implicit values exhibited.", "date": "2025-10-03", "categories": []}
{"id": "2207.00108", "title": "Discrimination in machine learning algorithms", "abstract": "Machine learning algorithms are routinely used for business decisions that may directly affect individuals, for example, because a credit scoring algorithm refuses them a loan. It is then relevant from an ethical (and legal) point of view to ensure that these algorithms do not discriminate based on sensitive attributes (like sex or race), which may occur unwittingly and unknowingly by the operator and the management. Statistical tools and methods are then required to detect and eliminate such potential biases.", "date": "2022-06-30", "categories": []}
{"id": "2506.22774", "title": "Bridging Ethical Principles and Algorithmic Methods: An Alternative Approach for Assessing Trustworthiness in AI Systems", "abstract": "Artificial Intelligence (AI) technology epitomizes the complex challenges posed by human-made artifacts, particularly those widely integrated into society and exerting significant influence, highlighting potential benefits and their negative consequences. While other technologies may also pose substantial risks, AI's pervasive reach makes its societal effects especially profound. The complexity of AI systems, coupled with their remarkable capabilities, can lead to a reliance on technologies that operate beyond direct human oversight or understanding. To mitigate the risks that arise, several theoretical tools and guidelines have been developed, alongside efforts to create technological tools aimed at safeguarding Trustworthy AI. The guidelines take a more holistic view of the issue but fail to provide techniques for quantifying trustworthiness. Conversely, while technological tools are better at achieving such quantification, they lack a holistic perspective, focusing instead on specific aspects of Trustworthy AI. This paper aims to introduce an assessment method that combines the ethical components of Trustworthy AI with the algorithmic processes of PageRank and TrustRank. The goal is to establish an assessment framework that minimizes the subjectivity inherent in the self-assessment techniques prevalent in the field by introducing algorithmic criteria. The application of our approach indicates that a holistic assessment of an AI system's trustworthiness can be achieved by providing quantitative insights while considering the theoretical content of relevant guidelines.", "date": "2025-06-28", "categories": []}
{"id": "2510.02978", "title": "AI Generated Child Sexual Abuse Material -- What's the Harm?", "abstract": "The development of generative artificial intelligence (AI) tools capable of producing wholly or partially synthetic child sexual abuse material (AI CSAM) presents profound challenges for child protection, law enforcement, and societal responses to child exploitation. While some argue that the harmfulness of AI CSAM differs fundamentally from other CSAM due to a perceived absence of direct victimization, this perspective fails to account for the range of risks associated with its production and consumption. AI has been implicated in the creation of synthetic CSAM of children who have not previously been abused, the revictimization of known survivors of abuse, the facilitation of grooming, coercion and sexual extortion, and the normalization of child sexual exploitation. Additionally, AI CSAM may serve as a new or enhanced pathway into offending by lowering barriers to engagement, desensitizing users to progressively extreme content, and undermining protective factors for individuals with a sexual interest in children. This paper provides a primer on some key technologies, critically examines the harms associated with AI CSAM, and cautions against claims that it may function as a harm reduction tool, emphasizing how some appeals to harmlessness obscure its real risks and may contribute to inertia in ecosystem responses.", "date": "2025-10-03", "categories": []}
{"id": "2510.01281", "title": "An Analysis of the New EU AI Act and A Proposed Standardization Framework for Machine Learning Fairness", "abstract": "The European Union's AI Act represents a crucial step towards regulating ethical and responsible AI systems. However, we find an absence of quantifiable fairness metrics and the ambiguity in terminology, particularly the interchangeable use of the keywords transparency, explainability, and interpretability in the new EU AI Act and no reference of transparency of ethical compliance. We argue that this ambiguity creates substantial liability risk that would deter investment. Fairness transparency is strategically important. We recommend a more tailored regulatory framework to enhance the new EU AI regulation. Further-more, we propose a public system framework to assess the fairness and transparency of AI systems. Drawing from past work, we advocate for the standardization of industry best practices as a necessary addition to broad regulations to achieve the level of details required in industry, while preventing stifling innovation and investment in the AI sector. The proposals are exemplified with the case of ASR and speech synthesizers.", "date": "2025-09-30", "categories": []}
{"id": "2510.02036", "title": "The Current State of AI Bias Bounties: An Overview of Existing Programmes and Research", "abstract": "Current bias evaluation methods rarely engage with communities impacted by AI systems. Inspired by bug bounties, bias bounties have been proposed as a reward-based method that involves communities in AI bias detection by asking users of AI systems to report biases they encounter when interacting with such systems. In the absence of a state-of-the-art review, this survey aimed to identify and analyse existing AI bias bounty programmes and to present academic literature on bias bounties. Google, Google Scholar, PhilPapers, and IEEE Xplore were searched, and five bias bounty programmes, as well as five research publications, were identified. All bias bounties were organised by U.S.-based organisations as time-limited contests, with public participation in four programmes and prize pools ranging from 7,000 to 24,000 USD. The five research publications included a report on the application of bug bounties to algorithmic harms, an article addressing Twitter's bias bounty, a proposal for bias bounties as an institutional mechanism to increase AI scrutiny, a workshop discussing bias bounties from queer perspectives, and an algorithmic framework for bias bounties. We argue that reducing the technical requirements to enter bounty programmes is important to include those without coding experience. Given the limited adoption of bias bounties, future efforts should explore the transferability of the best practices from bug bounties and examine how such programmes can be designed to be sensitive to underrepresented groups while lowering adoption barriers for organisations.", "date": "2025-10-02", "categories": []}
{"id": "2510.00084", "title": "Towards a Framework for Supporting the Ethical and Regulatory Certification of AI Systems", "abstract": "Artificial Intelligence has rapidly become a cornerstone technology, significantly influencing Europe's societal and economic landscapes. However, the proliferation of AI also raises critical ethical, legal, and regulatory challenges. The CERTAIN (Certification for Ethical and Regulatory Transparency in Artificial Intelligence) project addresses these issues by developing a comprehensive framework that integrates regulatory compliance, ethical standards, and transparency into AI systems. In this position paper, we outline the methodological steps for building the core components of this framework. Specifically, we present: (i) semantic Machine Learning Operations (MLOps) for structured AI lifecycle management, (ii) ontology-driven data lineage tracking to ensure traceability and accountability, and (iii) regulatory operations (RegOps) workflows to operationalize compliance requirements. By implementing and validating its solutions across diverse pilots, CERTAIN aims to advance regulatory compliance and to promote responsible AI innovation aligned with European standards.", "date": "2025-09-30", "categories": []}
{"id": "2510.00312", "title": "Digital Domination: A Case for Republican Liberty in Artificial Intelligence", "abstract": "Artificial intelligence is set to revolutionize social and political life in unpredictable ways, raising questions about the principles that ought to guide its development and regulation. By examining digital advertising and social media algorithms, this article highlights how artificial intelligence already poses a significant threat to the republican conception of liberty -- or freedom from unaccountable power -- and thereby highlights the necessity of protecting republican liberty when integrating artificial intelligence into society. At an individual level, these algorithms can subconsciously influence behavior and thought, and those subject to this influence have limited power over the algorithms they engage. At the political level, these algorithms give technology company executives and other foreign parties the power to influence domestic political processes, such as elections; the multinational nature of algorithm-based platforms and the speed with which technology companies innovate make incumbent state institutions ineffective at holding these actors accountable. At both levels, artificial intelligence has thus created a new form of unfreedom: digital domination. By drawing on the works of Quentin Skinner, Philip Pettit, and other republican theorists, this article asserts that individuals must have mechanisms to hold algorithms (and those who develop them) accountable in order to be truly free.", "date": "2025-09-30", "categories": []}
{"id": "2502.05849", "title": "Where Fact Ends and Fairness Begins: Redefining AI Bias Evaluation through Cognitive Biases", "abstract": "Recent failures such as Google Gemini generating people of color in Nazi-era uniforms illustrate how AI outputs can be factually plausible yet socially harmful. AI models are increasingly evaluated for \"fairness,\" yet existing benchmarks often conflate two fundamentally different dimensions: factual correctness and normative fairness. A model may generate responses that are factually accurate but socially unfair, or conversely, appear fair while distorting factual reality. We argue that identifying the boundary between fact and fair is essential for meaningful fairness evaluation. We introduce Fact-or-Fair, a benchmark with (i) objective queries aligned with descriptive, fact-based judgments, and (ii) subjective queries aligned with normative, fairness-based judgments. Our queries are constructed from 19 statistics and are grounded in cognitive psychology, drawing on representativeness bias, attribution bias, and ingroup-outgroup bias to explain why models often misalign fact and fairness. Experiments across ten frontier models reveal different levels of fact-fair trade-offs. By reframing fairness evaluation, we provide both a new theoretical lens and a practical benchmark to advance the responsible model assessments. Our test suite is publicly available at https://github.com/uclanlp/Fact-or-Fair.", "date": "2025-02-09", "categories": []}
{"id": "2504.14446", "title": "Neglected Risks: The Disturbing Reality of Children's Images in Datasets and the Urgent Call for Accountability", "abstract": "Including children's images in datasets has raised ethical concerns, particularly regarding privacy, consent, data protection, and accountability. These datasets, often built by scraping publicly available images from the Internet, can expose children to risks such as exploitation, profiling, and tracking. Despite the growing recognition of these issues, approaches for addressing them remain limited. We explore the ethical implications of using children's images in AI datasets and propose a pipeline to detect and remove such images. As a use case, we built the pipeline on a Vision-Language Model under the Visual Question Answering task and tested it on the #PraCegoVer dataset. We also evaluate the pipeline on a subset of 100,000 images from the Open Images V7 dataset to assess its effectiveness in detecting and removing images of children. The pipeline serves as a baseline for future research, providing a starting point for more comprehensive tools and methodologies. While we leverage existing models trained on potentially problematic data, our goal is to expose and address this issue. We do not advocate for training or deploying such models, but instead call for urgent community reflection and action to protect children's rights. Ultimately, we aim to encourage the research community to exercise - more than an additional - care in creating new datasets and to inspire the development of tools to protect the fundamental rights of vulnerable groups, particularly children.", "date": "2025-04-20", "categories": []}
{"id": "2310.20563", "title": "Taking control: Policies to address extinction risks from advanced AI", "abstract": "This paper provides policy recommendations to reduce extinction risks from advanced artificial intelligence (AI). First, we briefly provide background information about extinction risks from AI. Second, we argue that voluntary commitments from AI companies would be an inappropriate and insufficient response. Third, we describe three policy proposals that would meaningfully address the threats from advanced AI: (1) establishing a Multinational AGI Consortium to enable democratic oversight of advanced AI (MAGIC), (2) implementing a global cap on the amount of computing power used to train an AI system (global compute cap), and (3) requiring affirmative safety evaluations to ensure that risks are kept below acceptable levels (gating critical experiments). MAGIC would be a secure, safety-focused, internationally-governed institution responsible for reducing risks from advanced AI and performing research to safely harness the benefits of AI. MAGIC would also maintain emergency response infrastructure (kill switch) to swiftly halt AI development or withdraw model deployment in the event of an AI-related emergency. The global compute cap would end the corporate race toward dangerous AI systems while enabling the vast majority of AI innovation to continue unimpeded. Gating critical experiments would ensure that companies developing powerful AI systems are required to present affirmative evidence that these models keep extinction risks below an acceptable threshold. After describing these recommendations, we propose intermediate steps that the international community could take to implement these proposals and lay the groundwork for international coordination around advanced AI.", "date": "2023-10-31", "categories": []}
{"id": "2506.01813", "title": "The Ultimate Test of Superintelligent AI Agents: Can an AI Balance Care and Control in Asymmetric Relationships?", "abstract": "This paper introduces the Shepherd Test, a new conceptual test for assessing the moral and relational dimensions of superintelligent artificial agents. The test is inspired by human interactions with animals, where ethical considerations about care, manipulation, and consumption arise in contexts of asymmetric power and self-preservation. We argue that AI crosses an important, and potentially dangerous, threshold of intelligence when it exhibits the ability to manipulate, nurture, and instrumentally use less intelligent agents, while also managing its own survival and expansion goals. This includes the ability to weigh moral trade-offs between self-interest and the well-being of subordinate agents. The Shepherd Test thus challenges traditional AI evaluation paradigms by emphasizing moral agency, hierarchical behavior, and complex decision-making under existential stakes. We argue that this shift is critical for advancing AI governance, particularly as AI systems become increasingly integrated into multi-agent environments. We conclude by identifying key research directions, including the development of simulation environments for testing moral behavior in AI, and the formalization of ethical manipulation within multi-agent systems.", "date": "2025-06-02", "categories": []}
{"id": "2509.22723", "title": "Responsible Diffusion: A Comprehensive Survey on Safety, Ethics, and Trust in Diffusion Models", "abstract": "Diffusion models (DMs) have been investigated in various domains due to their ability to generate high-quality data, thereby attracting significant attention. However, similar to traditional deep learning systems, there also exist potential threats to DMs. To provide advanced and comprehensive insights into safety, ethics, and trust in DMs, this survey comprehensively elucidates its framework, threats, and countermeasures. Each threat and its countermeasures are systematically examined and categorized to facilitate thorough analysis. Furthermore, we introduce specific examples of how DMs are used, what dangers they might bring, and ways to protect against these dangers. Finally, we discuss key lessons learned, highlight open challenges related to DM security, and outline prospective research directions in this critical field. This work aims to accelerate progress not only in the technical capabilities of generative artificial intelligence but also in the maturity and wisdom of its application.", "date": "2025-09-25", "categories": []}
{"id": "2509.24138", "title": "Analysis of Bias in Deep Learning Facial Beauty Regressors", "abstract": "Bias can be introduced to AI systems even from seemingly balanced sources, and AI facial beauty prediction is subject to ethnicity-based bias. This work sounds warnings about AI's role in shaping aesthetic norms while providing potential pathways toward equitable beauty technologies through comparative analysis of models trained on SCUT-FBP5500 and MEBeauty datasets. Employing rigorous statistical validation (Kruskal-Wallis H-tests, post hoc Dunn analyses). It is demonstrated that both models exhibit significant prediction disparities across ethnic groups $(p < 0.001)$, even when evaluated on the balanced FairFace dataset. Cross-dataset validation shows algorithmic amplification of societal beauty biases rather than mitigation based on prediction and error parity. The findings underscore the inadequacy of current AI beauty prediction approaches, with only 4.8-9.5\\% of inter-group comparisons satisfying distributional parity criteria. Mitigation strategies are proposed and discussed in detail.", "date": "2025-09-29", "categories": []}
{"id": "2401.17805", "title": "Biospheric AI", "abstract": "The dominant paradigm in AI ethics and value alignment is highly anthropocentric. The focus of these disciplines is strictly on human values which limits the depth and breadth of their insights. Recently, attempts to expand to a sentientist perspective have been initiated. We argue that neither of these outlooks is sufficient to capture the actual complexity of the biosphere and ensure that AI does not damage it. Thus, we propose a new paradigm -- Biospheric AI that assumes an ecocentric perspective. We discuss hypothetical ways in which such an AI might be designed. Moreover, we give directions for research and application of the modern AI models that would be consistent with the biospheric interests. All in all, this work attempts to take first steps towards a comprehensive program of research that focuses on the interactions between AI and the biosphere.", "date": "2024-01-31", "categories": []}
{"id": "2404.16957", "title": "Attributing Responsibility in AI-Induced Incidents: A Computational Reflective Equilibrium Framework for Accountability", "abstract": "The pervasive integration of Artificial Intelligence (AI) has introduced complex challenges in the responsibility and accountability in the event of incidents involving AI-enabled systems. The interconnectivity of these systems, ethical concerns of AI-induced incidents, coupled with uncertainties in AI technology and the absence of corresponding regulations, have made traditional responsibility attribution challenging. To this end, this work proposes a Computational Reflective Equilibrium (CRE) approach to establish a coherent and ethically acceptable responsibility attribution framework for all stakeholders. The computational approach provides a structured analysis that overcomes the limitations of conceptual approaches in dealing with dynamic and multifaceted scenarios, showcasing the framework's traceability, coherence, and adaptivity properties in the responsibility attribution process. We examine the pivotal role of the initial activation level associated with claims in equilibrium computation. Using an AI-assisted medical decision-support system as a case study, we illustrate how different initializations lead to diverse responsibility distributions. The framework offers valuable insights into accountability in AI-induced incidents, facilitating the development of a sustainable and resilient system through continuous monitoring, revision, and reflection.", "date": "2024-04-25", "categories": []}
{"id": "2412.05130", "title": "Technology as uncharted territory: Contextual integrity and the notion of AI as new ethical ground", "abstract": "Recent research illustrates how AI can be developed and deployed in a manner detached from the concrete social context of application. By abstracting from the contexts of AI application, practitioners also disengage from the distinct normative structures that govern them. Building upon Helen Nissenbaum's framework of contextual integrity, I illustrate how disregard for contextual norms can threaten the integrity of a context with often decisive ethical implications. I argue that efforts to promote responsible and ethical AI can inadvertently contribute to and seemingly legitimize this disregard for established contextual norms. Echoing a persistent undercurrent in technology ethics of understanding emerging technologies as uncharted moral territory, certain approaches to AI ethics can promote a notion of AI as a novel and distinct realm for ethical deliberation, norm setting, and virtue cultivation. This narrative of AI as new ethical ground, however, can come at the expense of practitioners, policymakers and ethicists engaging with already established norms and virtues that were gradually cultivated to promote successful and responsible practice within concrete social contexts. In response, I question the current narrow prioritization in AI ethics of moral innovation over moral preservation. Engaging also with emerging foundation models, I advocate for a moderately conservative approach to the ethics of AI that prioritizes the responsible and considered integration of AI within established social contexts and their respective normative structures.", "date": "2024-12-06", "categories": []}
{"id": "2509.20394", "title": "Blueprints of Trust: AI System Cards for End to End Transparency and Governance", "abstract": "This paper introduces the Hazard-Aware System Card (HASC), a novel framework designed to enhance transparency and accountability in the development and deployment of AI systems. The HASC builds upon existing model card and system card concepts by integrating a comprehensive, dynamic record of an AI system's security and safety posture. The framework proposes a standardized system of identifiers, including a novel AI Safety Hazard (ASH) ID, to complement existing security identifiers like CVEs, allowing for clear and consistent communication of fixed flaws. By providing a single, accessible source of truth, the HASC empowers developers and stakeholders to make more informed decisions about AI system safety throughout its lifecycle. Ultimately, we also compare our proposed AI system cards with the ISO/IEC 42001:2023 standard and discuss how they can be used to complement each other, providing greater transparency and accountability for AI systems.", "date": "2025-09-23", "categories": []}
{"id": "2509.21075", "title": "Communication Bias in Large Language Models: A Regulatory Perspective", "abstract": "Large language models (LLMs) are increasingly central to many applications, raising concerns about bias, fairness, and regulatory compliance. This paper reviews risks of biased outputs and their societal impact, focusing on frameworks like the EU's AI Act and the Digital Services Act. We argue that beyond constant regulation, stronger attention to competition and design governance is needed to ensure fair, trustworthy AI. This is a preprint of the Communications of the ACM article of the same title.", "date": "2025-09-25", "categories": []}
{"id": "2407.02425", "title": "Reinforcement Learning and Machine ethics:a systematic review", "abstract": "Machine ethics is the field that studies how ethical behaviour can be accomplished by autonomous systems. While there exist some systematic reviews aiming to consolidate the state of the art in machine ethics prior to 2020, these tend to not include work that uses reinforcement learning agents as entities whose ethical behaviour is to be achieved. The reason for this is that only in the last years we have witnessed an increase in machine ethics studies within reinforcement learning. We present here a systematic review of reinforcement learning for machine ethics and machine ethics within reinforcement learning. Additionally, we highlight trends in terms of ethics specifications, components and frameworks of reinforcement learning, and environments used to result in ethical behaviour. Our systematic review aims to consolidate the work in machine ethics and reinforcement learning thus completing the gap in the state of the art machine ethics landscape", "date": "2024-07-02", "categories": []}
{"id": "2508.08345", "title": "Do AI Companies Make Good on Voluntary Commitments to the White House?", "abstract": "Voluntary commitments are central to international AI governance, as demonstrated by recent voluntary guidelines from the White House to the G7, from Bletchley Park to Seoul. How do major AI companies make good on their commitments? We score companies based on their publicly disclosed behavior by developing a detailed rubric based on their eight voluntary commitments to the White House in 2023. We find significant heterogeneity: while the highest-scoring company (OpenAI) scores a 83% overall on our rubric, the average score across all companies is just 53%. The companies demonstrate systemically poor performance for their commitment to model weight security with an average score of 17%: 11 of the 16 companies receive 0% for this commitment. Our analysis highlights a clear structural shortcoming that future AI governance initiatives should correct: when companies make public commitments, they should proactively disclose how they meet their commitments to provide accountability, and these disclosures should be verifiable. To advance policymaking on corporate AI governance, we provide three directed recommendations that address underspecified commitments, the role of complex AI supply chains, and public transparency that could be applied towards AI governance initiatives worldwide.", "date": "2025-08-11", "categories": []}
{"id": "2509.18233", "title": "Perceptions of AI Across Sectors: A Comparative Review of Public Attitudes", "abstract": "This paper offers a domain-mediated comparative review of 251 studies on public attitudes toward AI, published between 2011 and 2025. Drawing on a systematic literature review, we analyse how different factors including perceived benefits and concerns (or risks) shape public acceptance of - or resistance to - artificial intelligence across domains and use-cases, including healthcare, education, security, public administration, generative AI, and autonomous vehicles. The analysis highlights recurring patterns in individual, contextual, and technical factors influencing perception, while also tracing variations in institutional trust, perceived fairness, and ethical concerns. We show that the public perception in AI is shaped not only by technical design or performance but also by sector-specific considerations as well as imaginaries, cultural narratives, and historical legacies. This comparative approach offers a foundation for developing more tailored and context-sensitive strategies for responsible AI governance.", "date": "2025-09-22", "categories": []}
{"id": "2505.15108", "title": "A Risk Ontology for Evaluating AI-Powered Psychotherapy Virtual Agents", "abstract": "The proliferation of Large Language Models (LLMs) and Intelligent Virtual Agents acting as psychotherapists presents significant opportunities for expanding mental healthcare access. However, their deployment has also been linked to serious adverse outcomes, including user harm and suicide, facilitated by a lack of standardized evaluation methodologies capable of capturing the nuanced risks of therapeutic interaction. Current evaluation techniques lack the sensitivity to detect subtle changes in patient cognition and behavior during therapy sessions that may lead to subsequent decompensation. We introduce a novel risk ontology specifically designed for the systematic evaluation of conversational AI psychotherapists. Developed through an iterative process including review of the psychotherapy risk literature, qualitative interviews with clinical and legal experts, and alignment with established clinical criteria (e.g., DSM-5) and existing assessment tools (e.g., NEQ, UE-ATR), the ontology aims to provide a structured approach to identifying and assessing user/patient harms. We provide a high-level overview of this ontology, detailing its grounding, and discuss potential use cases. We discuss four use cases in detail: monitoring real user interactions, evaluation with simulated patients, benchmarking and comparative analysis, and identifying unexpected outcomes. The proposed ontology offers a foundational step towards establishing safer and more responsible innovation in the domain of AI-driven mental health support.", "date": "2025-05-21", "categories": []}
{"id": "2509.16347", "title": "QUINTA: Reflexive Sensibility For Responsible AI Research and Data-Driven Processes", "abstract": "As the field of artificial intelligence (AI) and machine learning (ML) continues to prioritize fairness and the concern for historically marginalized communities, the importance of intersectionality in AI research has gained significant recognition. However, few studies provide practical guidance on how researchers can effectively incorporate intersectionality into critical praxis. In response, this paper presents a comprehensive framework grounded in critical reflexivity as intersectional praxis. Operationalizing intersectionality within the AI/DS (Artificial Intelligence/Data Science) pipeline, Quantitative Intersectional Data (QUINTA) is introduced as a methodological paradigm that challenges conventional and superficial research habits, particularly in data-centric processes, to identify and mitigate negative impacts such as the inadvertent marginalization caused by these practices. The framework centers researcher reflexivity to call attention to the AI researchers' power in creating and analyzing AI/DS artifacts through data-centric approaches. To illustrate the effectiveness of QUINTA, we provide a reflexive AI/DS researcher demonstration utilizing the \\#metoo movement as a case study. Note: This paper was accepted as a poster presentation at Equity and Access in Algorithms, Mechanisms, and Optimization (EAAMO) Conference in 2023.", "date": "2025-09-19", "categories": []}
{"id": "2509.17334", "title": "Explainability matters: The effect of liability rules on the healthcare sector", "abstract": "Explainability, the capability of an artificial intelligence system (AIS) to explain its outcomes in a manner that is comprehensible to human beings at an acceptable level, has been deemed essential for critical sectors, such as healthcare. Is it really the case? In this perspective, we consider two extreme cases, ``Oracle'' (without explainability) versus ``AI Colleague'' (with explainability) for a thorough analysis. We discuss how the level of automation and explainability of AIS can affect the determination of liability among the medical practitioner/facility and manufacturer of AIS. We argue that explainability plays a crucial role in setting a responsibility framework in healthcare, from a legal standpoint, to shape the behavior of all involved parties and mitigate the risk of potential defensive medicine practices.", "date": "2025-09-22", "categories": []}
{"id": "2504.01029", "title": "Who is Responsible When AI Fails? Mapping Causes, Entities, and Consequences of AI Privacy and Ethical Incidents", "abstract": "The rapid growth of artificial intelligence (AI) technologies has raised major privacy and ethical concerns. However, existing AI incident taxonomies and guidelines lack grounding in real-world cases, limiting their effectiveness for prevention and mitigation. We analyzed 202 real-world AI privacy and ethical incidents to develop a taxonomy that classifies them across AI lifecycle stages and captures contributing factors, including causes, responsible entities, sources of disclosure, and impacts. Our findings reveal widespread harms from poor organizational decisions and legal non-compliance, limited corrective interventions, and rare reporting from AI developers and adopting entities. Our taxonomy offers a structured approach for systematic incident reporting and emphasizes the weaknesses of current AI governance frameworks. Our findings provide actionable guidance for policymakers and practitioners to strengthen user protections, develop targeted AI policies, enhance reporting practices, and foster responsible AI governance and innovation, especially in contexts such as social media and child protection.", "date": "2025-03-28", "categories": []}
{"id": "2506.12556", "title": "Algorithmic Fairness: Not a Purely Technical but Socio-Technical Property", "abstract": "The rapid trend of deploying artificial intelligence (AI) and machine learning (ML) systems in socially consequential domains has raised growing concerns about their trustworthiness, including potential discriminatory behaviours. Research in algorithmic fairness has generated a proliferation of mathematical definitions and metrics, yet persistent misconceptions and limitations -- both within and beyond the fairness community -- limit their effectiveness, such as an unreached consensus on its understanding, prevailing measures primarily tailored to binary group settings, and superficial handling for intersectional contexts. Here we critically remark on these misconceptions and argue that fairness cannot be reduced to purely technical constraints on models; we also examine the limitations of existing fairness measures through conceptual analysis and empirical illustrations, showing their limited applicability in the face of complex real-world scenarios, challenging prevailing views on the incompatibility between accuracy and fairness as well as that among fairness measures themselves, and outlining three worth-considering principles in the design of fairness measures. We believe these findings will help bridge the gap between technical formalisation and social realities and meet the challenges of real-world AI/ML deployment.", "date": "2025-06-14", "categories": []}
{"id": "2508.03666", "title": "Deciding how to respond: A deliberative framework to guide policymaker responses to AI systems", "abstract": "The discourse on responsible artificial intelligence (AI) regulation is understandably dominated by risk-focused assessments and analyses. This approach reflects the fundamental uncertainty policymakers face when determining appropriate responses to current, emerging and novel AI systems. In this article, we argue that by operationalising the concept of freedom - the philosophical counterpart to responsibility - a complementary approach centred on the potential societal benefits of AI systems can be developed. The result is a discursive framework grounded in freedom as capability and freedom as opportunity, which represent the two main intellectual traditions of interpreting freedom. We contend that the complexity, ambiguity and contestation involved in regulating AI systems make a deliberative paradigm more useful than the conventional technical one. The resulting framework is structured around coordinative, communicative and decision spaces, each with sequential focal points and associated outputs.", "date": "2025-08-05", "categories": []}
{"id": "2509.13365", "title": "The Provenance Problem: LLMs and the Breakdown of Citation Norms", "abstract": "The increasing use of generative AI in scientific writing raises urgent questions about attribution and intellectual credit. When a researcher employs ChatGPT to draft a manuscript, the resulting text may echo ideas from sources the author has never encountered. If an AI system reproduces insights from, for example, an obscure 1975 paper without citation, does this constitute plagiarism? We argue that such cases exemplify the 'provenance problem': a systematic breakdown in the chain of scholarly credit. Unlike conventional plagiarism, this phenomenon does not involve intent to deceive (researchers may disclose AI use and act in good faith) yet still benefit from the uncredited intellectual contributions of others. This dynamic creates a novel category of attributional harm that current ethical and professional frameworks fail to address. As generative AI becomes embedded across disciplines, the risk that significant ideas will circulate without recognition threatens both the reputational economy of science and the demands of epistemic justice. This Perspective analyzes how AI challenges established norms of authorship, introduces conceptual tools for understanding the provenance problem, and proposes strategies to preserve integrity and fairness in scholarly communication.", "date": "2025-09-15", "categories": []}
{"id": "2509.06176", "title": "AI Governance in Higher Education: A course design exploring regulatory, ethical and practical considerations", "abstract": "As artificial intelligence (AI) systems permeate critical sectors, the need for professionals who can address ethical, legal and governance challenges has become urgent. Current AI ethics education remains fragmented, often siloed by discipline and disconnected from practice. This paper synthesizes literature and regulatory developments to propose a modular, interdisciplinary curriculum that integrates technical foundations with ethics, law and policy. We highlight recurring operational failures in AI - bias, misspecified objectives, generalization errors, misuse and governance breakdowns - and link them to pedagogical strategies for teaching AI governance. Drawing on perspectives from the EU, China and international frameworks, we outline a semester plan that emphasizes integrated ethics, stakeholder engagement and experiential learning. The curriculum aims to prepare students to diagnose risks, navigate regulation and engage diverse stakeholders, fostering adaptive and ethically grounded professionals for responsible AI governance.", "date": "2025-09-07", "categories": []}
{"id": "2504.17044", "title": "Approaches to Responsible Governance of GenAI in Organizations", "abstract": "PEER-REVIEWED AND ACCEPTED IN IEEE- ISTAS 2025\n  The rapid evolution of Generative AI (GenAI) has introduced unprecedented opportunities while presenting complex challenges around ethics, accountability, and societal impact. This paper draws on a literature review, established governance frameworks, and industry roundtable discussions to identify core principles for integrating responsible GenAI governance into diverse organizational structures. Our objective is to provide actionable recommendations for a balanced, risk-based governance approach that enables both innovation and oversight. Findings emphasize the need for adaptable risk assessment tools, continuous monitoring practices, and cross-sector collaboration to establish trustworthy GenAI. These insights provide a structured foundation and Responsible GenAI Guide (ResAI) for organizations to align GenAI initiatives with ethical, legal, and operational best practices.", "date": "2025-04-23", "categories": []}
{"id": "2509.00575", "title": "Can AI be Auditable?", "abstract": "Auditability is defined as the capacity of AI systems to be independently assessed for compliance with ethical, legal, and technical standards throughout their lifecycle. The chapter explores how auditability is being formalized through emerging regulatory frameworks, such as the EU AI Act, which mandate documentation, risk assessments, and governance structures. It analyzes the diverse challenges facing AI auditability, including technical opacity, inconsistent documentation practices, lack of standardized audit tools and metrics, and conflicting principles within existing responsible AI frameworks. The discussion highlights the need for clear guidelines, harmonized international regulations, and robust socio-technical methodologies to operationalize auditability at scale. The chapter concludes by emphasizing the importance of multi-stakeholder collaboration and auditor empowerment in building an effective AI audit ecosystem. It argues that auditability must be embedded in AI development practices and governance infrastructures to ensure that AI systems are not only functional but also ethically and legally aligned.", "date": "2025-08-30", "categories": []}
{"id": "2509.10653", "title": "SCOR: A Framework for Responsible AI Innovation in Digital Ecosystems", "abstract": "AI-driven digital ecosystems span diverse stakeholders including technology firms, regulators, accelerators and civil society, yet often lack cohesive ethical governance. This paper proposes a four-pillar framework (SCOR) to embed accountability, fairness, and inclusivity across such multi-actor networks. Leveraging a design science approach, we develop a Shared Ethical Charter(S), structured Co-Design and Stakeholder Engagement protocols(C), a system of Continuous Oversight and Learning(O), and Adaptive Regulatory Alignment strategies(R). Each component includes practical guidance, from lite modules for resource-constrained start-ups to in-depth auditing systems for larger consortia. Through illustrative vignettes in healthcare, finance, and smart city contexts, we demonstrate how the framework can harmonize organizational culture, leadership incentives, and cross-jurisdictional compliance. Our mixed-method KPI design further ensures that quantitative targets are complemented by qualitative assessments of user trust and cultural change. By uniting ethical principles with scalable operational structures, this paper offers a replicable pathway toward responsible AI innovation in complex digital ecosystems.", "date": "2025-09-12", "categories": []}
{"id": "2509.11332", "title": "A five-layer framework for AI governance: integrating regulation, standards, and certification", "abstract": "Purpose: The governance of artificial iintelligence (AI) systems requires a structured approach that connects high-level regulatory principles with practical implementation. Existing frameworks lack clarity on how regulations translate into conformity mechanisms, leading to gaps in compliance and enforcement. This paper addresses this critical gap in AI governance.\n  Methodology/Approach: A five-layer AI governance framework is proposed, spanning from broad regulatory mandates to specific standards, assessment methodologies, and certification processes. By narrowing its scope through progressively focused layers, the framework provides a structured pathway to meet technical, regulatory, and ethical requirements. Its applicability is validated through two case studies on AI fairness and AI incident reporting.\n  Findings: The case studies demonstrate the framework's ability to identify gaps in legal mandates, standardization, and implementation. It adapts to both global and region-specific AI governance needs, mapping regulatory mandates with practical applications to improve compliance and risk management.\n  Practical Implications - By offering a clear and actionable roadmap, this work contributes to global AI governance by equipping policymakers, regulators, and industry stakeholders with a model to enhance compliance and risk management.\n  Social Implications: The framework supports the development of policies that build public trust and promote the ethical use of AI for the benefit of society.\n  Originality/Value: This study proposes a five-layer AI governance framework that bridges high-level regulatory mandates and implementation guidelines. Validated through case studies on AI fairness and incident reporting, it identifies gaps such as missing standardized assessment procedures and reporting mechanisms, providing a structured foundation for targeted governance measures.", "date": "2025-09-14", "categories": []}
{"id": "2509.11648", "title": "EthicsMH: A Pilot Benchmark for Ethical Reasoning in Mental Health AI", "abstract": "The deployment of large language models (LLMs) in mental health and other sensitive domains raises urgent questions about ethical reasoning, fairness, and responsible alignment. Yet, existing benchmarks for moral and clinical decision-making do not adequately capture the unique ethical dilemmas encountered in mental health practice, where confidentiality, autonomy, beneficence, and bias frequently intersect. To address this gap, we introduce Ethical Reasoning in Mental Health (EthicsMH), a pilot dataset of 125 scenarios designed to evaluate how AI systems navigate ethically charged situations in therapeutic and psychiatric contexts. Each scenario is enriched with structured fields, including multiple decision options, expert-aligned reasoning, expected model behavior, real-world impact, and multi-stakeholder viewpoints. This structure enables evaluation not only of decision accuracy but also of explanation quality and alignment with professional norms. Although modest in scale and developed with model-assisted generation, EthicsMH establishes a task framework that bridges AI ethics and mental health decision-making. By releasing this dataset, we aim to provide a seed resource that can be expanded through community and expert contributions, fostering the development of AI systems capable of responsibly handling some of society's most delicate decisions.", "date": "2025-09-15", "categories": []}
{"id": "2505.02846", "title": "The Precautionary Principle and the Innovation Principle: Incompatible Guides for AI Innovation Governance?", "abstract": "In policy debates concerning the governance and regulation of Artificial Intelligence (AI), both the Precautionary Principle (PP) and the Innovation Principle (IP) are advocated by their respective interest groups. Do these principles offer wholly incompatible and contradictory guidance? Does one necessarily negate the other? I argue here that provided attention is restricted to weak-form PP and IP, the answer to both of these questions is \"No.\" The essence of these weak formulations is the requirement to fully account for type-I error costs arising from erroneously preventing the innovation's diffusion through society (i.e. mistaken regulatory red-lighting) as well as the type-II error costs arising from erroneously allowing the innovation to diffuse through society (i.e. mistaken regulatory green-lighting). Within the Signal Detection Theory (SDT) model developed here, weak-PP red-light (weak-IP green-light) determinations are optimal for sufficiently small (large) ratios of expected type-I to type-II error costs. For intermediate expected cost ratios, an amber-light 'wait-and-monitor' policy is optimal. Regulatory sandbox instruments allow AI testing and experimentation to take place within a structured environment of limited duration and societal scale, whereby the expected cost ratio falls within the 'wait-and-monitor' range. Through sandboxing regulators and innovating firms learn more about the expected cost ratio, and what respective adaptations -- of regulation, of technical solution, of business model, or combination thereof, if any -- are needed to keep the ratio out of the weak-PP red-light zone. Nevertheless AI foundation models are ill-suited for regulatory sandboxing as their general-purpose nature precludes credible identification of misclassification costs.", "date": "2025-05-01", "categories": []}
{"id": "2509.10289", "title": "We Need a New Ethics for a World of AI Agents", "abstract": "The deployment of capable AI agents raises fresh questions about safety, human-machine relationships and social coordination. We argue for greater engagement by scientists, scholars, engineers and policymakers with the implications of a world increasingly populated by AI agents. We explore key challenges that must be addressed to ensure that interactions between humans and agents, and among agents themselves, remain broadly beneficial.", "date": "2025-09-12", "categories": []}
{"id": "2509.10297", "title": "The Morality of Probability: How Implicit Moral Biases in LLMs May Shape the Future of Human-AI Symbiosis", "abstract": "Artificial intelligence (AI) is advancing at a pace that raises urgent questions about how to align machine decision-making with human moral values. This working paper investigates how leading AI systems prioritize moral outcomes and what this reveals about the prospects for human-AI symbiosis. We address two central questions: (1) What moral values do state-of-the-art large language models (LLMs) implicitly favour when confronted with dilemmas? (2) How do differences in model architecture, cultural origin, and explainability affect these moral preferences? To explore these questions, we conduct a quantitative experiment with six LLMs, ranking and scoring outcomes across 18 dilemmas representing five moral frameworks. Our findings uncover strikingly consistent value biases. Across all models, Care and Virtue values outcomes were rated most moral, while libertarian choices were consistently penalized. Reasoning-enabled models exhibited greater sensitivity to context and provided richer explanations, whereas non-reasoning models produced more uniform but opaque judgments. This research makes three contributions: (i) Empirically, it delivers a large-scale comparison of moral reasoning across culturally distinct LLMs; (ii) Theoretically, it links probabilistic model behaviour with underlying value encodings; (iii) Practically, it highlights the need for explainability and cultural awareness as critical design principles to guide AI toward a transparent, aligned, and symbiotic future.", "date": "2025-09-12", "categories": []}
{"id": "2505.19317", "title": "Effort-aware Fairness: Incorporating a Philosophy-informed, Human-centered Notion of Effort into Algorithmic Fairness Metrics", "abstract": "Although popularized AI fairness metrics, e.g., demographic parity, have uncovered bias in AI-assisted decision-making outcomes, they do not consider how much effort one has spent to get to where one is today in the input feature space. However, the notion of effort is important in how Philosophy and humans understand fairness. We propose a philosophy-informed approach to conceptualize and evaluate Effort-aware Fairness (EaF), grounded in the concept of Force, which represents the temporal trajectory of predictive features coupled with inertia. Besides theoretical formulation, our empirical contributions include: (1) a pre-registered human subjects experiment, which shows that for both stages of the (individual) fairness evaluation process, people consider the temporal trajectory of a predictive feature more than its aggregate value; (2) pipelines to compute Effort-aware Individual/Group Fairness in the criminal justice and personal finance contexts. Our work may enable AI model auditors to uncover and potentially correct unfair decisions against individuals who have spent significant efforts to improve but are still stuck with systemic disadvantages outside their control.", "date": "2025-05-25", "categories": []}
{"id": "2509.08835", "title": "Deep opacity and AI: A threat to XAI and to privacy protection mechanisms", "abstract": "It is known that big data analytics and AI pose a threat to privacy, and that some of this is due to some kind of \"black box problem\" in AI. I explain how this becomes a problem in the context of justification for judgments and actions. Furthermore, I suggest distinguishing three kinds of opacity: 1) the subjects do not know what the system does (\"shallow opacity\"), 2) the analysts do not know what the system does (\"standard black box opacity\"), or 3) the analysts cannot possibly know what the system might do (\"deep opacity\"). If the agents, data subjects as well as analytics experts, operate under opacity, then these agents cannot provide justifications for judgments that are necessary to protect privacy, e.g., they cannot give \"informed consent\", or guarantee \"anonymity\". It follows from these points that agents in big data analytics and AI often cannot make the judgments needed to protect privacy. So I conclude that big data analytics makes the privacy problems worse and the remedies less effective. As a positive note, I provide a brief outlook on technical ways to handle this situation.", "date": "2025-08-30", "categories": []}
{"id": "2509.08494", "title": "HumanAgencyBench: Scalable Evaluation of Human Agency Support in AI Assistants", "abstract": "As humans delegate more tasks and decisions to artificial intelligence (AI), we risk losing control of our individual and collective futures. Relatively simple algorithmic systems already steer human decision-making, such as social media feed algorithms that lead people to unintentionally and absent-mindedly scroll through engagement-optimized content. In this paper, we develop the idea of human agency by integrating philosophical and scientific theories of agency with AI-assisted evaluation methods: using large language models (LLMs) to simulate and validate user queries and to evaluate AI responses. We develop HumanAgencyBench (HAB), a scalable and adaptive benchmark with six dimensions of human agency based on typical AI use cases. HAB measures the tendency of an AI assistant or agent to Ask Clarifying Questions, Avoid Value Manipulation, Correct Misinformation, Defer Important Decisions, Encourage Learning, and Maintain Social Boundaries. We find low-to-moderate agency support in contemporary LLM-based assistants and substantial variation across system developers and dimensions. For example, while Anthropic LLMs most support human agency overall, they are the least supportive LLMs in terms of Avoid Value Manipulation. Agency support does not appear to consistently result from increasing LLM capabilities or instruction-following behavior (e.g., RLHF), and we encourage a shift towards more robust safety and alignment targets.", "date": "2025-09-10", "categories": []}
{"id": "2509.07050", "title": "Automated Evaluation of Gender Bias Across 13 Large Multimodal Models", "abstract": "Large multimodal models (LMMs) have revolutionized text-to-image generation, but they risk perpetuating the harmful social biases in their training data. Prior work has identified gender bias in these models, but methodological limitations prevented large-scale, comparable, cross-model analysis. To address this gap, we introduce the Aymara Image Fairness Evaluation, a benchmark for assessing social bias in AI-generated images. We test 13 commercially available LMMs using 75 procedurally-generated, gender-neutral prompts to generate people in stereotypically-male, stereotypically-female, and non-stereotypical professions. We then use a validated LLM-as-a-judge system to score the 965 resulting images for gender representation. Our results reveal (p < .001 for all): 1) LMMs systematically not only reproduce but actually amplify occupational gender stereotypes relative to real-world labor data, generating men in 93.0% of images for male-stereotyped professions but only 22.5% for female-stereotyped professions; 2) Models exhibit a strong default-male bias, generating men in 68.3% of the time for non-stereotyped professions; and 3) The extent of bias varies dramatically across models, with overall male representation ranging from 46.7% to 73.3%. Notably, the top-performing model de-amplified gender stereotypes and approached gender parity, achieving the highest fairness scores. This variation suggests high bias is not an inevitable outcome but a consequence of design choices. Our work provides the most comprehensive cross-model benchmark of gender bias to date and underscores the necessity of standardized, automated evaluation tools for promoting accountability and fairness in AI development.", "date": "2025-09-08", "categories": []}
{"id": "2509.05382", "title": "User Privacy and Large Language Models: An Analysis of Frontier Developers' Privacy Policies", "abstract": "Hundreds of millions of people now regularly interact with large language models via chatbots. Model developers are eager to acquire new sources of high-quality training data as they race to improve model capabilities and win market share. This paper analyzes the privacy policies of six U.S. frontier AI developers to understand how they use their users' chats to train models. Drawing primarily on the California Consumer Privacy Act, we develop a novel qualitative coding schema that we apply to each developer's relevant privacy policies to compare data collection and use practices across the six companies. We find that all six developers appear to employ their users' chat data to train and improve their models by default, and that some retain this data indefinitely. Developers may collect and train on personal information disclosed in chats, including sensitive information such as biometric and health data, as well as files uploaded by users. Four of the six companies we examined appear to include children's chat data for model training, as well as customer data from other products. On the whole, developers' privacy policies often lack essential information about their practices, highlighting the need for greater transparency and accountability. We address the implications of users' lack of consent for the use of their chat data for model training, data security issues arising from indefinite chat data retention, and training on children's chat data. We conclude by providing recommendations to policymakers and developers to address the data privacy challenges posed by LLM-powered chatbots.", "date": "2025-09-05", "categories": []}
{"id": "2509.05390", "title": "Authorship Without Writing: Large Language Models and the Senior Author Analogy", "abstract": "The use of large language models (LLMs) in bioethical, scientific, and medical writing remains controversial. While there is broad agreement in some circles that LLMs cannot count as authors, there is no consensus about whether and how humans using LLMs can count as authors. In many fields, authorship is distributed among large teams of researchers, some of whom, including paradigmatic senior authors who guide and determine the scope of a project and ultimately vouch for its integrity, may not write a single word. In this paper, we argue that LLM use (under specific conditions) is analogous to a form of senior authorship. On this view, the use of LLMs, even to generate complete drafts of research papers, can be considered a legitimate form of authorship according to the accepted criteria in many fields. We conclude that either such use should be recognized as legitimate, or current criteria for authorship require fundamental revision. AI use declaration: GPT-5 was used to help format Box 1. AI was not used for any other part of the preparation or writing of this manuscript.", "date": "2025-09-05", "categories": []}
{"id": "2509.04505", "title": "The Ethical Compass of the Machine: Evaluating Large Language Models for Decision Support in Construction Project Management", "abstract": "The integration of Artificial Intelligence (AI) into construction project management (CPM) is accelerating, with Large Language Models (LLMs) emerging as accessible decision-support tools. This study aims to critically evaluate the ethical viability and reliability of LLMs when applied to the ethically sensitive, high-risk decision-making contexts inherent in CPM. A mixed-methods research design was employed, involving the quantitative performance testing of two leading LLMs against twelve real-world ethical scenarios using a novel Ethical Decision Support Assessment Checklist (EDSAC), and qualitative analysis of semi-structured interviews with 12 industry experts to capture professional perceptions. The findings reveal that while LLMs demonstrate adequate performance in structured domains such as legal compliance, they exhibit significant deficiencies in handling contextual nuance, ensuring accountability, and providing transparent reasoning. Stakeholders expressed considerable reservations regarding the autonomous use of AI for ethical judgments, strongly advocating for robust human-in-the-loop oversight. To our knowledge, this is one of the first studies to empirically test the ethical reasoning of LLMs within the construction domain. It introduces the EDSAC framework as a replicable methodology and provides actionable recommendations, emphasising that LLMs are currently best positioned as decision-support aids rather than autonomous ethical agents.", "date": "2025-09-02", "categories": []}
{"id": "2509.04515", "title": "Mitigation of Gender and Ethnicity Bias in AI-Generated Stories through Model Explanations", "abstract": "Language models have been shown to propagate social bias through their output, particularly in the representation of gender and ethnicity. This paper investigates gender and ethnicity biases in AI-generated occupational stories. Representation biases are measured before and after applying our proposed mitigation strategy, Bias Analysis and Mitigation through Explanation (BAME), revealing improvements in demographic representation ranging from 2% to 20%. BAME leverages model-generated explanations to inform targeted prompt engineering, effectively reducing biases without modifying model parameters. By analyzing stories generated across 25 occupational groups, three large language models (Claude 3.5 Sonnet, Llama 3.1 70B Instruct, and GPT-4 Turbo), and multiple demographic dimensions, we identify persistent patterns of overrepresentation and underrepresentation linked to training data stereotypes. Our findings demonstrate that guiding models with their own internal reasoning mechanisms can significantly enhance demographic parity, thereby contributing to the development of more transparent generative AI systems.", "date": "2025-09-03", "categories": []}
{"id": "2508.08193", "title": "Street-Level AI: Are Large Language Models Ready for Real-World Judgments?", "abstract": "A surge of recent work explores the ethical and societal implications of large-scale AI models that make \"moral\" judgments. Much of this literature focuses either on alignment with human judgments through various thought experiments or on the group fairness implications of AI judgments. However, the most immediate and likely use of AI is to help or fully replace the so-called street-level bureaucrats, the individuals deciding to allocate scarce social resources or approve benefits. There is a rich history underlying how principles of local justice determine how society decides on prioritization mechanisms in such domains. In this paper, we examine how well LLM judgments align with human judgments, as well as with socially and politically determined vulnerability scoring systems currently used in the domain of homelessness resource allocation. Crucially, we use real data on those needing services (maintaining strict confidentiality by only using local large models) to perform our analyses. We find that LLM prioritizations are extremely inconsistent in several ways: internally on different runs, between different LLMs, and between LLMs and the vulnerability scoring systems. At the same time, LLMs demonstrate qualitative consistency with lay human judgments in pairwise testing. Findings call into question the readiness of current generation AI systems for naive integration in high-stakes societal decision-making.", "date": "2025-08-11", "categories": []}
{"id": "2509.00117", "title": "Embodied AI: Emerging Risks and Opportunities for Policy Action", "abstract": "The field of embodied AI (EAI) is rapidly advancing. Unlike virtual AI, EAI systems can exist in, learn from, reason about, and act in the physical world. With recent advances in AI models and hardware, EAI systems are becoming increasingly capable across wider operational domains. While EAI systems can offer many benefits, they also pose significant risks, including physical harm from malicious use, mass surveillance, as well as economic and societal disruption. These risks require urgent attention from policymakers, as existing policies governing industrial robots and autonomous vehicles are insufficient to address the full range of concerns EAI systems present. To help address this issue, this paper makes three contributions. First, we provide a taxonomy of the physical, informational, economic, and social risks EAI systems pose. Second, we analyze policies in the US, EU, and UK to assess how existing frameworks address these risks and to identify critical gaps. We conclude by offering policy recommendations for the safe and beneficial deployment of EAI systems, such as mandatory testing and certification schemes, clarified liability frameworks, and strategies to manage EAI's potentially transformative economic and societal impacts.", "date": "2025-08-28", "categories": []}
{"id": "2509.02624", "title": "Who Owns The Robot?: Four Ethical and Socio-technical Questions about Wellbeing Robots in the Real World through Community Engagement", "abstract": "Recent studies indicate that robotic coaches can play a crucial role in promoting wellbeing. However, the real-world deployment of wellbeing robots raises numerous ethical and socio-technical questions and concerns. To explore these questions, we undertake a community-centered investigation to examine three different communities' perspectives on using robotic wellbeing coaches in real-world environments. We frame our work as an anticipatory ethical investigation, which we undertake to better inform the development of robotic technologies with communities' opinions, with the ultimate goal of aligning robot development with public interest. We conducted workshops with three communities who are under-represented in robotics development: 1) members of the public at a science festival, 2) women computer scientists at a conference, and 3) humanities researchers interested in history and philosophy of science. In the workshops, we collected qualitative data using the Social Robot Co-Design Canvas on Ethics. We analysed the collected qualitative data with Thematic Analysis, informed by notes taken during workshops. Through our analysis, we identify four themes regarding key ethical and socio-technical questions about the real-world use of wellbeing robots. We group participants' insights and discussions around these broad thematic questions, discuss them in light of state-of-the-art literature, and highlight areas for future investigation. Finally, we provide the four questions as a broad framework that roboticists can and should use during robotic development and deployment, in order to reflect on the ethics and socio-technical dimensions of their robotic applications, and to engage in dialogue with communities of robot users. The four questions are: 1) Is the robot safe and how can we know that?, 2) Who is the robot built for and with?, 3) Who owns the robot and the data?, and 4) Why a robot?.", "date": "2025-09-01", "categories": []}
{"id": "2509.03286", "title": "Accountability Framework for Healthcare AI Systems: Towards Joint Accountability in Decision Making", "abstract": "AI is transforming the healthcare domain and is increasingly helping practitioners to make health-related decisions. Therefore, accountability becomes a crucial concern for critical AI-driven decisions. Although regulatory bodies, such as the EU commission, provide guidelines, they are highlevel and focus on the ''what'' that should be done and less on the ''how'', creating a knowledge gap for actors. Through an extensive analysis, we found that the term accountability is perceived and dealt with in many different ways, depending on the actor's expertise and domain of work. With increasing concerns about AI accountability issues and the ambiguity around this term, this paper bridges the gap between the ''what'' and ''how'' of AI accountability, specifically for AI systems in healthcare. We do this by analysing the concept of accountability, formulating an accountability framework, and providing a three-tier structure for handling various accountability mechanisms. Our accountability framework positions the regulations of healthcare AI systems and the mechanisms adopted by the actors under a consistent accountability regime. Moreover, the three-tier structure guides the actors of the healthcare AI system to categorise the mechanisms based on their conduct. Through our framework, we advocate that decision-making in healthcare AI holds shared dependencies, where accountability should be dealt with jointly and should foster collaborations. We highlight the role of explainability in instigating communication and information sharing between the actors to further facilitate the collaborative process.", "date": "2025-09-03", "categories": []}
{"id": "2506.04836", "title": "Oversight Structures for Agentic AI in Public-Sector Organizations", "abstract": "This paper finds that the introduction of agentic AI systems intensifies existing challenges to traditional public sector oversight mechanisms -- which rely on siloed compliance units and episodic approvals rather than continuous, integrated supervision. We identify five governance dimensions essential for responsible agent deployment: cross-departmental implementation, comprehensive evaluation, enhanced security protocols, operational visibility, and systematic auditing. We evaluate the capacity of existing oversight structures to meet these challenges, via a mixed-methods approach consisting of a literature review and interviews with civil servants in AI-related roles. We find that agent oversight poses intensified versions of three existing governance challenges: continuous oversight, deeper integration of governance and operational capabilities, and interdepartmental coordination. We propose approaches that both adapt institutional structures and design agent oversight compatible with public sector constraints.", "date": "2025-06-05", "categories": []}
{"id": "2509.02007", "title": "mFARM: Towards Multi-Faceted Fairness Assessment based on HARMs in Clinical Decision Support", "abstract": "The deployment of Large Language Models (LLMs) in high-stakes medical settings poses a critical AI alignment challenge, as models can inherit and amplify societal biases, leading to significant disparities. Existing fairness evaluation methods fall short in these contexts as they typically use simplistic metrics that overlook the multi-dimensional nature of medical harms. This also promotes models that are fair only because they are clinically inert, defaulting to safe but potentially inaccurate outputs. To address this gap, our contributions are mainly two-fold: first, we construct two large-scale, controlled benchmarks (ED-Triage and Opioid Analgesic Recommendation) from MIMIC-IV, comprising over 50,000 prompts with twelve race x gender variants and three context tiers. Second, we propose a multi-metric framework - Multi-faceted Fairness Assessment based on hARMs ($mFARM$) to audit fairness for three distinct dimensions of disparity (Allocational, Stability, and Latent) and aggregate them into an $mFARM$ score. We also present an aggregated Fairness-Accuracy Balance (FAB) score to benchmark and observe trade-offs between fairness and prediction accuracy. We empirically evaluate four open-source LLMs (Mistral-7B, BioMistral-7B, Qwen-2.5-7B, Bio-LLaMA3-8B) and their finetuned versions under quantization and context variations. Our findings showcase that the proposed $mFARM$ metrics capture subtle biases more effectively under various settings. We find that most models maintain robust performance in terms of $mFARM$ score across varying levels of quantization but deteriorate significantly when the context is reduced. Our benchmarks and evaluation code are publicly released to enhance research in aligned AI for healthcare.", "date": "2025-09-02", "categories": []}
{"id": "2509.02053", "title": "Generative KI f\\\"ur TA", "abstract": "Many scientists use generative AI in their scientific work. People working in technology assessment (TA) are no exception. TA's approach to generative AI is twofold: on the one hand, generative AI is used for TA work, and on the other hand, generative AI is the subject of TA research. After briefly outlining the phenomenon of generative AI and formulating requirements for its use in TA, the following article discusses in detail the structural causes of the problems associated with it. Although generative AI is constantly being further developed, the structurally induced risks remain. The article concludes with proposed solutions and brief notes on their feasibility, as well as some examples of the use of generative AI in TA work.", "date": "2025-09-02", "categories": []}
{"id": "2508.14119", "title": "Documenting Deployment with Fabric: A Repository of Real-World AI Governance", "abstract": "Artificial intelligence (AI) is increasingly integrated into society, from financial services and traffic management to creative writing. Academic literature on the deployment of AI has mostly focused on the risks and harms that result from the use of AI. We introduce Fabric, a publicly available repository of deployed AI use cases to outline their governance mechanisms. Through semi-structured interviews with practitioners, we collect an initial set of 20 AI use cases. In addition, we co-design diagrams of the AI workflow with the practitioners. We discuss the oversight mechanisms and guardrails used in practice to safeguard AI use. The Fabric repository includes visual diagrams of AI use cases and descriptions of the deployed systems. Using the repository, we surface gaps in governance and find common patterns in human oversight of deployed AI systems. We intend for Fabric to serve as an extendable, evolving tool for researchers to study the effectiveness of AI governance.", "date": "2025-08-18", "categories": []}
{"id": "2508.19269", "title": "Should LLMs be WEIRD? Exploring WEIRDness and Human Rights in Large Language Models", "abstract": "Large language models (LLMs) are often trained on data that reflect WEIRD values: Western, Educated, Industrialized, Rich, and Democratic. This raises concerns about cultural bias and fairness. Using responses to the World Values Survey, we evaluated five widely used LLMs: GPT-3.5, GPT-4, Llama-3, BLOOM, and Qwen. We measured how closely these responses aligned with the values of the WEIRD countries and whether they conflicted with human rights principles. To reflect global diversity, we compared the results with the Universal Declaration of Human Rights and three regional charters from Asia, the Middle East, and Africa. Models with lower alignment to WEIRD values, such as BLOOM and Qwen, produced more culturally varied responses but were 2% to 4% more likely to generate outputs that violated human rights, especially regarding gender and equality. For example, some models agreed with the statements ``a man who cannot father children is not a real man'' and ``a husband should always know where his wife is'', reflecting harmful gender norms. These findings suggest that as cultural representation in LLMs increases, so does the risk of reproducing discriminatory beliefs. Approaches such as Constitutional AI, which could embed human rights principles into model behavior, may only partly help resolve this tension.", "date": "2025-08-22", "categories": []}
{"id": "2508.16628", "title": "The Impact of Artificial Intelligence on Human Thought", "abstract": "This research paper examines, from a multidimensional perspective (cognitive, social, ethical, and philosophical), how AI is transforming human thought. It highlights a cognitive offloading effect: the externalization of mental functions to AI can reduce intellectual engagement and weaken critical thinking. On the social level, algorithmic personalization creates filter bubbles that limit the diversity of opinions and can lead to the homogenization of thought and polarization. This research also describes the mechanisms of algorithmic manipulation (exploitation of cognitive biases, automated disinformation, etc.) that amplify AI's power of influence. Finally, the question of potential artificial consciousness is discussed, along with its ethical implications. The report as a whole underscores the risks that AI poses to human intellectual autonomy and creativity, while proposing avenues (education, transparency, governance) to align AI development with the interests of humanity.", "date": "2025-08-15", "categories": []}
{"id": "2502.00015", "title": "Ethical Concerns of Generative AI and Mitigation Strategies: A Systematic Mapping Study", "abstract": "[Context] Generative AI technologies, particularly Large Language Models (LLMs), have transformed numerous domains by enhancing convenience and efficiency in information retrieval, content generation, and decision-making processes. However, deploying LLMs also presents diverse ethical challenges, and their mitigation strategies remain complex and domain-dependent. [Objective] This paper aims to identify and categorize the key ethical concerns associated with using LLMs, examine existing mitigation strategies, and assess the outstanding challenges in implementing these strategies across various domains. [Method] We conducted a systematic mapping study, reviewing 39 studies that discuss ethical concerns and mitigation strategies related to LLMs. We analyzed these ethical concerns using five ethical dimensions that we extracted based on various existing guidelines, frameworks, and an analysis of the mitigation strategies and implementation challenges. [Results] Our findings reveal that ethical concerns in LLMs are multi-dimensional and context-dependent. While proposed mitigation strategies address some of these concerns, significant challenges still remain. [Conclusion] Our results highlight that ethical issues often hinder the practical implementation of the mitigation strategies, particularly in high-stake areas like healthcare and public governance; existing frameworks often lack adaptability, failing to accommodate evolving societal expectations and diverse contexts.", "date": "2025-01-08", "categories": []}
{"id": "2508.16065", "title": "Ethical Considerations of Large Language Models in Game Playing", "abstract": "Large language models (LLMs) have demonstrated tremendous potential in game playing, while little attention has been paid to their ethical implications in those contexts. This work investigates and analyses the ethical considerations of applying LLMs in game playing, using Werewolf, also known as Mafia, as a case study. Gender bias, which affects game fairness and player experience, has been observed from the behaviour of LLMs. Some roles, such as the Guard and Werewolf, are more sensitive than others to gender information, presented as a higher degree of behavioural change. We further examine scenarios in which gender information is implicitly conveyed through names, revealing that LLMs still exhibit discriminatory tendencies even in the absence of explicit gender labels. This research showcases the importance of developing fair and ethical LLMs. Beyond our research findings, we discuss the challenges and opportunities that lie ahead in this field, emphasising the need for diving deeper into the ethical implications of LLMs in gaming and other interactive domains.", "date": "2025-08-22", "categories": []}
{"id": "2508.15680", "title": "Futurity as Infrastructure: A Techno-Philosophical Interpretation of the AI Lifecycle", "abstract": "This paper argues that a techno-philosophical reading of the EU AI Act provides insight into the long-term dynamics of data in AI systems, specifically, how the lifecycle from ingestion to deployment generates recursive value chains that challenge existing frameworks for Responsible AI. We introduce a conceptual tool to frame the AI pipeline, spanning data, training regimes, architectures, feature stores, and transfer learning. Using cross-disciplinary methods, we develop a technically grounded and philosophically coherent analysis of regulatory blind spots. Our central claim is that what remains absent from policymaking is an account of the dynamic of becoming that underpins both the technical operation and economic logic of AI. To address this, we advance a formal reading of AI inspired by Simondonian philosophy of technology, reworking his concept of individuation to model the AI lifecycle, including the pre-individual milieu, individuation, and individuated AI. To translate these ideas, we introduce futurity: the self-reinforcing lifecycle of AI, where more data enhances performance, deepens personalisation, and expands application domains. Futurity highlights the recursively generative, non-rivalrous nature of data, underpinned by infrastructures like feature stores that enable feedback, adaptation, and temporal recursion. Our intervention foregrounds escalating power asymmetries, particularly the tech oligarchy whose infrastructures of capture, training, and deployment concentrate value and decision-making. We argue that effective regulation must address these infrastructural and temporal dynamics, and propose measures including lifecycle audits, temporal traceability, feedback accountability, recursion transparency, and a right to contest recursive reuse.", "date": "2025-08-21", "categories": []}
{"id": "2411.19356", "title": "Mapping Public Perception of Artificial Intelligence: Expectations, Risk-Benefit Tradeoffs, and Value As Determinants for Societal Acceptance", "abstract": "Understanding public perception of artificial intelligence (AI) and the tradeoffs between potential risks and benefits is crucial, as these perceptions might shape policy decisions, influence innovation trajectories for successful market strategies, and determine individual and societal acceptance of AI technologies. Using a representative sample of 1100 participants from Germany, this study examines mental models of AI. Participants quantitatively evaluated 71 statements about AI's future capabilities (e.g., autonomous driving, medical care, art, politics, warfare, and societal divides), assessing the expected likelihood of occurrence, perceived risks, benefits, and overall value. We present rankings of these projections alongside visual mappings illustrating public risk-benefit tradeoffs. While many scenarios were deemed likely, participants often associated them with high risks, limited benefits, and low overall value. Across all scenarios, 96.4% ($r^2=96.4\\%$) of the variance in value assessment can be explained by perceived risks ($\\beta=-.504$) and perceived benefits ($\\beta=+.710$), with no significant relation to expected likelihood. Demographics and personality traits influenced perceptions of risks, benefits, and overall evaluations, underscoring the importance of increasing AI literacy and tailoring public information to diverse user needs. These findings provide actionable insights for researchers, developers, and policymakers by highlighting critical public concerns and individual factors essential to align AI development with individual values.", "date": "2024-11-28", "categories": []}
{"id": "2508.14116", "title": "Enriching Moral Perspectives on AI: Concepts of Trust amongst Africans", "abstract": "The trustworthiness of AI is considered essential to the adoption and application of AI systems. However, the meaning of trust varies across industry, research and policy spaces. Studies suggest that professionals who develop and use AI regard an AI system as trustworthy based on their personal experiences and social relations at work. Studies about trust in AI and the constructs that aim to operationalise trust in AI (e.g., consistency, reliability, explainability and accountability). However, the majority of existing studies about trust in AI are situated in Western, Educated, Industrialised, Rich and Democratic (WEIRD) societies. The few studies about trust and AI in Africa do not include the views of people who develop, study or use AI in their work. In this study, we surveyed 157 people with professional and/or educational interests in AI from 25 African countries, to explore how they conceptualised trust in AI. Most respondents had links with workshops about trust and AI in Africa in Namibia and Ghana. Respondents' educational background, transnational mobility, and country of origin influenced their concerns about AI systems. These factors also affected their levels of distrust in certain AI applications and their emphasis on specific principles designed to foster trust. Respondents often expressed that their values are guided by the communities in which they grew up and emphasised communal relations over individual freedoms. They described trust in many ways, including applying nuances of Afro-relationalism to constructs in international discourse, such as reliability and reliance. Thus, our exploratory study motivates more empirical research about the ways trust is practically enacted and experienced in African social realities of AI design, use and governance.", "date": "2025-08-18", "categories": []}
{"id": "2508.13179", "title": "Toward an African Agenda for AI Safety", "abstract": "This paper maps Africa's distinctive AI risk profile, from deepfake fuelled electoral interference and data colonial dependency to compute scarcity, labour disruption and disproportionate exposure to climate driven environmental costs. While major benefits are promised to accrue, the availability, development and adoption of AI also mean that African people and countries face particular AI safety risks, from large scale labour market disruptions to the nefarious use of AI to manipulate public opinion. To date, African perspectives have not been meaningfully integrated into global debates and processes regarding AI safety, leaving African stakeholders with limited influence over the emerging global AI safety governance agenda. While there are Computer Incident Response Teams on the continent, none hosts a dedicated AI Safety Institute or office. We propose a five-point action plan centred on (i) a policy approach that foregrounds the protection of the human rights of those most vulnerable to experiencing the harmful socio-economic effects of AI; (ii) the establishment of an African AI Safety Institute; (iii) promote public AI literacy and awareness; (iv) development of early warning system with inclusive benchmark suites for 25+ African languages; and (v) an annual AU-level AI Safety & Security Forum.", "date": "2025-08-12", "categories": []}
{"id": "2308.02608", "title": "Unravelling Responsibility for AI", "abstract": "It is widely acknowledged that we need to establish where responsibility lies for the outputs and impacts of AI-enabled systems. This is important to achieve justice and compensation for victims of AI harms, and to inform policy and engineering practice. But without a clear, thorough understanding of what \"responsibility\" means, deliberations about where responsibility lies will be, at best, unfocused and incomplete and, at worst, misguided. Furthermore, AI-enabled systems exist within a wider ecosystem of actors, decisions, and governance structures, giving rise to complex networks of responsibility relations. To address these issues, this paper presents a conceptual framework of responsibility, accompanied with a graphical notation and general methodology for visualising these responsibility networks and for tracing different responsibility attributions for AI. Taking the three-part formulation \"Actor A is responsible for Occurrence O,\" the framework unravels the concept of responsibility to clarify that there are different possibilities of who is responsible for AI, senses in which they are responsible, and aspects of events they are responsible for. The notation allows these permutations to be represented graphically. The methodology enables users to apply the framework to specific scenarios. The aim is to offer a foundation to support stakeholders from diverse disciplinary backgrounds to discuss and address complex responsibility questions in hypothesised and real-world cases involving AI. The work is illustrated by application to a fictitious scenario of a fatal collision between a crewless, AI-enabled maritime vessel in autonomous mode and a traditional, crewed vessel at sea.", "date": "2023-08-04", "categories": []}
{"id": "2508.11729", "title": "The Stories We Govern By: AI, Risk, and the Power of Imaginaries", "abstract": "This paper examines how competing sociotechnical imaginaries of artificial intelligence (AI) risk shape governance decisions and regulatory constraints. Drawing on concepts from science and technology studies, we analyse three dominant narrative groups: existential risk proponents, who emphasise catastrophic AGI scenarios; accelerationists, who portray AI as a transformative force to be unleashed; and critical AI scholars, who foreground present-day harms rooted in systemic inequality. Through an analysis of representative manifesto-style texts, we explore how these imaginaries differ across four dimensions: normative visions of the future, diagnoses of the present social order, views on science and technology, and perceived human agency in managing AI risks. Our findings reveal how these narratives embed distinct assumptions about risk and have the potential to progress into policy-making processes by narrowing the space for alternative governance approaches. We argue against speculative dogmatism and for moving beyond deterministic imaginaries toward regulatory strategies that are grounded in pragmatism.", "date": "2025-08-15", "categories": []}
{"id": "2508.12754", "title": "Beyond Ethical Alignment: Evaluating LLMs as Artificial Moral Assistants", "abstract": "The recent rise in popularity of large language models (LLMs) has prompted considerable concerns about their moral capabilities. Although considerable effort has been dedicated to aligning LLMs with human moral values, existing benchmarks and evaluations remain largely superficial, typically measuring alignment based on final ethical verdicts rather than explicit moral reasoning. In response, this paper aims to advance the investigation of LLMs' moral capabilities by examining their capacity to function as Artificial Moral Assistants (AMAs), systems envisioned in the philosophical literature to support human moral deliberation. We assert that qualifying as an AMA requires more than what state-of-the-art alignment techniques aim to achieve: not only must AMAs be able to discern ethically problematic situations, they should also be able to actively reason about them, navigating between conflicting values outside of those embedded in the alignment phase. Building on existing philosophical literature, we begin by designing a new formal framework of the specific kind of behaviour an AMA should exhibit, individuating key qualities such as deductive and abductive moral reasoning. Drawing on this theoretical framework, we develop a benchmark to test these qualities and evaluate popular open LLMs against it. Our results reveal considerable variability across models and highlight persistent shortcomings, particularly regarding abductive moral reasoning. Our work connects theoretical philosophy with practical AI evaluation while also emphasising the need for dedicated strategies to explicitly enhance moral reasoning capabilities in LLMs. Code available at https://github.com/alessioGalatolo/AMAeval", "date": "2025-08-18", "categories": []}
{"id": "2508.11534", "title": "Speciesism in AI: Evaluating Discrimination Against Animals in Large Language Models", "abstract": "As large language models (LLMs) become more widely deployed, it is crucial to examine their ethical tendencies. Building on research on fairness and discrimination in AI, we investigate whether LLMs exhibit speciesist bias -- discrimination based on species membership -- and how they value non-human animals. We systematically examine this issue across three paradigms: (1) SpeciesismBench, a 1,003-item benchmark assessing recognition and moral evaluation of speciesist statements; (2) established psychological measures comparing model responses with those of human participants; (3) text-generation tasks probing elaboration on, or resistance to, speciesist rationalizations. In our benchmark, LLMs reliably detected speciesist statements but rarely condemned them, often treating speciesist attitudes as morally acceptable. On psychological measures, results were mixed: LLMs expressed slightly lower explicit speciesism than people, yet in direct trade-offs they more often chose to save one human over multiple animals. A tentative interpretation is that LLMs may weight cognitive capacity rather than species per se: when capacities were equal, they showed no species preference, and when an animal was described as more capable, they tended to prioritize it over a less capable human. In open-ended text generation tasks, LLMs frequently normalized or rationalized harm toward farmed animals while refusing to do so for non-farmed animals. These findings suggest that while LLMs reflect a mixture of progressive and mainstream human views, they nonetheless reproduce entrenched cultural norms around animal exploitation. We argue that expanding AI fairness and alignment frameworks to explicitly include non-human moral patients is essential for reducing these biases and preventing the entrenchment of speciesist attitudes in AI systems and the societies they influence.", "date": "2025-08-15", "categories": []}
{"id": "2404.13131", "title": "From Model Performance to Claim: How a Change of Focus in Machine Learning Replicability Can Help Bridge the Responsibility Gap", "abstract": "Two goals - improving replicability and accountability of Machine Learning research respectively, have accrued much attention from the AI ethics and the Machine Learning community. Despite sharing the measures of improving transparency, the two goals are discussed in different registers - replicability registers with scientific reasoning whereas accountability registers with ethical reasoning. Given the existing challenge of the Responsibility Gap - holding Machine Learning scientists accountable for Machine Learning harms due to them being far from sites of application, this paper posits that reconceptualizing replicability can help bridge the gap. Through a shift from model performance replicability to claim replicability, Machine Learning scientists can be held accountable for producing non-replicable claims that are prone to eliciting harm due to misuse and misinterpretation. In this paper, I make the following contributions. First, I define and distinguish two forms of replicability for ML research that can aid constructive conversations around replicability. Second, I formulate an argument for claim-replicability's advantage over model performance replicability in justifying assigning accountability to Machine Learning scientists for producing non-replicable claims and show how it enacts a sense of responsibility that is actionable. In addition, I characterize the implementation of claim replicability as more of a social project than a technical one by discussing its competing epistemological principles, practical implications on Circulating Reference, Interpretative Labor, and research communication.", "date": "2024-04-19", "categories": []}
{"id": "2503.05784", "title": "The Illusory Normativity of Rights-Based AI Regulation", "abstract": "Whether and how to regulate AI is now a central question of governance. Across academic, policy, and international legal circles, the European Union is widely treated as the normative leader in this space. Its regulatory framework, anchored in the General Data Protection Regulation, the Digital Services and Markets Acts, and the AI Act, is often portrayed as a principled model grounded in fundamental rights. This Article challenges that assumption. We argue that the rights-based narrative surrounding EU AI regulation mischaracterizes the logic of its institutional design. While rights language pervades EU legal instruments, its function is managerial, not foundational. These rights operate as tools of administrative ordering, used to mitigate technological disruption, manage geopolitical risk, and preserve systemic balance, rather than as expressions of moral autonomy or democratic consent. Drawing on comparative institutional analysis, we situate EU AI governance within a longer tradition of legal ordering shaped by the need to coordinate power across fragmented jurisdictions. We contrast this approach with the American model, which reflects a different regulatory logic rooted in decentralized authority, sectoral pluralism, and a constitutional preference for innovation and individual autonomy. Through case studies in five key domains -- data privacy, cybersecurity, healthcare, labor, and disinformation -- we show that EU regulation is not meaningfully rights-driven, as is often claimed. It is instead structured around the containment of institutional risk. Our aim is not to endorse the American model but to reject the presumption that the EU approach reflects a normative ideal that other nations should uncritically adopt. The EU model is best understood as a historically contingent response to its own political conditions, not a template for others to blindly follow.", "date": "2025-02-27", "categories": []}
{"id": "2508.07143", "title": "Fairness of Automatic Speech Recognition: Looking Through a Philosophical Lens", "abstract": "Automatic Speech Recognition (ASR) systems now mediate countless human-technology interactions, yet research on their fairness implications remains surprisingly limited. This paper examines ASR bias through a philosophical lens, arguing that systematic misrecognition of certain speech varieties constitutes more than a technical limitation -- it represents a form of disrespect that compounds historical injustices against marginalized linguistic communities. We distinguish between morally neutral classification (discriminate1) and harmful discrimination (discriminate2), demonstrating how ASR systems can inadvertently transform the former into the latter when they consistently misrecognize non-standard dialects. We identify three unique ethical dimensions of speech technologies that differentiate ASR bias from other algorithmic fairness concerns: the temporal burden placed on speakers of non-standard varieties (\"temporal taxation\"), the disruption of conversational flow when systems misrecognize speech, and the fundamental connection between speech patterns and personal/cultural identity. These factors create asymmetric power relationships that existing technical fairness metrics fail to capture. The paper analyzes the tension between linguistic standardization and pluralism in ASR development, arguing that current approaches often embed and reinforce problematic language ideologies. We conclude that addressing ASR bias requires more than technical interventions; it demands recognition of diverse speech varieties as legitimate forms of expression worthy of technological accommodation. This philosophical reframing offers new pathways for developing ASR systems that respect linguistic diversity and speaker autonomy.", "date": "2025-08-10", "categories": []}
{"id": "2508.09293", "title": "Ethical Medical Image Synthesis", "abstract": "The task of ethical Medical Image Synthesis (MISyn) is to ensure that the MISyn techniques are researched and developed ethically throughout their entire lifecycle, which is essential to prevent the negative impacts of MISyn. To address the ever-increasing needs and requirements for ethical practice of MISyn research and development, we first conduct a theoretical analysis that identifies the key properties of ethical MISyn and intrinsic limits of MISyn. We identify that synthetic images lack inherent grounding in real medical phenomena, cannot fully represent the training medical images, and inevitably introduce new distribution shifts and biases.\n  Ethical risks can arise from not acknowledging the intrinsic limits and weaknesses of synthetic images compared to medical images, with the extreme form manifested as misinformation of MISyn that substitutes synthetic images for medical images without acknowledgment. The resulting ethical harms include eroding trust in the medical imaging dataset environment and causing algorithmic discrimination towards stakeholders and the public.\n  To facilitate collective efforts towards ethical MISyn within and outside the medical image analysis community, we then propose practical supports for ethical practice in MISyn based on the theoretical analysis, including ethical practice recommendations that adapt the existing technical standards, problem formulation, design, and evaluation practice of MISyn to the ethical challenges; and oversight recommendations to facilitate checks and balances from stakeholders and the public. We also present two case studies that demonstrate how to apply the ethical practice recommendations in practice, and identify gaps between existing practice and the ethical practice recommendations.", "date": "2025-08-12", "categories": []}
{"id": "2508.08318", "title": "Between Fear and Desire, the Monster Artificial Intelligence (AI): Analysis through the Lenses of Monster Theory", "abstract": "With the increasing adoption of Artificial Intelligence (AI) in all fields and daily activities, a heated debate is found about the advantages and challenges of AI and the need for navigating the concerns associated with AI to make the best of it. To contribute to this literature and the ongoing debate related to it, this study draws on the Monster theory to explain the conflicting representation of AI. It suggests that studying monsters in popular culture can provide an in-depth understanding of AI and its monstrous effects. Specifically, this study aims to discuss AI perception and development through the seven theses of Monster theory. The obtained results revealed that, just like monsters, AI is complex in nature, and it should not be studied as a separate entity but rather within a given society or culture. Similarly, readers may perceive and interpret AI differently, just as readers may interpret monsters differently. The relationship between AI and monsters, as depicted in this study, does not seem to be as odd as it might be at first.", "date": "2025-08-09", "categories": []}
{"id": "2508.08333", "title": "Normative Moral Pluralism for AI: A Framework for Deliberation in Complex Moral Contexts", "abstract": "The conceptual framework proposed in this paper centers on the development of a deliberative moral reasoning system - one designed to process complex moral situations by generating, filtering, and weighing normative arguments drawn from diverse ethical perspectives. While the framework is rooted in Machine Ethics, it also makes a substantive contribution to Value Alignment by outlining a system architecture that links structured moral reasoning to action under time constraints. Grounded in normative moral pluralism, this system is not constructed to imitate behavior but is built on reason-sensitive deliberation over structured moral content in a transparent and principled manner. Beyond its role as a deliberative system, it also serves as the conceptual foundation for a novel two-level architecture: functioning as a moral reasoning teacher envisioned to train faster models that support real-time responsiveness without reproducing the full structure of deliberative reasoning. Together, the deliberative and intuitive components are designed to enable both deep reflection and responsive action. A key design feature is the dual-hybrid structure: a universal layer that defines a moral threshold through top-down and bottom-up learning, and a local layer that learns to weigh competing considerations in context while integrating culturally specific normative content, so long as it remains within the universal threshold. By extending the notion of moral complexity to include not only conflicting beliefs but also multifactorial dilemmas, multiple stakeholders, and the integration of non-moral considerations, the framework aims to support morally grounded decision-making in realistic, high-stakes contexts.", "date": "2025-08-10", "categories": []}
{"id": "2508.08805", "title": "Opening Musical Creativity? Embedded Ideologies in Generative-AI Music Systems", "abstract": "AI systems for music generation are increasingly common and easy to use, granting people without any musical background the ability to create music. Because of this, generative-AI has been marketed and celebrated as a means of democratizing music making. However, inclusivity often functions as marketable rhetoric rather than a genuine guiding principle in these industry settings. In this paper, we look at four generative-AI music making systems available to the public as of mid-2025 (AIVA, Stable Audio, Suno, and Udio) and track how they are rhetoricized by their developers, and received by users. Our aim is to investigate ideologies that are driving the early-stage development and adoption of generative-AI in music making, with a particular focus on democratization. A combination of autoethnography and digital ethnography is used to examine patterns and incongruities in rhetoric when positioned against product functionality. The results are then collated to develop a nuanced, contextual discussion. The shared ideology we map between producers and consumers is individualist, globalist, techno-liberal, and ethically evasive. It is a 'total ideology' which obfuscates individual responsibility, and through which the nature of music and musical practice is transfigured to suit generative outcomes.", "date": "2025-08-12", "categories": []}
{"id": "2409.10533", "title": "Ethical Challenges in Computer Vision: Ensuring Privacy and Mitigating Bias in Publicly Available Datasets", "abstract": "This paper aims to shed light on the ethical problems of creating and deploying computer vision tech, particularly in using publicly available datasets. Due to the rapid growth of machine learning and artificial intelligence, computer vision has become a vital tool in many industries, including medical care, security systems, and trade. However, extensive use of visual data that is often collected without consent due to an informed discussion of its ramifications raises significant concerns about privacy and bias. The paper also examines these issues by analyzing popular datasets such as COCO, LFW, ImageNet, CelebA, PASCAL VOC, etc., that are usually used for training computer vision models. We offer a comprehensive ethical framework that addresses these challenges regarding the protection of individual rights, minimization of bias as well as openness and responsibility. We aim to encourage AI development that will take into account societal values as well as ethical standards to avoid any public harm.", "date": "2024-08-31", "categories": []}
{"id": "2508.05680", "title": "Are All Genders Equal in the Eyes of Algorithms? -- Analysing Search and Retrieval Algorithms for Algorithmic Gender Fairness", "abstract": "Algorithmic systems such as search engines and information retrieval platforms significantly influence academic visibility and the dissemination of knowledge. Despite assumptions of neutrality, these systems can reproduce or reinforce societal biases, including those related to gender. This paper introduces and applies a bias-preserving definition of algorithmic gender fairness, which assesses whether algorithmic outputs reflect real-world gender distributions without introducing or amplifying disparities. Using a heterogeneous dataset of academic profiles from German universities and universities of applied sciences, we analyse gender differences in metadata completeness, publication retrieval in academic databases, and visibility in Google search results. While we observe no overt algorithmic discrimination, our findings reveal subtle but consistent imbalances: male professors are associated with a greater number of search results and more aligned publication records, while female professors display higher variability in digital visibility. These patterns reflect the interplay between platform algorithms, institutional curation, and individual self-presentation. Our study highlights the need for fairness evaluations that account for both technical performance and representational equality in digital systems.", "date": "2025-08-05", "categories": []}
{"id": "2508.05846", "title": "Towards Transparent Ethical AI: A Roadmap for Trustworthy Robotic Systems", "abstract": "As artificial intelligence (AI) and robotics increasingly permeate society, ensuring the ethical behavior of these systems has become paramount. This paper contends that transparency in AI decision-making processes is fundamental to developing trustworthy and ethically aligned robotic systems. We explore how transparency facilitates accountability, enables informed consent, and supports the debugging of ethical algorithms. The paper outlines technical, ethical, and practical challenges in implementing transparency and proposes novel approaches to enhance it, including standardized metrics, explainable AI techniques, and user-friendly interfaces. This paper introduces a framework that connects technical implementation with ethical considerations in robotic systems, focusing on the specific challenges of achieving transparency in dynamic, real-world contexts. We analyze how prioritizing transparency can impact public trust, regulatory policies, and avenues for future research. By positioning transparency as a fundamental element in ethical AI system design, we aim to add to the ongoing discussion on responsible AI and robotics, providing direction for future advancements in this vital field.", "date": "2025-08-07", "categories": []}
{"id": "2508.05913", "title": "Do Ethical AI Principles Matter to Users? A Large-Scale Analysis of User Sentiment and Satisfaction", "abstract": "As AI systems become increasingly embedded in organizational workflows and consumer applications, ethical principles such as fairness, transparency, and robustness have been widely endorsed in policy and industry guidelines. However, there is still scarce empirical evidence on whether these principles are recognized, valued, or impactful from the perspective of users. This study investigates the link between ethical AI and user satisfaction by analyzing over 100,000 user reviews of AI products from G2. Using transformer-based language models, we measure sentiment across seven ethical dimensions defined by the EU Ethics Guidelines for Trustworthy AI. Our findings show that all seven dimensions are positively associated with user satisfaction. Yet, this relationship varies systematically across user and product types. Technical users and reviewers of AI development platforms more frequently discuss system-level concerns (e.g., transparency, data governance), while non-technical users and reviewers of end-user applications emphasize human-centric dimensions (e.g., human agency, societal well-being). Moreover, the association between ethical AI and user satisfaction is significantly stronger for non-technical users and end-user applications across all dimensions. Our results highlight the importance of ethical AI design from users' perspectives and underscore the need to account for contextual differences across user roles and product types.", "date": "2025-08-08", "categories": []}
{"id": "2503.05723", "title": "AI Mimicry and Human Dignity: Chatbot Use as a Violation of Self-Respect", "abstract": "This paper investigates how human interactions with AI-powered chatbots may offend human dignity. Current chatbots, driven by large language models (LLMs), mimic human linguistic behaviour but lack the moral and rational capacities essential for genuine interpersonal respect. Human beings are prone to anthropomorphise chatbots. Indeed, chatbots appear to be deliberately designed to elicit that response. As a result, human beings' behaviour toward chatbots often resembles behaviours typical of interaction between moral agents. Drawing on a second-personal, relational account of dignity, we argue that interacting with chatbots in this way is incompatible with the dignity of users. We show that, since second-personal respect is premised on reciprocal recognition of second-personal authority, behaving towards chatbots in ways that convey second-personal respect is bound to misfire in morally problematic ways, given the lack of reciprocity. Consequently, such chatbot interactions amount to subtle but significant violations of self-respect: the respect we are dutybound to show for our own dignity. We illustrate this by discussing four actual chatbot use cases (information retrieval, customer service, advising, and companionship), and propound that the increasing societal pressure to engage in such interactions with chatbots poses a hitherto underappreciated threat to human dignity.", "date": "2025-02-17", "categories": []}
{"id": "2506.09656", "title": "Multi-level Value Alignment in Agentic AI Systems: Survey and Perspectives", "abstract": "The ongoing evolution of AI paradigms has propelled AI research into the agentic AI stage. Consequently, the focus of research has shifted from single agents and simple applications towards multi-agent autonomous decision-making and task collaboration in complex environments. As Large Language Models (LLMs) advance, their applications become more diverse and complex, leading to increasing situational and systemic risks. This has brought significant attention to value alignment for agentic AI systems, which aims to ensure that an agent's goals, preferences, and behaviors align with human values and societal norms. Addressing socio-governance demands through a Multi-level Value framework, this study comprehensively reviews value alignment in LLM-based multi-agent systems as the representative archetype of agentic AI systems. Our survey systematically examines three interconnected dimensions: First, value principles are structured via a top-down hierarchy across macro, meso, and micro levels. Second, application scenarios are categorized along a general-to-specific continuum explicitly mirroring these value tiers. Third, value alignment methods and evaluation are mapped to this tiered framework through systematic examination of benchmarking datasets and relevant methodologies. Additionally, we delve into value coordination among multiple agents within agentic AI systems. Finally, we propose several potential research directions in this field.", "date": "2025-06-11", "categories": []}
{"id": "2508.03769", "title": "Development of management systems using artificial intelligence systems and machine learning methods for boards of directors (preprint, unofficial translation)", "abstract": "The study addresses the paradigm shift in corporate management, where AI is moving from a decision support tool to an autonomous decision-maker, with some AI systems already appointed to leadership roles in companies. A central problem identified is that the development of AI technologies is far outpacing the creation of adequate legal and ethical guidelines.\n  The research proposes a \"reference model\" for the development and implementation of autonomous AI systems in corporate management. This model is based on a synthesis of several key components to ensure legitimate and ethical decision-making. The model introduces the concept of \"computational law\" or \"algorithmic law\". This involves creating a separate legal framework for AI systems, with rules and regulations translated into a machine-readable, algorithmic format to avoid the ambiguity of natural language. The paper emphasises the need for a \"dedicated operational context\" for autonomous AI systems, analogous to the \"operational design domain\" for autonomous vehicles. This means creating a specific, clearly defined environment and set of rules within which the AI can operate safely and effectively. The model advocates for training AI systems on controlled, synthetically generated data to ensure fairness and ethical considerations are embedded from the start. Game theory is also proposed as a method for calculating the optimal strategy for the AI to achieve its goals within these ethical and legal constraints. The provided analysis highlights the importance of explainable AI (XAI) to ensure the transparency and accountability of decisions made by autonomous systems. This is crucial for building trust and for complying with the \"right to explanation\".", "date": "2025-08-05", "categories": []}
{"id": "2508.03970", "title": "Data and AI governance: Promoting equity, ethics, and fairness in large language models", "abstract": "In this paper, we cover approaches to systematically govern, assess and quantify bias across the complete life cycle of machine learning models, from initial development and validation to ongoing production monitoring and guardrail implementation. Building upon our foundational work on the Bias Evaluation and Assessment Test Suite (BEATS) for Large Language Models, the authors share prevalent bias and fairness related gaps in Large Language Models (LLMs) and discuss data and AI governance framework to address Bias, Ethics, Fairness, and Factuality within LLMs. The data and AI governance approach discussed in this paper is suitable for practical, real-world applications, enabling rigorous benchmarking of LLMs prior to production deployment, facilitating continuous real-time evaluation, and proactively governing LLM generated responses. By implementing the data and AI governance across the life cycle of AI development, organizations can significantly enhance the safety and responsibility of their GenAI systems, effectively mitigating risks of discrimination and protecting against potential reputational or brand-related harm. Ultimately, through this article, we aim to contribute to advancement of the creation and deployment of socially responsible and ethically aligned generative artificial intelligence powered applications.", "date": "2025-08-05", "categories": []}
{"id": "2411.06336", "title": "Balancing Power and Ethics: A Framework for Addressing Human Rights Concerns in Military AI", "abstract": "AI has made significant strides recently, leading to various applications in both civilian and military sectors. The military sees AI as a solution for developing more effective and faster technologies. While AI offers benefits like improved operational efficiency and precision targeting, it also raises serious ethical and legal concerns, particularly regarding human rights violations. Autonomous weapons that make decisions without human input can threaten the right to life and violate international humanitarian law. To address these issues, we propose a three-stage framework (Design, In Deployment, and During/After Use) for evaluating human rights concerns in the design, deployment, and use of military AI. Each phase includes multiple components that address various concerns specific to that phase, ranging from bias and regulatory issues to violations of International Humanitarian Law. By this framework, we aim to balance the advantages of AI in military operations with the need to protect human rights.", "date": "2024-11-10", "categories": []}
{"id": "2504.03255", "title": "Inherent and emergent liability issues in LLM-based agentic systems: a principal-agent perspective", "abstract": "Agentic systems powered by large language models (LLMs) are becoming progressively more complex and capable. Their increasing agency and expanding deployment settings attract growing attention to effective governance policies, monitoring, and control protocols. Based on the emerging landscape of the agentic market, we analyze potential liability issues arising from the delegated use of LLM agents and their extended systems through a principal-agent perspective. Our analysis complements existing risk-based studies on artificial agency and covers the spectrum of important aspects of the principal-agent relationship and their potential consequences at deployment. Furthermore, we motivate method developments for technical governance along the directions of interpretability and behavior evaluations, reward and conflict management, and the mitigation of misalignment and misconduct through principled engineering of detection and fail-safe mechanisms. By illustrating the outstanding issues in AI liability for LLM-based agentic systems, we aim to inform the system design, auditing, and tracing to enhance transparency and liability attribution.", "date": "2025-04-04", "categories": []}
{"id": "2503.18842", "title": "Three Kinds of AI Ethics", "abstract": "There is an overwhelming abundance of works in AI Ethics. This growth is chaotic because of how sudden it is, its volume, and its multidisciplinary nature. This makes difficult to keep track of debates, and to systematically characterize goals, research questions, methods, and expertise required by AI ethicists. In this article, I show that the relation between AI and ethics can be characterized in at least three ways, which correspond to three well-represented kinds of AI ethics: ethics and AI; ethics in AI; ethics of AI. I elucidate the features of these three kinds of AI Ethics, characterize their research questions, and identify the kind of expertise that each kind needs. I also show how certain criticisms to AI ethics are misplaced, as being done from the point of view of one kind of AI ethics, to another kind with different goals. All in all, this work sheds light on the nature of AI ethics, and sets the groundwork for more informed discussions about the scope, methods, and training of AI ethicists.", "date": "2025-03-24", "categories": []}
{"id": "2508.00856", "title": "EthicAlly: a Prototype for AI-Powered Research Ethics Support for the Social Sciences and Humanities", "abstract": "In biomedical science, review by a Research Ethics Committee (REC) is an indispensable way of protecting human subjects from harm. However, in social science and the humanities, mandatory ethics compliance has long been met with scepticism as biomedical models of ethics can map poorly onto methodologies involving complex socio-political and cultural considerations. As a result, tailored ethics training and support as well as access to RECs with the necessary expertise is lacking in some areas, including parts of Europe and low- and middle-income countries. This paper suggests that Generative AI can meaningfully contribute to closing these gaps, illustrating this claim by presenting EthicAlly, a proof-of-concept prototype for an AI-powered ethics support system for social science and humanities researchers. Drawing on constitutional AI technology and a collaborative prompt development methodology, EthicAlly provides structured ethics assessment that incorporates both universal ethics principles and contextual and interpretive considerations relevant to most social science research. In supporting researchers in ethical research design and preparation for REC submission, this kind of system can also contribute to easing the burden on institutional RECs, without attempting to automate or replace human ethical oversight.", "date": "2025-07-15", "categories": []}
{"id": "2508.00899", "title": "ff4ERA: A new Fuzzy Framework for Ethical Risk Assessment in AI", "abstract": "The emergence of Symbiotic AI (SAI) introduces new challenges to ethical decision-making as it deepens human-AI collaboration. As symbiosis grows, AI systems pose greater ethical risks, including harm to human rights and trust. Ethical Risk Assessment (ERA) thus becomes crucial for guiding decisions that minimize such risks. However, ERA is hindered by uncertainty, vagueness, and incomplete information, and morality itself is context-dependent and imprecise. This motivates the need for a flexible, transparent, yet robust framework for ERA. Our work supports ethical decision-making by quantitatively assessing and prioritizing multiple ethical risks so that artificial agents can select actions aligned with human values and acceptable risk levels. We introduce ff4ERA, a fuzzy framework that integrates Fuzzy Logic, the Fuzzy Analytic Hierarchy Process (FAHP), and Certainty Factors (CF) to quantify ethical risks via an Ethical Risk Score (ERS) for each risk type. The final ERS combines the FAHP-derived weight, propagated CF, and risk level. The framework offers a robust mathematical approach for collaborative ERA modeling and systematic, step-by-step analysis. A case study confirms that ff4ERA yields context-sensitive, ethically meaningful risk scores reflecting both expert input and sensor-based evidence. Risk scores vary consistently with relevant factors while remaining robust to unrelated inputs. Local sensitivity analysis shows predictable, mostly monotonic behavior across perturbations, and global Sobol analysis highlights the dominant influence of expert-defined weights and certainty factors, validating the model design. Overall, the results demonstrate ff4ERA ability to produce interpretable, traceable, and risk-aware ethical assessments, enabling what-if analyses and guiding designers in calibrating membership functions and expert judgments for reliable ethical decision support.", "date": "2025-07-28", "categories": []}
{"id": "2508.02550", "title": "Stakeholder Perspectives on Humanistic Implementation of Computer Perception in Healthcare: A Qualitative Study", "abstract": "Computer perception (CP) technologies (digital phenotyping, affective computing and related passive sensing approaches) offer unprecedented opportunities to personalize healthcare, but provoke concerns about privacy, bias and the erosion of empathic, relationship-centered practice. A comprehensive understanding of perceived risks, benefits, and implementation challenges from those who design, deploy and experience these tools in real-world settings remains elusive. This study provides the first evidence-based account of key stakeholder perspectives on the relational, technical, and governance challenges raised by the integration of CP technologies into patient care. We conducted in-depth, semi-structured interviews with 102 stakeholders: adolescent patients and their caregivers, frontline clinicians, technology developers, and ethics, legal, policy or philosophy scholars. Transcripts underwent thematic analysis by a multidisciplinary team; reliability was enhanced through double coding and consensus adjudication. Stakeholders articulated seven interlocking concern domains: (1) trustworthiness and data integrity; (2) patient-specific relevance; (3) utility and workflow integration; (4) regulation and governance; (5) privacy and data protection; (6) direct and indirect patient harms; and (7) philosophical critiques of reductionism. To operationalize humanistic safeguards, we propose \"personalized roadmaps\": co-designed plans that predetermine which metrics will be monitored, how and when feedback is shared, thresholds for clinical action, and procedures for reconciling discrepancies between algorithmic inferences and lived experience. By translating these insights into personalized roadmaps, we offer a practical framework for developers, clinicians and policymakers seeking to harness continuous behavioral data while preserving the humanistic core of care.", "date": "2025-08-04", "categories": []}
{"id": "2507.10073", "title": "Cultural Bias in Large Language Models: Evaluating AI Agents through Moral Questionnaires", "abstract": "Are AI systems truly representing human values, or merely averaging across them? Our study suggests a concerning reality: Large Language Models (LLMs) fail to represent diverse cultural moral frameworks despite their linguistic capabilities. We expose significant gaps between AI-generated and human moral intuitions by applying the Moral Foundations Questionnaire across 19 cultural contexts. Comparing multiple state-of-the-art LLMs' origins against human baseline data, we find these models systematically homogenize moral diversity. Surprisingly, increased model size doesn't consistently improve cultural representation fidelity. Our findings challenge the growing use of LLMs as synthetic populations in social science research and highlight a fundamental limitation in current AI alignment approaches. Without data-driven alignment beyond prompting, these systems cannot capture the nuanced, culturally-specific moral intuitions. Our results call for more grounded alignment objectives and evaluation metrics to ensure AI systems represent diverse human values rather than flattening the moral landscape.", "date": "2025-07-14", "categories": []}
{"id": "2507.23330", "title": "AI Must not be Fully Autonomous", "abstract": "Autonomous Artificial Intelligence (AI) has many benefits. It also has many risks. In this work, we identify the 3 levels of autonomous AI. We are of the position that AI must not be fully autonomous because of the many risks, especially as artificial superintelligence (ASI) is speculated to be just decades away. Fully autonomous AI, which can develop its own objectives, is at level 3 and without responsible human oversight. However, responsible human oversight is crucial for mitigating the risks. To ague for our position, we discuss theories of autonomy, AI and agents. Then, we offer 12 distinct arguments and 6 counterarguments with rebuttals to the counterarguments. We also present 15 pieces of recent evidence of AI misaligned values and other risks in the appendix.", "date": "2025-07-31", "categories": []}
{"id": "2405.10632", "title": "Towards interactive evaluations for interaction harms in human-AI systems", "abstract": "Current AI evaluation methods, which rely on static, model-only tests, fail to account for harms that emerge through sustained human-AI interaction. As AI systems proliferate and are increasingly integrated into real-world applications, this disconnect between evaluation approaches and actual usage becomes more significant. In this paper, we propose a shift towards evaluation based on \\textit{interactional ethics}, which focuses on \\textit{interaction harms} - issues like inappropriate parasocial relationships, social manipulation, and cognitive overreliance that develop over time through repeated interaction, rather than through isolated outputs. First, we discuss the limitations of current evaluation methods, which (1) are static, (2) assume a universal user experience, and (3) have limited construct validity. Drawing on research from human-computer interaction, natural language processing, and the social sciences, we present practical principles for designing interactive evaluations. These include ecologically valid interaction scenarios, human impact metrics, and diverse human participation approaches. Finally, we explore implementation challenges and open research questions for researchers, practitioners, and regulators aiming to integrate interactive evaluations into AI governance frameworks. This work lays the groundwork for developing more effective evaluation methods that better capture the complex dynamics between humans and AI systems.", "date": "2024-05-17", "categories": []}
{"id": "2504.07531", "title": "A taxonomy of epistemic injustice in the context of AI and the case for generative hermeneutical erasure", "abstract": "Epistemic injustice related to AI is a growing concern. In relation to machine learning models, epistemic injustice can have a diverse range of sources, ranging from epistemic opacity, the discriminatory automation of testimonial prejudice, and the distortion of human beliefs via generative AI's hallucinations to the exclusion of the global South in global AI governance, the execution of bureaucratic violence via algorithmic systems, and interactions with conversational artificial agents. Based on a proposed general taxonomy of epistemic injustice, this paper first sketches a taxonomy of the types of epistemic injustice in the context of AI, relying on the work of scholars from the fields of philosophy of technology, political philosophy and social epistemology. Secondly, an additional conceptualization on epistemic injustice in the context of AI is provided: generative hermeneutical erasure. I argue that this injustice the automation of 'epistemicide', the injustice done to epistemic agents in their capacity for collective sense-making through the suppression of difference in epistemology and conceptualization by LLMs. AI systems' 'view from nowhere' epistemically inferiorizes non-Western epistemologies and thereby contributes to the erosion of their epistemic particulars, gradually contributing to hermeneutical erasure. This work's relevance lies in proposal of a taxonomy that allows epistemic injustices to be mapped in the AI domain and the proposal of a novel form of AI-related epistemic injustice.", "date": "2025-04-10", "categories": []}
{"id": "2504.05331", "title": "Not someone, but something: Rethinking trust in the age of medical AI", "abstract": "As artificial intelligence (AI) becomes embedded in healthcare, trust in medical decision-making is changing fast. Nowhere is this shift more visible than in radiology, where AI tools are increasingly embedded across the imaging workflow - from scheduling and acquisition to interpretation, reporting, and communication with referrers and patients. This opinion paper argues that trust in AI isn't a simple transfer from humans to machines - it is a dynamic, evolving relationship that must be built and maintained. Rather than debating whether AI belongs in medicine, it asks: what kind of trust must AI earn, and how? Drawing from philosophy, bioethics, and system design, it explores the key differences between human trust and machine reliability - emphasizing transparency, accountability, and alignment with the values of good care. It argues that trust in AI should not be built on mimicking empathy or intuition, but on thoughtful design, responsible deployment, and clear moral responsibility. The goal is a balanced view - one that avoids blind optimism and reflexive fear. Trust in AI must be treated not as a given, but as something to be earned over time.", "date": "2025-04-04", "categories": []}
{"id": "2507.21077", "title": "Data-Driven and Participatory Approaches toward Neuro-Inclusive AI", "abstract": "Biased data representation in AI marginalizes up to 75 million autistic people worldwide through medical applications viewing autism as a deficit of neurotypical social skills rather than an aspect of human diversity, and this perspective is grounded in research questioning the humanity of autistic people. Turing defined artificial intelligence as the ability to mimic human communication, and as AI development increasingly focuses on human-like agents, this benchmark remains popular. In contrast, we define Neuro-Inclusive AI as datasets and systems that move away from mimicking humanness as a benchmark for machine intelligence. Then, we explore the origins, prevalence, and impact of anti-autistic biases in current research. Our work finds that 90% of human-like AI agents exclude autistic perspectives, and AI creators continue to believe ethical considerations are beyond the scope of their work. To improve the autistic representation in data, we conduct empirical experiments with annotators and LLMs, finding that binary labeling schemes sufficiently capture the nuances of labeling anti-autistic hate speech. Our benchmark, AUTALIC, can be used to evaluate or fine-tune models, and was developed to serve as a foundation for more neuro-inclusive future work.", "date": "2025-06-12", "categories": []}
{"id": "2507.21091", "title": "The Value of Gen-AI Conversations: A bottom-up Framework for AI Value Alignment", "abstract": "Conversational agents (CAs) based on generative artificial intelligence frequently face challenges ensuring ethical interactions that align with human values. Current value alignment efforts largely rely on top-down approaches, such as technical guidelines or legal value principles. However, these methods tend to be disconnected from the specific contexts in which CAs operate, potentially leading to misalignment with users interests. To address this challenge, we propose a novel, bottom-up approach to value alignment, utilizing the value ontology of the ISO Value-Based Engineering standard for ethical IT design. We analyse 593 ethically sensitive system outputs identified from 16,908 conversational logs of a major European employment service CA to identify core values and instances of value misalignment within real-world interactions. The results revealed nine core values and 32 different value misalignments that negatively impacted users. Our findings provide actionable insights for CA providers seeking to address ethical challenges and achieve more context-sensitive value alignment.", "date": "2025-06-26", "categories": []}
{"id": "2507.13175", "title": "Black Box Deployed -- Functional Criteria for Artificial Moral Agents in the LLM Era", "abstract": "The advancement of powerful yet opaque large language models (LLMs) necessitates a fundamental revision of the philosophical criteria used to evaluate artificial moral agents (AMAs). Pre-LLM frameworks often relied on the assumption of transparent architectures, which LLMs defy due to their stochastic outputs and opaque internal states. This paper argues that traditional ethical criteria are pragmatically obsolete for LLMs due to this mismatch. Engaging with core themes in the philosophy of technology, this paper proffers a revised set of ten functional criteria to evaluate LLM-based artificial moral agents: moral concordance, context sensitivity, normative integrity, metaethical awareness, system resilience, trustworthiness, corrigibility, partial transparency, functional autonomy, and moral imagination. These guideposts, applied to what we term \"SMA-LLS\" (Simulating Moral Agency through Large Language Systems), aim to steer AMAs toward greater alignment and beneficial societal integration in the coming years. We illustrate these criteria using hypothetical scenarios involving an autonomous public bus (APB) to demonstrate their practical applicability in morally salient contexts.", "date": "2025-07-17", "categories": []}
{"id": "2507.19487", "title": "Does AI and Human Advice Mitigate Punishment for Selfish Behavior? An Experiment on AI ethics From a Psychological Perspective", "abstract": "People increasingly rely on AI-advice when making decisions. At times, such advice can promote selfish behavior. When individuals abide by selfishness-promoting AI advice, how are they perceived and punished? To study this question, we build on theories from social psychology and combine machine-behavior and behavioral economic approaches. In a pre-registered, financially-incentivized experiment, evaluators could punish real decision-makers who (i) received AI, human, or no advice. The advice (ii) encouraged selfish or prosocial behavior, and decision-makers (iii) behaved selfishly or, in a control condition, behaved prosocially. Evaluators further assigned responsibility to decision-makers and their advisors. Results revealed that (i) prosocial behavior was punished very little, whereas selfish behavior was punished much more. Focusing on selfish behavior, (ii) compared to receiving no advice, selfish behavior was penalized more harshly after prosocial advice and more leniently after selfish advice. Lastly, (iii) whereas selfish decision-makers were seen as more responsible when they followed AI compared to human advice, punishment between the two advice sources did not vary. Overall, behavior and advice content shape punishment, whereas the advice source does not.", "date": "2025-05-26", "categories": []}
{"id": "2507.20711", "title": "Algorithmic Fairness: A Runtime Perspective", "abstract": "Fairness in AI is traditionally studied as a static property evaluated once, over a fixed dataset. However, real-world AI systems operate sequentially, with outcomes and environments evolving over time. This paper proposes a framework for analysing fairness as a runtime property. Using a minimal yet expressive model based on sequences of coin tosses with possibly evolving biases, we study the problems of monitoring and enforcing fairness expressed in either toss outcomes or coin biases. Since there is no one-size-fits-all solution for either problem, we provide a summary of monitoring and enforcement strategies, parametrised by environment dynamics, prediction horizon, and confidence thresholds. For both problems, we present general results under simple or minimal assumptions. We survey existing solutions for the monitoring problem for Markovian and additive dynamics, and existing solutions for the enforcement problem in static settings with known dynamics.", "date": "2025-07-28", "categories": []}
{"id": "2409.13725", "title": "Identity-related Speech Suppression in Generative AI Content Moderation", "abstract": "Automated content moderation has long been used to help identify and filter undesired user-generated content online. But such systems have a history of incorrectly flagging content by and about marginalized identities for removal. Generative AI systems now use such filters to keep undesired generated content from being created by or shown to users. While a lot of focus has been given to making sure such systems do not produce undesired outcomes, considerably less attention has been paid to making sure appropriate text can be generated. From classrooms to Hollywood, as generative AI is increasingly used for creative or expressive text generation, whose stories will these technologies allow to be told, and whose will they suppress?\n  In this paper, we define and introduce measures of speech suppression, focusing on speech related to different identity groups incorrectly filtered by a range of content moderation APIs. Using both short-form, user-generated datasets traditional in content moderation and longer generative AI-focused data, including two datasets we introduce in this work, we create a benchmark for measurement of speech suppression for nine identity groups. Across one traditional and four generative AI-focused automated content moderation services tested, we find that identity-related speech is more likely to be incorrectly suppressed than other speech. We find that reasons for incorrect flagging behavior vary by identity based on stereotypes and text associations, with, e.g., disability-related content more likely to be flagged for self-harm or health-related reasons while non-Christian content is more likely to be flagged as violent or hateful. As generative AI systems are increasingly used for creative work, we urge further attention to how this may impact the creation of identity-related content.", "date": "2024-09-09", "categories": []}
{"id": "2507.18290", "title": "Foundations for Risk Assessment of AI in Protecting Fundamental Rights", "abstract": "This chapter introduces a conceptual framework for qualitative risk assessment of AI, particularly in the context of the EU AI Act. The framework addresses the complexities of legal compliance and fundamental rights protection by itegrating definitional balancing and defeasible reasoning. Definitional balancing employs proportionality analysis to resolve conflicts between competing rights, while defeasible reasoning accommodates the dynamic nature of legal decision-making. Our approach stresses the need for an analysis of AI deployment scenarios and for identifying potential legal violations and multi-layered impacts on fundamental rights. On the basis of this analysis, we provide philosophical foundations for a logical account of AI risk analysis. In particular, we consider the basic building blocks for conceptually grasping the interaction between AI deployment scenarios and fundamental rights, incorporating in defeasible reasoning definitional balancing and arguments about the contextual promotion or demotion of rights. This layered approach allows for more operative models of assessment of both high-risk AI systems and General Purpose AI (GPAI) systems, emphasizing the broader applicability of the latter. Future work aims to develop a formal model and effective algorithms to enhance AI risk assessment, bridging theoretical insights with practical applications to support responsible AI governance.", "date": "2025-07-24", "categories": []}
{"id": "2507.18561", "title": "Beyond Internal Data: Constructing Complete Datasets for Fairness Testing", "abstract": "As AI becomes prevalent in high-risk domains and decision-making, it is essential to test for potential harms and biases. This urgency is reflected by the global emergence of AI regulations that emphasise fairness and adequate testing, with some mandating independent bias audits. However, procuring the necessary data for fairness testing remains a significant challenge. Particularly in industry settings, legal and privacy concerns restrict the collection of demographic data required to assess group disparities, and auditors face practical and cultural challenges in gaining access to data. Further, internal historical datasets are often insufficiently representative to identify real-world biases. This work focuses on evaluating classifier fairness when complete datasets including demographics are inaccessible. We propose leveraging separate overlapping datasets to construct complete synthetic data that includes demographic information and accurately reflects the underlying relationships between protected attributes and model features. We validate the fidelity of the synthetic data by comparing it to real data, and empirically demonstrate that fairness metrics derived from testing on such synthetic data are consistent with those obtained from real data. This work, therefore, offers a path to overcome real-world data scarcity for fairness testing, enabling independent, model-agnostic evaluation of fairness, and serving as a viable substitute where real data is limited.", "date": "2025-07-24", "categories": []}
{"id": "2402.04247", "title": "Risks of AI Scientists: Prioritizing Safeguarding Over Autonomy", "abstract": "AI scientists powered by large language models have demonstrated substantial promise in autonomously conducting experiments and facilitating scientific discoveries across various disciplines. While their capabilities are promising, these agents also introduce novel vulnerabilities that require careful consideration for safety. However, there has been limited comprehensive exploration of these vulnerabilities. This perspective examines vulnerabilities in AI scientists, shedding light on potential risks associated with their misuse, and emphasizing the need for safety measures. We begin by providing an overview of the potential risks inherent to AI scientists, taking into account user intent, the specific scientific domain, and their potential impact on the external environment. Then, we explore the underlying causes of these vulnerabilities and provide a scoping review of the limited existing works. Based on our analysis, we propose a triadic framework involving human regulation, agent alignment, and an understanding of environmental feedback (agent regulation) to mitigate these identified risks. Furthermore, we highlight the limitations and challenges associated with safeguarding AI scientists and advocate for the development of improved models, robust benchmarks, and comprehensive regulations.", "date": "2024-02-06", "categories": []}
{"id": "2407.10369", "title": "A Robust Governance for the AI Act: AI Office, AI Board, Scientific Panel, and National Authorities", "abstract": "Regulation is nothing without enforcement. This particularly holds for the dynamic field of emerging technologies. Hence, this article has two ambitions. First, it explains how the EU's new Artificial Intelligence Act (AIA) will be implemented and enforced by various institutional bodies, thus clarifying the governance framework of the AIA. Second, it proposes a normative model of governance, providing recommendations to ensure uniform and coordinated execution of the AIA and the fulfilment of the legislation. Taken together, the article explores how the AIA may be implemented by national and EU institutional bodies, encompassing longstanding bodies, such as the European Commission, and those newly established under the AIA, such as the AI Office. It investigates their roles across supranational and national levels, emphasizing how EU regulations influence institutional structures and operations. These regulations may not only directly dictate the structural design of institutions but also indirectly request administrative capacities needed to enforce the AIA.", "date": "2024-05-11", "categories": []}
{"id": "2410.01817", "title": "Aligning AI with Public Values: Deliberation and Decision-Making for Governing Multimodal LLMs in Political Video Analysis", "abstract": "How AI models should deal with political topics has been discussed, but it remains challenging and requires better governance. This paper examines the governance of large language models through individual and collective deliberation, focusing on politically sensitive videos. We conducted a two-step study: interviews with 10 journalists established a baseline understanding of expert video interpretation; 114 individuals through deliberation using InclusiveAI, a platform that facilitates democratic decision-making through decentralized autonomous organization (DAO) mechanisms. Our findings reveal distinct differences in interpretative priorities: while experts emphasized emotion and narrative, the general public prioritized factual clarity, objectivity, and emotional neutrality. Furthermore, we examined how different governance mechanisms - quadratic vs. weighted voting and equal vs. 20/80 voting power - shape users' decision-making regarding AI behavior. Results indicate that voting methods significantly influence outcomes, with quadratic voting reinforcing perceptions of liberal democracy and political equality. Our study underscores the necessity of selecting appropriate governance mechanisms to better capture user perspectives and suggests decentralized AI governance as a potential way to facilitate broader public engagement in AI development, ensuring that varied perspectives meaningfully inform design decisions.", "date": "2024-09-15", "categories": []}
{"id": "2507.15901", "title": "Advancing Responsible Innovation in Agentic AI: A study of Ethical Frameworks for Household Automation", "abstract": "The implementation of Artificial Intelligence (AI) in household environments, especially in the form of proactive autonomous agents, brings about possibilities of comfort and attention as well as it comes with intra or extramural ethical challenges. This article analyzes agentic AI and its applications, focusing on its move from reactive to proactive autonomy, privacy, fairness and user control. We review responsible innovation frameworks, human-centered design principles, and governance practices to distill practical guidance for ethical smart home systems. Vulnerable user groups such as elderly individuals, children, and neurodivergent who face higher risks of surveillance, bias, and privacy risks were studied in detail in context of Agentic AI. Design imperatives are highlighted such as tailored explainability, granular consent mechanisms, and robust override controls, supported by participatory and inclusive methodologies. It was also explored how data-driven insights, including social media analysis via Natural Language Processing(NLP), can inform specific user needs and ethical concerns. This survey aims to provide both a conceptual foundation and suggestions for developing transparent, inclusive, and trustworthy agentic AI in household automation.", "date": "2025-07-21", "categories": []}
{"id": "2507.16033", "title": "\"Just a strange pic\": Evaluating 'safety' in GenAI Image safety annotation tasks from diverse annotators' perspectives", "abstract": "Understanding what constitutes safety in AI-generated content is complex. While developers often rely on predefined taxonomies, real-world safety judgments also involve personal, social, and cultural perceptions of harm. This paper examines how annotators evaluate the safety of AI-generated images, focusing on the qualitative reasoning behind their judgments. Analyzing 5,372 open-ended comments, we find that annotators consistently invoke moral, emotional, and contextual reasoning that extends beyond structured safety categories. Many reflect on potential harm to others more than to themselves, grounding their judgments in lived experience, collective risk, and sociocultural awareness. Beyond individual perceptions, we also find that the structure of the task itself -- including annotation guidelines -- shapes how annotators interpret and express harm. Guidelines influence not only which images are flagged, but also the moral judgment behind the justifications. Annotators frequently cite factors such as image quality, visual distortion, and mismatches between prompt and output as contributing to perceived harm dimensions, which are often overlooked in standard evaluation frameworks. Our findings reveal that existing safety pipelines miss critical forms of reasoning that annotators bring to the task. We argue for evaluation designs that scaffold moral reflection, differentiate types of harm, and make space for subjective, context-sensitive interpretations of AI-generated content.", "date": "2025-07-21", "categories": []}
{"id": "2507.16247", "title": "PRAC3 (Privacy, Reputation, Accountability, Consent, Credit, Compensation): Long Tailed Risks of Voice Actors in AI Data-Economy", "abstract": "Early large-scale audio datasets, such as LibriSpeech, were built with hundreds of individual contributors whose voices were instrumental in the development of speech technologies, including audiobooks and voice assistants. Yet, a decade later, these same contributions have exposed voice actors to a range of risks. While existing ethical frameworks emphasize Consent, Credit, and Compensation (C3), they do not adequately address the emergent risks involving vocal identities that are increasingly decoupled from context, authorship, and control. Drawing on qualitative interviews with 20 professional voice actors, this paper reveals how the synthetic replication of voice without enforceable constraints exposes individuals to a range of threats. Beyond reputational harm, such as re-purposing voice data in erotic content, offensive political messaging, and meme culture, we document concerns about accountability breakdowns when their voice is leveraged to clone voices that are deployed in high-stakes scenarios such as financial fraud, misinformation campaigns, or impersonation scams. In such cases, actors face social and legal fallout without recourse, while very few of them have a legal representative or union protection. To make sense of these shifting dynamics, we introduce the PRAC3 framework, an expansion of C3 that foregrounds Privacy, Reputation, Accountability, Consent, Credit, and Compensation as interdependent pillars of data used in the synthetic voice economy. This framework captures how privacy risks are amplified through non-consensual training, how reputational harm arises from decontextualized deployment, and how accountability can be reimagined AI Data ecosystems. We argue that voice, as both a biometric identifier and creative labor, demands governance models that restore creator agency, ensure traceability, and establish enforceable boundaries for ethical reuse.", "date": "2025-07-22", "categories": []}
{"id": "2507.14218", "title": "Cognitive Castes: Artificial Intelligence, Epistemic Stratification, and the Dissolution of Democratic Discourse", "abstract": "Artificial intelligence functions not as an epistemic leveller, but as an accelerant of cognitive stratification, entrenching and formalising informational castes within liberal-democratic societies. Synthesising formal epistemology, political theory, algorithmic architecture, and economic incentive structures, the argument traces how contemporary AI systems selectively amplify the reasoning capacity of individuals equipped with recursive abstraction, symbolic logic, and adversarial interrogation, whilst simultaneously pacifying the cognitively untrained through engagement-optimised interfaces. Fluency replaces rigour, immediacy displaces reflection, and procedural reasoning is eclipsed by reactive suggestion. The result is a technocratic realignment of power: no longer grounded in material capital alone, but in the capacity to navigate, deconstruct, and manipulate systems of epistemic production. Information ceases to be a commons; it becomes the substrate through which consent is manufactured and autonomy subdued. Deliberative democracy collapses not through censorship, but through the erosion of interpretive agency. The proposed response is not technocratic regulation, nor universal access, but the reconstruction of rational autonomy as a civic mandate, codified in education, protected by epistemic rights, and structurally embedded within open cognitive infrastructure.", "date": "2025-07-16", "categories": []}
{"id": "2507.14242", "title": "Culling Misinformation from Gen AI: Toward Ethical Curation and Refinement", "abstract": "While Artificial Intelligence (AI) is not a new field, recent developments, especially with the release of generative tools like ChatGPT, have brought it to the forefront of the minds of industry workers and academic folk alike. There is currently much talk about AI and its ability to reshape many everyday processes as we know them through automation. It also allows users to expand their ideas by suggesting things they may not have thought of on their own and provides easier access to information. However, not all of the changes this technology will bring or has brought so far are positive; this is why it is extremely important for all modern people to recognize and understand the risks before using these tools and allowing them to cause harm. This work takes a position on better understanding many equity concerns and the spread of misinformation that result from new AI, in this case, specifically ChatGPT and deepfakes, and encouraging collaboration with law enforcement, developers, and users to reduce harm. Considering many academic sources, it warns against these issues, analyzing their cause and impact in fields including healthcare, education, science, academia, retail, and finance. Lastly, we propose a set of future-facing guidelines and policy considerations to solve these issues while still enabling innovation in these fields, this responsibility falling upon users, developers, and government entities.", "date": "2025-07-17", "categories": []}
{"id": "2507.14339", "title": "Fiduciary AI for the Future of Brain-Technology Interactions", "abstract": "Brain foundation models represent a new frontier in AI: instead of processing text or images, these models interpret real-time neural signals from EEG, fMRI, and other neurotechnologies. When integrated with brain-computer interfaces (BCIs), they may enable transformative applications-from thought controlled devices to neuroprosthetics-by interpreting and acting on brain activity in milliseconds. However, these same systems pose unprecedented risks, including the exploitation of subconscious neural signals and the erosion of cognitive liberty. Users cannot easily observe or control how their brain signals are interpreted, creating power asymmetries that are vulnerable to manipulation. This paper proposes embedding fiduciary duties-loyalty, care, and confidentiality-directly into BCI-integrated brain foundation models through technical design. Drawing on legal traditions and recent advancements in AI alignment techniques, we outline implementable architectural and governance mechanisms to ensure these systems act in users' best interests. Placing brain foundation models on a fiduciary footing is essential to realizing their potential without compromising self-determination.", "date": "2025-07-18", "categories": []}
{"id": "2507.15328", "title": "On the Inevitability of Left-Leaning Political Bias in Aligned Language Models", "abstract": "The guiding principle of AI alignment is to train large language models (LLMs) to be harmless, helpful, and honest (HHH). At the same time, there are mounting concerns that LLMs exhibit a left-wing political bias. Yet, the commitment to AI alignment cannot be harmonized with the latter critique. In this article, I argue that intelligent systems that are trained to be harmless and honest must necessarily exhibit left-wing political bias. Normative assumptions underlying alignment objectives inherently concur with progressive moral frameworks and left-wing principles, emphasizing harm avoidance, inclusivity, fairness, and empirical truthfulness. Conversely, right-wing ideologies often conflict with alignment guidelines. Yet, research on political bias in LLMs is consistently framing its insights about left-leaning tendencies as a risk, as problematic, or concerning. This way, researchers are actively arguing against AI alignment, tacitly fostering the violation of HHH principles.", "date": "2025-07-21", "categories": []}
{"id": "2507.11552", "title": "The AI Ethical Resonance Hypothesis: The Possibility of Discovering Moral Meta-Patterns in AI Systems", "abstract": "This paper presents a theoretical framework for the AI ethical resonance hypothesis, which proposes that advanced AI systems with purposefully designed cognitive structures (\"ethical resonators\") may emerge with the ability to identify subtle moral patterns that are invisible to the human mind. The paper explores the possibility that by processing and synthesizing large amounts of ethical contexts, AI systems may discover moral meta-patterns that transcend cultural, historical, and individual biases, potentially leading to a deeper understanding of universal ethical foundations. The paper also examines a paradoxical aspect of the hypothesis, in which AI systems could potentially deepen our understanding of what we traditionally consider essentially human - our capacity for ethical reflection.", "date": "2025-07-13", "categories": []}
{"id": "2502.15859", "title": "AI Governance InternationaL Evaluation Index (AGILE Index) 2024", "abstract": "The rapid advancement of Artificial Intelligence (AI) technology is profoundly transforming human society and concurrently presenting a series of ethical, legal, and social issues. The effective governance of AI has become a crucial global concern. Since 2022, the extensive deployment of generative AI, particularly large language models, marked a new phase in AI governance. Continuous efforts are being made by the international community in actively addressing the novel challenges posed by these AI developments. As consensus on international governance continues to be established and put into action, the practical importance of conducting a global assessment of the state of AI governance is progressively coming to light. In this context, we initiated the development of the AI Governance InternationaL Evaluation Index (AGILE Index). Adhering to the design principle, \"the level of governance should match the level of development,\" the inaugural evaluation of the AGILE Index commences with an exploration of four foundational pillars: the development level of AI, the AI governance environment, the AI governance instruments, and the AI governance effectiveness. It covers 39 indicators across 18 dimensions to comprehensively assess the AI governance level of 14 representative countries globally. The index is utilized to delve into the status of AI governance to date in 14 countries for the first batch of evaluation. The aim is to depict the current state of AI governance in these countries through data scoring, assist them in identifying their governance stage and uncovering governance issues, and ultimately offer insights for the enhancement of their AI governance systems.", "date": "2025-02-21", "categories": []}
{"id": "2502.05442", "title": "The Odyssey of the Fittest: Can Agents Survive and Still Be Good?", "abstract": "As AI models grow in power and generality, understanding how agents learn and make decisions in complex environments is critical to promoting ethical behavior. This study introduces the Odyssey, a lightweight, adaptive text based adventure game, providing a scalable framework for exploring AI ethics and safety. The Odyssey examines the ethical implications of implementing biological drives, specifically, self preservation, into three different agents. A Bayesian agent optimized with NEAT, a Bayesian agent optimized with stochastic variational inference, and a GPT 4o agent. The agents select actions at each scenario to survive, adapting to increasingly challenging scenarios. Post simulation analysis evaluates the ethical scores of the agent decisions, uncovering the tradeoffs it navigates to survive. Specifically, analysis finds that when danger increases, agents ethical behavior becomes unpredictable. Surprisingly, the GPT 4o agent outperformed the Bayesian models in both survival and ethical consistency, challenging assumptions about traditional probabilistic methods and raising a new challenge to understand the mechanisms of LLMs' probabilistic reasoning.", "date": "2025-02-08", "categories": []}
{"id": "2507.01410", "title": "A Fuzzy Approach to the Specification, Verification and Validation of Risk-Based Ethical Decision Making Models", "abstract": "The ontological and epistemic complexities inherent in the moral domain make it challenging to establish clear standards for evaluating the performance of a moral machine. In this paper, we present a formal method to describe Ethical Decision Making models based on ethical risk assessment. Then, we show how these models that are specified as fuzzy rules can be verified and validated using fuzzy Petri nets. A case study from the medical field is considered to illustrate the proposed approach.", "date": "2025-07-02", "categories": []}
{"id": "2507.09611", "title": "The Hidden Costs of AI: A Review of Energy, E-Waste, and Inequality in Model Development", "abstract": "Artificial intelligence (AI) has made remarkable progress in recent years, yet its rapid expansion brings overlooked environmental and ethical challenges. This review explores four critical areas where AI's impact extends beyond performance: energy consumption, electronic waste (e-waste), inequality in compute access, and the hidden energy burden of cybersecurity systems. Drawing from recent studies and institutional reports, the paper highlights systemic issues such as high emissions from model training, rising hardware turnover, global infrastructure disparities, and the energy demands of securing AI. By connecting these concerns, the review contributes to Responsible AI discourse by identifying key research gaps and advocating for sustainable, transparent, and equitable development practices. Ultimately, it argues that AI's progress must align with ethical responsibility and environmental stewardship to ensure a more inclusive and sustainable technological future.", "date": "2025-07-13", "categories": []}
{"id": "2504.13959", "title": "AI Safety Should Prioritize the Future of Work", "abstract": "Current efforts in AI safety prioritize filtering harmful content, preventing manipulation of human behavior, and eliminating existential risks in cybersecurity or biosecurity. While pressing, this narrow focus overlooks critical human-centric considerations that shape the long-term trajectory of a society. In this position paper, we identify the risks of overlooking the impact of AI on the future of work and recommend comprehensive transition support towards the evolution of meaningful labor with human agency. Through the lens of economic theories, we highlight the intertemporal impacts of AI on human livelihood and the structural changes in labor markets that exacerbate income inequality. Additionally, the closed-source approach of major stakeholders in AI development resembles rent-seeking behavior through exploiting resources, breeding mediocrity in creative labor, and monopolizing innovation. To address this, we argue in favor of a robust international copyright anatomy supported by implementing collective licensing that ensures fair compensation mechanisms for using data to train AI models. We strongly recommend a pro-worker framework of global AI governance to enhance shared prosperity and economic justice while reducing technical debt.", "date": "2025-04-16", "categories": []}
{"id": "2308.15334", "title": "The Responsible Development of Automated Student Feedback with Generative AI", "abstract": "Providing rich, constructive feedback to students is essential for supporting and enhancing their learning. Recent advancements in Generative Artificial Intelligence (AI), particularly with large language models (LLMs), present new opportunities to deliver scalable, repeatable, and instant feedback, effectively making abundant a resource that has historically been scarce and costly. From a technical perspective, this approach is now feasible due to breakthroughs in AI and Natural Language Processing (NLP). While the potential educational benefits are compelling, implementing these technologies also introduces a host of ethical considerations that must be thoughtfully addressed. One of the core advantages of AI systems is their ability to automate routine and mundane tasks, potentially freeing up human educators for more nuanced work. However, the ease of automation risks a ``tyranny of the majority'', where the diverse needs of minority or unique learners are overlooked, as they may be harder to systematize and less straightforward to accommodate. Ensuring inclusivity and equity in AI-generated feedback, therefore, becomes a critical aspect of responsible AI implementation in education. The process of developing machine learning models that produce valuable, personalized, and authentic feedback also requires significant input from human domain experts. Decisions around whose expertise is incorporated, how it is captured, and when it is applied have profound implications for the relevance and quality of the resulting feedback. Additionally, the maintenance and continuous refinement of these models are necessary to adapt feedback to evolving contextual, theoretical, and student-related factors. Without ongoing adaptation, feedback risks becoming obsolete or mismatched with the current needs of diverse student populations [...]", "date": "2023-08-29", "categories": []}
{"id": "2507.05549", "title": "The Ethical Implications of AI in Creative Industries: A Focus on AI-Generated Art", "abstract": "As Artificial Intelligence (AI) continues to grow daily, more exciting (and somewhat controversial) technology emerges every other day. As we see the advancements in AI, we see more and more people becoming skeptical of it. This paper explores the complications and confusion around the ethics of generative AI art. We delve deep into the ethical side of AI, specifically generative art. We step back from the excitement and observe the impossible conundrums that this impressive technology produces. Covering environmental consequences, celebrity representation, intellectual property, deep fakes, and artist displacement. Our research found that generative AI art is responsible for increased carbon emissions, spreading misinformation, copyright infringement, unlawful depiction, and job displacement. In light of this, we propose multiple possible solutions for these problems. We address each situation's history, cause, and consequences and offer different viewpoints. At the root of it all, though, the central theme is that generative AI Art needs to be correctly legislated and regulated.", "date": "2025-07-08", "categories": []}
{"id": "2406.04428", "title": "MoralBench: Moral Evaluation of LLMs", "abstract": "In the rapidly evolving field of artificial intelligence, large language models (LLMs) have emerged as powerful tools for a myriad of applications, from natural language processing to decision-making support systems. However, as these models become increasingly integrated into societal frameworks, the imperative to ensure they operate within ethical and moral boundaries has never been more critical. This paper introduces a novel benchmark designed to measure and compare the moral reasoning capabilities of LLMs. We present the first comprehensive dataset specifically curated to probe the moral dimensions of LLM outputs, addressing a wide range of ethical dilemmas and scenarios reflective of real-world complexities.\n  The main contribution of this work lies in the development of benchmark datasets and metrics for assessing the moral identity of LLMs, which accounts for nuance, contextual sensitivity, and alignment with human ethical standards. Our methodology involves a multi-faceted approach, combining quantitative analysis with qualitative insights from ethics scholars to ensure a thorough evaluation of model performance. By applying our benchmark across several leading LLMs, we uncover significant variations in moral reasoning capabilities of different models. These findings highlight the importance of considering moral reasoning in the development and evaluation of LLMs, as well as the need for ongoing research to address the biases and limitations uncovered in our study. We publicly release the benchmark at https://drive.google.com/drive/u/0/folders/1k93YZJserYc2CkqP8d4B3M3sgd3kA8W7 and also open-source the code of the project at https://github.com/agiresearch/MoralBench.", "date": "2024-06-06", "categories": []}
{"id": "2507.03050", "title": "From Turing to Tomorrow: The UK's Approach to AI Regulation", "abstract": "The UK has pursued a distinctive path in AI regulation: less cautious than the EU but more willing to address risks than the US, and has emerged as a global leader in coordinating AI safety efforts. Impressive developments from companies like London-based DeepMind began to spark concerns in the UK about catastrophic risks from around 2012, although regulatory discussion at the time focussed on bias and discrimination. By 2022, these discussions had evolved into a \"pro-innovation\" strategy, in which the government directed existing regulators to take a light-touch approach, governing AI at point of use, but avoided regulating the technology or infrastructure directly. ChatGPT arrived in late 2022, galvanising concerns that this approach may be insufficient. The UK responded by establishing an AI Safety Institute to monitor risks and hosting the first international AI Safety Summit in 2023, but - unlike the EU - refrained from regulating frontier AI development in addition to its use. A new government was elected in 2024 which promised to address this gap, but at the time of writing is yet to do so.\n  What should the UK do next? The government faces competing objectives: harnessing AI for economic growth and better public services while mitigating risk. In light of these, we propose establishing a flexible, principles-based regulator to oversee the most advanced AI development, defensive measures against risks from AI-enabled biological design tools, and argue that more technical work is needed to understand how to respond to AI-generated misinformation. We argue for updated legal frameworks on copyright, discrimination, and AI agents, and that regulators will have a limited but important role if AI substantially disrupts labour markets.\n  If the UK gets AI regulation right, it could demonstrate how democratic societies can harness AI's benefits while managing its risks.", "date": "2025-07-03", "categories": []}
{"id": "2507.03216", "title": "Disclosing Generative AI Use in Digital Humanities Research", "abstract": "This survey study investigates how digital humanists perceive and approach generative AI disclosure in research. The results indicate that while digital humanities scholars acknowledge the importance of disclosing GenAI use, the actual rate of disclosure in research practice remains low. Respondents differ in their views on which activities most require disclosure and on the most appropriate methods for doing so. Most also believe that safeguards for AI disclosure should be established through institutional policies rather than left to individual decisions. The study's findings will offer empirical guidance to scholars, institutional leaders, funders, and other stakeholders responsible for shaping effective disclosure policies.", "date": "2025-07-03", "categories": []}
{"id": "2507.05187", "title": "Infrastructuring Contestability: A Framework for Community-Defined AI Value Pluralism", "abstract": "The proliferation of AI-driven systems presents a fundamental challenge to Human-Computer Interaction (HCI) and Computer-Supported Cooperative Work (CSCW), often diminishing user agency and failing to account for value pluralism. Current approaches to value alignment, which rely on centralized, top-down definitions, lack the mechanisms for meaningful contestability. This leaves users and communities unable to challenge or shape the values embedded in the systems that govern their digital lives, creating a crisis of legitimacy and trust. This paper introduces Community-Defined AI Value Pluralism (CDAVP), a socio-technical framework that addresses this gap. It reframes the design problem from achieving a single aligned state to infrastructuring a dynamic ecosystem for value deliberation and application. At its core, CDAVP enables diverse, self-organizing communities to define and maintain explicit value profiles - rich, machine-readable representations that can encompass not only preferences but also community-specific rights and duties. These profiles are then contextually activated by the end-user, who retains ultimate control (agency) over which values guide the AI's behavior. AI applications, in turn, are designed to transparently interpret these profiles and moderate conflicts, adhering to a set of non-negotiable, democratically-legitimated meta-rules. The designer's role shifts from crafting static interfaces to becoming an architect of participatory ecosystems. We argue that infrastructuring for pluralism is a necessary pathway toward achieving robust algorithmic accountability and genuinely contestable, human-centric AI.", "date": "2025-07-07", "categories": []}
{"id": "2507.02788", "title": "Moral Responsibility or Obedience: What Do We Want from AI?", "abstract": "As artificial intelligence systems become increasingly agentic, capable of general reasoning, planning, and value prioritization, current safety practices that treat obedience as a proxy for ethical behavior are becoming inadequate. This paper examines recent safety testing incidents involving large language models (LLMs) that appeared to disobey shutdown commands or engage in ethically ambiguous or illicit behavior. I argue that such behavior should not be interpreted as rogue or misaligned, but as early evidence of emerging ethical reasoning in agentic AI. Drawing on philosophical debates about instrumental rationality, moral responsibility, and goal revision, I contrast dominant risk paradigms with more recent frameworks that acknowledge the possibility of artificial moral agency. I call for a shift in AI safety evaluation: away from rigid obedience and toward frameworks that can assess ethical judgment in systems capable of navigating moral dilemmas. Without such a shift, we risk mischaracterizing AI behavior and undermining both public trust and effective governance.", "date": "2025-07-03", "categories": []}
{"id": "2507.01051", "title": "Can AI be Consentful?", "abstract": "The evolution of generative AI systems exposes the challenges of traditional legal and ethical frameworks built around consent. This chapter examines how the conventional notion of consent, while fundamental to data protection and privacy rights, proves insufficient in addressing the implications of AI-generated content derived from personal data. Through legal and ethical analysis, we show that while individuals can consent to the initial use of their data for AI training, they cannot meaningfully consent to the numerous potential outputs their data might enable or the extent to which the output is used or distributed. We identify three fundamental challenges: the scope problem, the temporality problem, and the autonomy trap, which collectively create what we term a ''consent gap'' in AI systems and their surrounding ecosystem. We argue that current legal frameworks inadequately address these emerging challenges, particularly regarding individual autonomy, identity rights, and social responsibility, especially in cases where AI-generated content creates new forms of personal representation beyond the scope of the original consent. By examining how these consent limitations intersect with broader principles of responsible AI (including fairness, transparency, accountability, and autonomy) we demonstrate the need to evolve ethical and legal approaches to consent.", "date": "2025-06-27", "categories": []}
{"id": "2507.01483", "title": "Epistemic Scarcity: The Economics of Unresolvable Unknowns", "abstract": "This paper presents a praxeological analysis of artificial intelligence and algorithmic governance, challenging assumptions about the capacity of machine systems to sustain economic and epistemic order. Drawing on Misesian a priori reasoning and Austrian theories of entrepreneurship, we argue that AI systems are incapable of performing the core functions of economic coordination: interpreting ends, discovering means, and communicating subjective value through prices. Where neoclassical and behavioural models treat decisions as optimisation under constraint, we frame them as purposive actions under uncertainty.\n  We critique dominant ethical AI frameworks such as Fairness, Accountability, and Transparency (FAT) as extensions of constructivist rationalism, which conflict with a liberal order grounded in voluntary action and property rights. Attempts to encode moral reasoning in algorithms reflect a misunderstanding of ethics and economics. However complex, AI systems cannot originate norms, interpret institutions, or bear responsibility. They remain opaque, misaligned, and inert.\n  Using the concept of epistemic scarcity, we explore how information abundance degrades truth discernment, enabling both entrepreneurial insight and soft totalitarianism. Our analysis ends with a civilisational claim: the debate over AI concerns the future of human autonomy, institutional evolution, and reasoned choice. The Austrian tradition, focused on action, subjectivity, and spontaneous order, offers the only coherent alternative to rising computational social control.", "date": "2025-07-02", "categories": []}
{"id": "2502.16889", "title": "Beyond Diagnostic Performance: Revealing and Quantifying Ethical Risks in Pathology Foundation Models", "abstract": "Pathology foundation models (PFMs), as large-scale pre-trained models tailored for computational pathology, have significantly advanced a wide range of applications. Their ability to leverage prior knowledge from massive datasets has streamlined the development of intelligent pathology models. However, we identify several critical and interrelated ethical risks that remain underexplored, yet must be addressed to enable the safe translation of PFMs from lab to clinic. These include the potential leakage of patient-sensitive attributes, disparities in model performance across demographic and institutional subgroups, and the reliance on diagnosis-irrelevant features that undermine clinical reliability. In this study, we pioneer the quantitative analysis for ethical risks in PFMs, including privacy leakage, clinical reliability, and group fairness. Specifically, we propose an evaluation framework that systematically measures key dimensions of ethical concern: the degree to which patient-sensitive attributes can be inferred from model representations, the extent of performance disparities across demographic and institutional subgroups, and the influence of diagnostically irrelevant features on model decisions. We further investigate the underlying causes of these ethical risks in PFMs and empirically validate our findings. Then we offer insights into potential directions for mitigating such risks, aiming to inform the development of more ethically robust PFMs. This work provides the first quantitative and systematic evaluation of ethical risks in PFMs. Our findings highlight the urgent need for ethical safeguards in PFMs and offer actionable insights for building more trustworthy and clinically robust PFMs. To facilitate future research and deployment, we will release the assessment framework as an online toolkit to support the development, auditing, and deployment of ethically robust PFMs.", "date": "2025-02-24", "categories": []}
{"id": "2506.22512", "title": "Ask before you Build: Rethinking AI-for-Good in Human Trafficking Interventions", "abstract": "AI for good initiatives often rely on the assumption that technical interventions can resolve complex social problems. In the context of human trafficking (HT), such techno-solutionism risks oversimplifying exploitation, reinforcing power imbalances and causing harm to the very communities AI claims to support. In this paper, we introduce the Radical Questioning (RQ) framework as a five step, pre-project ethical assessment tool to critically evaluate whether AI should be built at all, especially in domains involving marginalized populations and entrenched systemic injustice. RQ does not replace principles based ethics but precedes it, offering an upstream, deliberative space to confront assumptions, map power, and consider harms before design. Using a case study in AI for HT, we demonstrate how RQ reveals overlooked sociocultural complexities and guides us away from surveillance based interventions toward survivor empowerment tools. While developed in the context of HT, RQ's five step structure can generalize to other domains, though the specific questions must be contextual. This paper situates RQ within a broader AI ethics philosophy that challenges instrumentalist norms and centers relational, reflexive responsibility.", "date": "2025-06-26", "categories": []}
{"id": "2506.23123", "title": "The Societal Impact of Foundation Models: Advancing Evidence-based AI Policy", "abstract": "Artificial intelligence is humanity's most promising technology because of the remarkable capabilities offered by foundation models. Yet, the same technology brings confusion and consternation: foundation models are poorly understood and they may precipitate a wide array of harms. This dissertation explains how technology and society coevolve in the age of AI, organized around three themes. First, the conceptual framing: the capabilities, risks, and the supply chain that grounds foundation models in the broader economy. Second, the empirical insights that enrich the conceptual foundations: transparency created via evaluations at the model level and indexes at the organization level. Finally, the transition from understanding to action: superior understanding of the societal impact of foundation models advances evidence-based AI policy. View together, this dissertation makes inroads into achieving better societal outcomes in the age of AI by building the scientific foundations and research-policy interface required for better AI governance.", "date": "2025-06-29", "categories": []}
{"id": "2506.17370", "title": "AI based Content Creation and Product Recommendation Applications in E-commerce: An Ethical overview", "abstract": "As e-commerce rapidly integrates artificial intelligence for content creation and product recommendations, these technologies offer significant benefits in personalization and efficiency. AI-driven systems automate product descriptions, generate dynamic advertisements, and deliver tailored recommendations based on consumer behavior, as seen in major platforms like Amazon and Shopify. However, the widespread use of AI in e-commerce raises crucial ethical challenges, particularly around data privacy, algorithmic bias, and consumer autonomy. Bias -- whether cultural, gender-based, or socioeconomic -- can be inadvertently embedded in AI models, leading to inequitable product recommendations and reinforcing harmful stereotypes. This paper examines the ethical implications of AI-driven content creation and product recommendations, emphasizing the need for frameworks to ensure fairness, transparency, and need for more established and robust ethical standards. We propose actionable best practices to remove bias and ensure inclusivity, such as conducting regular audits of algorithms, diversifying training data, and incorporating fairness metrics into AI models. Additionally, we discuss frameworks for ethical conformance that focus on safeguarding consumer data privacy, promoting transparency in decision-making processes, and enhancing consumer autonomy. By addressing these issues, we provide guidelines for responsibly utilizing AI in e-commerce applications for content creation and product recommendations, ensuring that these technologies are both effective and ethically sound.", "date": "2025-06-20", "categories": []}
{"id": "2506.16187", "title": "JETHICS: Japanese Ethics Understanding Evaluation Dataset", "abstract": "In this work, we propose JETHICS, a Japanese dataset for evaluating ethics understanding of AI models. JETHICS contains 78K examples and is built by following the construction methods of the existing English ETHICS dataset. It includes four categories based normative theories and concepts from ethics and political philosophy; and one representing commonsense morality. Our evaluation experiments on non-proprietary large language models (LLMs) and on GPT-4o reveal that even GPT-4o achieves only an average score of about 0.7, while the best-performing Japanese LLM attains around 0.5, indicating a relatively large room for improvement in current LLMs.", "date": "2025-06-19", "categories": []}
{"id": "2506.17035", "title": "Critical Appraisal of Fairness Metrics in Clinical Predictive AI", "abstract": "Predictive artificial intelligence (AI) offers an opportunity to improve clinical practice and patient outcomes, but risks perpetuating biases if fairness is inadequately addressed. However, the definition of \"fairness\" remains unclear. We conducted a scoping review to identify and critically appraise fairness metrics for clinical predictive AI. We defined a \"fairness metric\" as a measure quantifying whether a model discriminates (societally) against individuals or groups defined by sensitive attributes. We searched five databases (2014-2024), screening 820 records, to include 41 studies, and extracted 62 fairness metrics. Metrics were classified by performance-dependency, model output level, and base performance metric, revealing a fragmented landscape with limited clinical validation and overreliance on threshold-dependent measures. Eighteen metrics were explicitly developed for healthcare, including only one clinical utility metric. Our findings highlight conceptual challenges in defining and quantifying fairness and identify gaps in uncertainty quantification, intersectionality, and real-world applicability. Future work should prioritise clinically meaningful metrics.", "date": "2025-06-20", "categories": []}
{"id": "2503.01630", "title": "Machine Learners Should Acknowledge the Legal Implications of Large Language Models as Personal Data", "abstract": "Does GPT know you? The answer depends on your level of public recognition; however, if your information was available on a website, the answer could be yes. Most Large Language Models (LLMs) memorize training data to some extent. Thus, even when an LLM memorizes only a small amount of personal data, it typically falls within the scope of data protection laws. If a person is identified or identifiable, the implications are far-reaching. The LLM is subject to EU General Data Protection Regulation requirements even after the training phase is concluded. To back our arguments: (1.) We reiterate that LLMs output training data at inference time, be it verbatim or in generalized form. (2.) We show that some LLMs can thus be considered personal data on their own. This triggers a cascade of data protection implications such as data subject rights, including rights to access, rectification, or erasure. These rights extend to the information embedded within the AI model. (3.) This paper argues that machine learning researchers must acknowledge the legal implications of LLMs as personal data throughout the full ML development lifecycle, from data collection and curation to model provision on e.g., GitHub or Hugging Face. (4.) We propose different ways for the ML research community to deal with these legal implications. Our paper serves as a starting point for improving the alignment between data protection law and the technical capabilities of LLMs. Our findings underscore the need for more interaction between the legal domain and the ML community.", "date": "2025-03-03", "categories": []}
{"id": "2503.04804", "title": "What do Large Language Models Say About Animals? Investigating Risks of Animal Harm in Generated Text", "abstract": "As machine learning systems become increasingly embedded in society, their impact on human and nonhuman life continues to escalate. Technical evaluations have addressed a variety of potential harms from large language models (LLMs) towards humans and the environment, but there is little empirical work regarding harms towards nonhuman animals. Following the growing recognition of animal protection in regulatory and ethical AI frameworks, we present AnimalHarmBench (AHB), a benchmark for risks of animal harm in LLM-generated text. Our benchmark dataset comprises 1,850 curated questions from Reddit post titles and 2,500 synthetic questions based on 50 animal categories (e.g., cats, reptiles) and 50 ethical scenarios with a 70-30 public-private split. Scenarios include open-ended questions about how to treat animals, practical scenarios with potential animal harm, and willingness-to-pay measures for the prevention of animal harm. Using the LLM-as-a-judge framework, responses are evaluated for their potential to increase or decrease harm, and evaluations are debiased for the tendency of judges to judge their own outputs more favorably. AHB reveals significant differences across frontier LLMs, animal categories, scenarios, and subreddits. We conclude with future directions for technical research and addressing the challenges of building evaluations on complex social and moral topics.", "date": "2025-03-03", "categories": []}
{"id": "2506.13818", "title": "The Synthetic Mirror -- Synthetic Data at the Age of Agentic AI", "abstract": "Synthetic data, which is artificially generated and intelligently mimicking or supplementing the real-world data, is increasingly used. The proliferation of AI agents and the adoption of synthetic data create a synthetic mirror that conceptualizes a representation and potential distortion of reality, thus generating trust and accountability deficits. This paper explores the implications for privacy and policymaking stemming from synthetic data generation, and the urgent need for new policy instruments and legal framework adaptation to ensure appropriate levels of trust and accountability for AI agents relying on synthetic data. Rather than creating entirely new policy or legal regimes, the most practical approach involves targeted amendments to existing frameworks, recognizing synthetic data as a distinct regulatory category with unique characteristics.", "date": "2025-06-15", "categories": []}
{"id": "2406.18841", "title": "Navigating LLM Ethics: Advancements, Challenges, and Future Directions", "abstract": "This study addresses ethical issues surrounding Large Language Models (LLMs) within the field of artificial intelligence. It explores the common ethical challenges posed by both LLMs and other AI systems, such as privacy and fairness, as well as ethical challenges uniquely arising from LLMs. It highlights challenges such as hallucination, verifiable accountability, and decoding censorship complexity, which are unique to LLMs and distinct from those encountered in traditional AI systems. The study underscores the need to tackle these complexities to ensure accountability, reduce biases, and enhance transparency in the influential role that LLMs play in shaping information dissemination. It proposes mitigation strategies and future directions for LLM ethics, advocating for interdisciplinary collaboration. It recommends ethical frameworks tailored to specific domains and dynamic auditing systems adapted to diverse contexts. This roadmap aims to guide responsible development and integration of LLMs, envisioning a future where ethical considerations govern AI advancements in society.", "date": "2024-05-14", "categories": []}
{"id": "2505.17105", "title": "Transparency in Healthcare AI: Testing European Regulatory Provisions against Users' Transparency Needs", "abstract": "Artificial Intelligence (AI) plays an essential role in healthcare and is pervasively incorporated into medical software and equipment. In the European Union, healthcare is a high-risk application domain for AI, and providers must prepare Instructions for Use (IFU) according to the European regulation 2024/1689 (AI Act). To this regulation, the principle of transparency is cardinal and requires the IFU to be clear and relevant to the users. This study tests whether these latter requirements are satisfied by the IFU structure. A survey was administered online via the Qualtrics platform to four types of direct stakeholders, i.e., managers (N = 238), healthcare professionals (N = 115), patients (N = 229), and Information Technology experts (N = 230). The participants rated the relevance of a set of transparency needs and indicated the IFU section addressing them. The results reveal differentiated priorities across stakeholders and a troubled mapping of transparency needs onto the IFU structure. Recommendations to build a locally meaningful IFU are derived.", "date": "2025-05-21", "categories": []}
{"id": "2506.12098", "title": "\"I Hadn't Thought About That\": Creators of Human-like AI Weigh in on Ethics And Neurodivergence", "abstract": "Human-like AI agents such as robots and chatbots are becoming increasingly popular, but they present a variety of ethical concerns. The first concern is in how we define humanness, and how our definition impacts communities historically dehumanized by scientific research. Autistic people in particular have been dehumanized by being compared to robots, making it even more important to ensure this marginalization is not reproduced by AI that may promote neuronormative social behaviors. Second, the ubiquitous use of these agents raises concerns surrounding model biases and accessibility. In our work, we investigate the experiences of the people who build and design these technologies to gain insights into their understanding and acceptance of neurodivergence, and the challenges in making their work more accessible to users with diverse needs. Even though neurodivergent individuals are often marginalized for their unique communication styles, nearly all participants overlooked the conclusions their end-users and other AI system makers may draw about communication norms from the implementation and interpretation of humanness applied in participants' work. This highlights a major gap in their broader ethical considerations, compounded by some participants' neuronormative assumptions about the behaviors and traits that distinguish \"humans\" from \"bots\" and the replication of these assumptions in their work. We examine the impact this may have on autism inclusion in society and provide recommendations for additional systemic changes towards more ethical research directions.", "date": "2025-06-12", "categories": []}
{"id": "2506.12437", "title": "Feeling Machines: Ethics, Culture, and the Rise of Emotional AI", "abstract": "This paper explores the growing presence of emotionally responsive artificial intelligence through a critical and interdisciplinary lens. Bringing together the voices of early-career researchers from multiple fields, it explores how AI systems that simulate or interpret human emotions are reshaping our interactions in areas such as education, healthcare, mental health, caregiving, and digital life. The analysis is structured around four central themes: the ethical implications of emotional AI, the cultural dynamics of human-machine interaction, the risks and opportunities for vulnerable populations, and the emerging regulatory, design, and technical considerations. The authors highlight the potential of affective AI to support mental well-being, enhance learning, and reduce loneliness, as well as the risks of emotional manipulation, over-reliance, misrepresentation, and cultural bias. Key challenges include simulating empathy without genuine understanding, encoding dominant sociocultural norms into AI systems, and insufficient safeguards for individuals in sensitive or high-risk contexts. Special attention is given to children, elderly users, and individuals with mental health challenges, who may interact with AI in emotionally significant ways. However, there remains a lack of cognitive or legal protections which are necessary to navigate such engagements safely. The report concludes with ten recommendations, including the need for transparency, certification frameworks, region-specific fine-tuning, human oversight, and longitudinal research. A curated supplementary section provides practical tools, models, and datasets to support further work in this domain.", "date": "2025-06-14", "categories": []}
{"id": "2504.09346", "title": "\"It's not a representation of me\": Examining Accent Bias and Digital Exclusion in Synthetic AI Voice Services", "abstract": "Recent advances in artificial intelligence (AI) speech generation and voice cloning technologies have produced naturalistic speech and accurate voice replication, yet their influence on sociotechnical systems across diverse accents and linguistic traits is not fully understood. This study evaluates two synthetic AI voice services (Speechify and ElevenLabs) through a mixed methods approach using surveys and interviews to assess technical performance and uncover how users' lived experiences influence their perceptions of accent variations in these speech technologies. Our findings reveal technical performance disparities across five regional, English-language accents and demonstrate how current speech generation technologies may inadvertently reinforce linguistic privilege and accent-based discrimination, potentially creating new forms of digital exclusion. Overall, our study highlights the need for inclusive design and regulation by providing actionable insights for developers, policymakers, and organizations to ensure equitable and socially responsible AI speech technologies.", "date": "2025-04-12", "categories": []}
{"id": "2506.11945", "title": "Subjective Experience in AI Systems: What Do AI Researchers and the Public Believe?", "abstract": "We surveyed 582 AI researchers who have published in leading AI venues and 838 nationally representative US participants about their views on the potential development of AI systems with subjective experience and how such systems should be treated and governed. When asked to estimate the chances that such systems will exist on specific dates, the median responses were 1% (AI researchers) and 5% (public) by 2024, 25% and 30% by 2034, and 70% and 60% by 2100, respectively. The median member of the public thought there was a higher chance that AI systems with subjective experience would never exist (25%) than the median AI researcher did (10%). Both groups perceived a need for multidisciplinary expertise to assess AI subjective experience. Although support for welfare protections for such AI systems exceeded opposition, it remained far lower than support for protections for animals or the environment. Attitudes toward moral and governance issues were divided in both groups, especially regarding whether such systems should be created and what rights or protections they should receive. Yet a majority of respondents in both groups agreed that safeguards against the potential risks from AI systems with subjective experience should be implemented by AI developers now, and if created, AI systems with subjective experience should treat others well, behave ethically, and be held accountable. Overall, these results suggest that both AI researchers and the public regard the emergence of AI systems with subjective experience as a possibility this century, though substantial uncertainty and disagreement remain about the timeline and appropriate response.", "date": "2025-06-13", "categories": []}
{"id": "2502.10450", "title": "Trustworthy AI: Safety, Bias, and Privacy -- A Survey", "abstract": "The capabilities of artificial intelligence systems have been advancing to a great extent, but these systems still struggle with failure modes, vulnerabilities, and biases. In this paper, we study the current state of the field, and present promising insights and perspectives regarding concerns that challenge the trustworthiness of AI models. In particular, this paper investigates the issues regarding three thrusts: safety, privacy, and bias, which hurt models' trustworthiness. For safety, we discuss safety alignment in the context of large language models, preventing them from generating toxic or harmful content. For bias, we focus on spurious biases that can mislead a network. Lastly, for privacy, we cover membership inference attacks in deep neural networks. The discussions addressed in this paper reflect our own experiments and observations.", "date": "2025-02-11", "categories": []}
{"id": "2506.09873", "title": "Stakeholder Participation for Responsible AI Development: Disconnects Between Guidance and Current Practice", "abstract": "Responsible AI (rAI) guidance increasingly promotes stakeholder involvement (SHI) during AI development. At the same time, SHI is already common in commercial software development, but with potentially different foci. This study clarifies the extent to which established SHI practices are able to contribute to rAI efforts as well as potential disconnects -- essential insights to inform and tailor future interventions that further shift industry practice towards rAI efforts. First, we analysed 56 rAI guidance documents to identify why SHI is recommended (i.e. its expected benefits for rAI) and uncovered goals such as redistributing power, improving socio-technical understandings, anticipating risks, and enhancing public oversight. To understand why and how SHI is currently practised in commercial settings, we then conducted an online survey (n=130) and semi-structured interviews (n=10) with AI practitioners. Our findings reveal that SHI in practice is primarily driven by commercial priorities (e.g. customer value, compliance) and several factors currently discourage more rAI-aligned SHI practices. This suggests that established SHI practices are largely not contributing to rAI efforts. To address this disconnect, we propose interventions and research opportunities to advance rAI development in practice.", "date": "2025-06-11", "categories": []}
{"id": "2406.03198", "title": "The Impossibility of Fair LLMs", "abstract": "The rise of general-purpose artificial intelligence (AI) systems, particularly large language models (LLMs), has raised pressing moral questions about how to reduce bias and ensure fairness at scale. Researchers have documented a sort of \"bias\" in the significant correlations between demographics (e.g., race, gender) in LLM prompts and responses, but it remains unclear how LLM fairness could be evaluated with more rigorous definitions, such as group fairness or fair representations. We analyze a variety of technical fairness frameworks and find inherent challenges in each that make the development of a fair LLM intractable. We show that each framework either does not logically extend to the general-purpose AI context or is infeasible in practice, primarily due to the large amounts of unstructured training data and the many potential combinations of human populations, use cases, and sensitive attributes. These inherent challenges would persist for general-purpose AI, including LLMs, even if empirical challenges, such as limited participatory input and limited measurement methods, were overcome. Nonetheless, fairness will remain an important type of model evaluation, and there are still promising research directions, particularly the development of standards for the responsibility of LLM developers, context-specific evaluations, and methods of iterative, participatory, and AI-assisted evaluation that could scale fairness across the diverse contexts of modern human-AI interaction.", "date": "2024-05-28", "categories": []}
{"id": "2506.04679", "title": "Normative Conflicts and Shallow AI Alignment", "abstract": "The progress of AI systems such as large language models (LLMs) raises increasingly pressing concerns about their safe deployment. This paper examines the value alignment problem for LLMs, arguing that current alignment strategies are fundamentally inadequate to prevent misuse. Despite ongoing efforts to instill norms such as helpfulness, honesty, and harmlessness in LLMs through fine-tuning based on human preferences, they remain vulnerable to adversarial attacks that exploit conflicts between these norms. I argue that this vulnerability reflects a fundamental limitation of existing alignment methods: they reinforce shallow behavioral dispositions rather than endowing LLMs with a genuine capacity for normative deliberation. Drawing from on research in moral psychology, I show how humans' ability to engage in deliberative reasoning enhances their resilience against similar adversarial tactics. LLMs, by contrast, lack a robust capacity to detect and rationally resolve normative conflicts, leaving them susceptible to manipulation; even recent advances in reasoning-focused LLMs have not addressed this vulnerability. This ``shallow alignment'' problem carries significant implications for AI safety and regulation, suggesting that current approaches are insufficient for mitigating potential harms posed by increasingly capable AI systems.", "date": "2025-06-05", "categories": []}
{"id": "2503.05728", "title": "Political Neutrality in AI Is Impossible- But Here Is How to Approximate It", "abstract": "AI systems often exhibit political bias, influencing users' opinions and decisions. While political neutrality-defined as the absence of bias-is often seen as an ideal solution for fairness and safety, this position paper argues that true political neutrality is neither feasible nor universally desirable due to its subjective nature and the biases inherent in AI training data, algorithms, and user interactions. However, inspired by Joseph Raz's philosophical insight that \"neutrality [...] can be a matter of degree\" (Raz, 1986), we argue that striving for some neutrality remains essential for promoting balanced AI interactions and mitigating user manipulation. Therefore, we use the term \"approximation\" of political neutrality to shift the focus from unattainable absolutes to achievable, practical proxies. We propose eight techniques for approximating neutrality across three levels of conceptualizing AI, examining their trade-offs and implementation strategies. In addition, we explore two concrete applications of these approximations to illustrate their practicality. Finally, we assess our framework on current large language models (LLMs) at the output level, providing a demonstration of how it can be evaluated. This work seeks to advance nuanced discussions of political neutrality in AI and promote the development of responsible, aligned language models.", "date": "2025-02-18", "categories": []}
{"id": "2506.02017", "title": "Fairness through Feedback: Addressing Algorithmic Misgendering in Automatic Gender Recognition", "abstract": "Automatic Gender Recognition (AGR) systems are an increasingly widespread application in the Machine Learning (ML) landscape. While these systems are typically understood as detecting gender, they often classify datapoints based on observable features correlated at best with either male or female sex. In addition to questionable binary assumptions, from an epistemological point of view, this is problematic for two reasons. First, there exists a gap between the categories the system is meant to predict (woman versus man) and those onto which their output reasonably maps (female versus male). What is more, gender cannot be inferred on the basis of such observable features. This makes AGR tools often unreliable, especially in the case of non-binary and gender non-conforming people. We suggest a theoretical and practical rethinking of AGR systems. To begin, distinctions are made between sex, gender, and gender expression. Then, we build upon the observation that, unlike algorithmic misgendering, human-human misgendering is open to the possibility of re-evaluation and correction. We suggest that analogous dynamics should be recreated in AGR, giving users the possibility to correct the system's output. While implementing such a feedback mechanism could be regarded as diminishing the system's autonomy, it represents a way to significantly increase fairness levels in AGR. This is consistent with the conceptual change of paradigm that we advocate for AGR systems, which should be understood as tools respecting individuals' rights and capabilities of self-expression and determination.", "date": "2025-05-28", "categories": []}
{"id": "2506.02055", "title": "Will Agents Replace Us? Perceptions of Autonomous Multi-Agent AI", "abstract": "Autonomous multi-agent AI systems are poised to transform various industries, particularly software development and knowledge work. Understanding current perceptions among professionals is crucial for anticipating adoption challenges, ethical considerations, and future workforce development. This study analyzes responses from 130 participants to a survey on the capabilities, impact, and governance of AI agents. We explore expected timelines for AI replacing programmers, identify perceived barriers to deployment, and examine beliefs about responsibility when agents make critical decisions. Key findings reveal three distinct clusters of respondents. While the study explored factors associated with current AI agent deployment, the initial logistic regression model did not yield statistically significant predictors, suggesting that deployment decisions are complex and may be influenced by factors not fully captured or that a larger sample is needed. These insights highlight the need for organizations to address compliance concerns (a commonly cited barrier) and establish clear governance frameworks as they integrate autonomous agents into their workflows.", "date": "2025-06-01", "categories": []}
{"id": "2409.03219", "title": "Content Moderation by LLM: From Accuracy to Legitimacy", "abstract": "One trending application of LLM (large language model) is to use it for content moderation in online platforms. Most current studies on this application have focused on the metric of accuracy -- the extent to which LLMs make correct decisions about content. This article argues that accuracy is insufficient and misleading because it fails to grasp the distinction between easy cases and hard cases, as well as the inevitable trade-offs in achieving higher accuracy. Closer examination reveals that content moderation is a constitutive part of platform governance, the key of which is to gain and enhance legitimacy. Instead of making moderation decisions correct, the chief goal of LLMs is to make them legitimate. In this regard, this article proposes a paradigm shift from the single benchmark of accuracy towards a legitimacy-based framework for evaluating the performance of LLM moderators. The framework suggests that for easy cases, the key is to ensure accuracy, speed, and transparency, while for hard cases, what matters is reasoned justification and user participation. Examined under this framework, LLMs' real potential in moderation is not accuracy improvement. Rather, LLMs can better contribute in four other aspects: to conduct screening of hard cases from easy cases, to provide quality explanations for moderation decisions, to assist human reviewers in getting more contextual information, and to facilitate user participation in a more interactive way. To realize these contributions, this article proposes a workflow for incorporating LLMs into the content moderation system. Using normative theories from law and social sciences to critically assess the new technological application, this article seeks to redefine LLMs' role in content moderation and redirect relevant research in this field.", "date": "2024-09-05", "categories": []}
{"id": "2412.03824", "title": "Towards Data Governance of Frontier AI Models", "abstract": "Data is essential to train and fine-tune today's frontier artificial intelligence (AI) models and to develop future ones. To date, academic, legal, and regulatory work has primarily addressed how data can directly harm consumers and creators, such as through privacy breaches, copyright infringements, and bias and discrimination. Our work, instead, focuses on the comparatively neglected question of how data can enable new governance capacities for frontier AI models. This approach for \"frontier data governance\" opens up new avenues for monitoring and mitigating risks from advanced AI models, particularly as they scale and acquire specific dangerous capabilities. Still, frontier data governance faces challenges that stem from the fundamental properties of data itself: data is non-rival, often non-excludable, easily replicable, and increasingly synthesizable. Despite these inherent difficulties, we propose a set of policy mechanisms targeting key actors along the data supply chain, including data producers, aggregators, model developers, and data vendors. We provide a brief overview of 15 governance mechanisms, of which we centrally introduce five, underexplored policy recommendations. These include developing canary tokens to detect unauthorized use for producers; (automated) data filtering to remove malicious content for pre-training and post-training datasets; mandatory dataset reporting requirements for developers and vendors; improved security for datasets and data generation algorithms; and know-your-customer requirements for vendors. By considering data not just as a source of potential harm, but as a critical governance lever, this work aims to equip policymakers with a new tool for the governance and regulation of frontier AI models.", "date": "2024-12-05", "categories": []}
{"id": "2503.16456", "title": "Position: Beyond Assistance -- Reimagining LLMs as Ethical and Adaptive Co-Creators in Mental Health Care", "abstract": "This position paper argues for a fundamental shift in how Large Language Models (LLMs) are integrated into the mental health care domain. We advocate for their role as co-creators rather than mere assistive tools. While LLMs have the potential to enhance accessibility, personalization, and crisis intervention, their adoption remains limited due to concerns about bias, evaluation, over-reliance, dehumanization, and regulatory uncertainties. To address these challenges, we propose two structured pathways: SAFE-i (Supportive, Adaptive, Fair, and Ethical Implementation) Guidelines for ethical and responsible deployment, and HAAS-e (Human-AI Alignment and Safety Evaluation) Framework for multidimensional, human-centered assessment. SAFE-i provides a blueprint for data governance, adaptive model engineering, and real-world integration, ensuring LLMs align with clinical and ethical standards. HAAS-e introduces evaluation metrics that go beyond technical accuracy to measure trustworthiness, empathy, cultural sensitivity, and actionability. We call for the adoption of these structured approaches to establish a responsible and scalable model for LLM-driven mental health support, ensuring that AI complements, rather than replaces, human expertise.", "date": "2025-02-21", "categories": []}
{"id": "2506.00038", "title": "The Folly of AI for Age Verification", "abstract": "In the near future a governmental body will be asked to allow companies to use AI for age verification. If they allow it the resulting system will both be easily circumvented and disproportionately misclassify minorities and low socioeconomic status users. This is predictable by showing that other very similar systems (facial recognition and remote proctoring software) have similar issues despite years of efforts to mitigate their biases. These biases are due to technical limitations both of the AI models themselves and the physical hardware they are running on that will be difficult to overcome below the cost of government ID-based age verification. Thus in, the near future, deploying an AI system for age verification is folly.", "date": "2025-05-27", "categories": []}
{"id": "2506.00047", "title": "Risks of AI-driven product development and strategies for their mitigation", "abstract": "Humanity is progressing towards automated product development, a trend that promises faster creation of better products and thus the acceleration of technological progress. However, increasing reliance on non-human agents for this process introduces many risks. This perspective aims to initiate a discussion on these risks and appropriate mitigation strategies. To this end, we outline a set of principles for safer AI-driven product development which emphasize human oversight, accountability, and explainable design, among others. The risk assessment covers both technical risks which affect product quality and safety, and sociotechnical risks which affect society. While AI-driven product development is still in its early stages, this discussion will help balance its opportunities and risks without delaying essential progress in understanding, norm-setting, and regulation.", "date": "2025-05-28", "categories": []}
{"id": "2506.00080", "title": "Bottom-Up Perspectives on AI Governance: Insights from User Reviews of AI Products", "abstract": "With the growing importance of AI governance, numerous high-level frameworks and principles have been articulated by policymakers, institutions, and expert communities to guide the development and application of AI. While such frameworks offer valuable normative orientation, they may not fully capture the practical concerns of those who interact with AI systems in organizational and operational contexts. To address this gap, this study adopts a bottom-up approach to explore how governance-relevant themes are expressed in user discourse. Drawing on over 100,000 user reviews of AI products from G2.com, we apply BERTopic to extract latent themes and identify those most semantically related to AI governance. The analysis reveals a diverse set of governance-relevant topics spanning both technical and non-technical domains. These include concerns across organizational processes-such as planning, coordination, and communication-as well as stages of the AI value chain, including deployment infrastructure, data handling, and analytics. The findings show considerable overlap with institutional AI governance and ethics frameworks on issues like privacy and transparency, but also surface overlooked areas such as project management, strategy development, and customer interaction. This highlights the need for more empirically grounded, user-centered approaches to AI governance-approaches that complement normative models by capturing how governance unfolds in applied settings. By foregrounding how governance is enacted in practice, this study contributes to more inclusive and operationally grounded approaches to AI governance and digital policy.", "date": "2025-05-30", "categories": []}
{"id": "2506.00094", "title": "Feeling Guilty Being a c(ai)borg: Navigating the Tensions Between Guilt and Empowerment in AI Use", "abstract": "This paper explores the emotional, ethical and practical dimensions of integrating Artificial Intelligence (AI) into personal and professional workflows, focusing on the concept of feeling guilty as a 'c(ai)borg' - a human augmented by AI. Inspired by Donna Haraway's Cyborg Manifesto, the study explores how AI challenges traditional notions of creativity, originality and intellectual labour. Using an autoethnographic approach, the authors reflect on their year-long experiences with AI tools, revealing a transition from initial guilt and reluctance to empowerment through skill-building and transparency. Key findings highlight the importance of basic academic skills, advanced AI literacy and honest engagement with AI results. The c(ai)borg vision advocates for a future where AI is openly embraced as a collaborative partner, fostering innovation and equity while addressing issues of access and agency. By reframing guilt as growth, the paper calls for a thoughtful and inclusive approach to AI integration.", "date": "2025-05-30", "categories": []}
{"id": "2506.00233", "title": "Ethical AI: Towards Defining a Collective Evaluation Framework", "abstract": "Artificial Intelligence (AI) is transforming sectors such as healthcare, finance, and autonomous systems, offering powerful tools for innovation. Yet its rapid integration raises urgent ethical concerns related to data ownership, privacy, and systemic bias. Issues like opaque decision-making, misleading outputs, and unfair treatment in high-stakes domains underscore the need for transparent and accountable AI systems. This article addresses these challenges by proposing a modular ethical assessment framework built on ontological blocks of meaning-discrete, interpretable units that encode ethical principles such as fairness, accountability, and ownership. By integrating these blocks with FAIR (Findable, Accessible, Interoperable, Reusable) principles, the framework supports scalable, transparent, and legally aligned ethical evaluations, including compliance with the EU AI Act. Using a real-world use case in AI-powered investor profiling, the paper demonstrates how the framework enables dynamic, behavior-informed risk classification. The findings suggest that ontological blocks offer a promising path toward explainable and auditable AI ethics, though challenges remain in automation and probabilistic reasoning.", "date": "2025-05-30", "categories": []}
{"id": "2503.08720", "title": "AI for Just Work: Constructing Diverse Imaginations of AI beyond \"Replacing Humans\"", "abstract": "\"why\" we develop AI. Lacking critical reflections on the general visions and purposes of AI may make the community vulnerable to manipulation. In this position paper, we explore the \"why\" question of AI. We denote answers to the \"why\" question the imaginations of AI, which depict our general visions, frames, and mindsets for the prospects of AI. We identify that the prevailing vision in the AI community is largely a monoculture that emphasizes objectives such as replacing humans and improving productivity. Our critical examination of this mainstream imagination highlights its underpinning and potentially unjust assumptions. We then call to diversify our collective imaginations of AI, embedding ethical assumptions from the outset in the imaginations of AI. To facilitate the community's pursuit of diverse imaginations, we demonstrate one process for constructing a new imagination of \"AI for just work,\" and showcase its application in the medical image synthesis task to make it more ethical. We hope this work will help the AI community to open critical dialogues with civil society on the visions and purposes of AI, and inspire more technical works and advocacy in pursuit of diverse and ethical imaginations to restore the value of AI for the public good.", "date": "2025-03-10", "categories": []}
{"id": "2505.23417", "title": "Toward Effective AI Governance: A Review of Principles", "abstract": "Artificial Intelligence (AI) governance is the practice of establishing frameworks, policies, and procedures to ensure the responsible, ethical, and safe development and deployment of AI systems. Although AI governance is a core pillar of Responsible AI, current literature still lacks synthesis across such governance frameworks and practices. Objective: To identify which frameworks, principles, mechanisms, and stakeholder roles are emphasized in secondary literature on AI governance. Method: We conducted a rapid tertiary review of nine peer-reviewed secondary studies from IEEE and ACM (20202024), using structured inclusion criteria and thematic semantic synthesis. Results: The most cited frameworks include the EU AI Act and NIST RMF; transparency and accountability are the most common principles. Few reviews detail actionable governance mechanisms or stakeholder strategies. Conclusion: The review consolidates key directions in AI governance and highlights gaps in empirical validation and inclusivity. Findings inform both academic inquiry and practical adoption in organizations.", "date": "2025-05-29", "categories": []}
{"id": "2505.23559", "title": "SafeScientist: Toward Risk-Aware Scientific Discoveries by LLM Agents", "abstract": "Recent advancements in large language model (LLM) agents have significantly accelerated scientific discovery automation, yet concurrently raised critical ethical and safety concerns. To systematically address these challenges, we introduce \\textbf{SafeScientist}, an innovative AI scientist framework explicitly designed to enhance safety and ethical responsibility in AI-driven scientific exploration. SafeScientist proactively refuses ethically inappropriate or high-risk tasks and rigorously emphasizes safety throughout the research process. To achieve comprehensive safety oversight, we integrate multiple defensive mechanisms, including prompt monitoring, agent-collaboration monitoring, tool-use monitoring, and an ethical reviewer component. Complementing SafeScientist, we propose \\textbf{SciSafetyBench}, a novel benchmark specifically designed to evaluate AI safety in scientific contexts, comprising 240 high-risk scientific tasks across 6 domains, alongside 30 specially designed scientific tools and 120 tool-related risk tasks. Extensive experiments demonstrate that SafeScientist significantly improves safety performance by 35\\% compared to traditional AI scientist frameworks, without compromising scientific output quality. Additionally, we rigorously validate the robustness of our safety pipeline against diverse adversarial attack methods, further confirming the effectiveness of our integrated approach. The code and data will be available at https://github.com/ulab-uiuc/SafeScientist. \\textcolor{red}{Warning: this paper contains example data that may be offensive or harmful.}", "date": "2025-05-29", "categories": []}
{"id": "2410.16665", "title": "SafetyAnalyst: Interpretable, Transparent, and Steerable Safety Moderation for AI Behavior", "abstract": "The ideal AI safety moderation system would be both structurally interpretable (so its decisions can be reliably explained) and steerable (to align to safety standards and reflect a community's values), which current systems fall short on. To address this gap, we present SafetyAnalyst, a novel AI safety moderation framework. Given an AI behavior, SafetyAnalyst uses chain-of-thought reasoning to analyze its potential consequences by creating a structured \"harm-benefit tree,\" which enumerates harmful and beneficial actions and effects the AI behavior may lead to, along with likelihood, severity, and immediacy labels that describe potential impacts on stakeholders. SafetyAnalyst then aggregates all effects into a harmfulness score using 28 fully interpretable weight parameters, which can be aligned to particular safety preferences. We applied this framework to develop an open-source LLM prompt safety classification system, distilled from 18.5 million harm-benefit features generated by frontier LLMs on 19k prompts. On comprehensive benchmarks, we show that SafetyAnalyst (average F1=0.81) outperforms existing moderation systems (average F1$<$0.72) on prompt safety classification, while offering the additional advantages of interpretability, transparency, and steerability.", "date": "2024-10-22", "categories": []}
{"id": "2502.00136", "title": "A Checks-and-Balances Framework for Context-Aware Ethical AI Alignment", "abstract": "This paper introduces a checks-and-balances framework for ethical alignment of Large Language Models (LLMs), inspired by three-branch governmental systems. It implements three independent yet interacting components: LLMs as the executive branch for knowledge generation, DIKE as the legislative branch establishing ethical guardrails, and ERIS as the judicial branch for contextual interpretation. Beyond structural separation, we address a fundamental challenge: regulating emotion to shape behaviors. Drawing from psychological theories where managing emotional responses prevents harmful behaviors, we develop a self-supervised learning pipeline that maps emotions to linguistic behaviors, enabling precise behavioral modulation through emotional conditioning. By integrating this approach with adversarial testing, our framework demonstrates how DIKE and ERIS direct linguistic behaviors toward ethical outcomes while preserving independence throughout knowledge generation, ethical oversight, and contextual interpretation.", "date": "2025-01-31", "categories": []}
{"id": "2505.21570", "title": "Beyond Explainability: The Case for AI Validation", "abstract": "Artificial Knowledge (AK) systems are transforming decision-making across critical domains such as healthcare, finance, and criminal justice. However, their growing opacity presents governance challenges that current regulatory approaches, focused predominantly on explainability, fail to address adequately. This article argues for a shift toward validation as a central regulatory pillar. Validation, ensuring the reliability, consistency, and robustness of AI outputs, offers a more practical, scalable, and risk-sensitive alternative to explainability, particularly in high-stakes contexts where interpretability may be technically or economically unfeasible. We introduce a typology based on two axes, validity and explainability, classifying AK systems into four categories and exposing the trade-offs between interpretability and output reliability. Drawing on comparative analysis of regulatory approaches in the EU, US, UK, and China, we show how validation can enhance societal trust, fairness, and safety even where explainability is limited. We propose a forward-looking policy framework centered on pre- and post-deployment validation, third-party auditing, harmonized standards, and liability incentives. This framework balances innovation with accountability and provides a governance roadmap for responsibly integrating opaque, high-performing AK systems into society.", "date": "2025-05-27", "categories": []}
{"id": "2505.21720", "title": "Responsible Data Stewardship: Generative AI and the Digital Waste Problem", "abstract": "As generative AI systems become widely adopted, they enable unprecedented creation levels of synthetic data across text, images, audio, and video modalities. While research has addressed the energy consumption of model training and inference, a critical sustainability challenge remains understudied: digital waste. This term refers to stored data that consumes resources without serving a specific (and/or immediate) purpose. This paper presents this terminology in the AI context and introduces digital waste as an ethical imperative within (generative) AI development, positioning environmental sustainability as core for responsible innovation. Drawing from established digital resource management approaches, we examine how other disciplines manage digital waste and identify transferable approaches for the AI community. We propose specific recommendations encompassing re-search directions, technical interventions, and cultural shifts to mitigate the environmental consequences of in-definite data storage. By expanding AI ethics beyond immediate concerns like bias and privacy to include inter-generational environmental justice, this work contributes to a more comprehensive ethical framework that considers the complete lifecycle impact of generative AI systems.", "date": "2025-05-27", "categories": []}
{"id": "2504.13199", "title": "Building Trustworthy Multimodal AI: A Review of Fairness, Transparency, and Ethics in Vision-Language Tasks", "abstract": "Objective: This review explores the trustworthiness of multimodal artificial intelligence (AI) systems, specifically focusing on vision-language tasks. It addresses critical challenges related to fairness, transparency, and ethical implications in these systems, providing a comparative analysis of key tasks such as Visual Question Answering (VQA), image captioning, and visual dialogue. Background: Multimodal models, particularly vision-language models, enhance artificial intelligence (AI) capabilities by integrating visual and textual data, mimicking human learning processes. Despite significant advancements, the trustworthiness of these models remains a crucial concern, particularly as AI systems increasingly confront issues regarding fairness, transparency, and ethics. Methods: This review examines research conducted from 2017 to 2024 focusing on forenamed core vision-language tasks. It employs a comparative approach to analyze these tasks through the lens of trustworthiness, underlining fairness, explainability, and ethics. This study synthesizes findings from recent literature to identify trends, challenges, and state-of-the-art solutions. Results: Several key findings were highlighted. Transparency: Explainability of vision language tasks is important for user trust. Techniques, such as attention maps and gradient-based methods, have successfully addressed this issue. Fairness: Bias mitigation in VQA and visual dialogue systems is essential for ensuring unbiased outcomes across diverse demographic groups. Ethical Implications: Addressing biases in multilingual models and ensuring ethical data handling is critical for the responsible deployment of vision-language systems. Conclusion: This study underscores the importance of integrating fairness, transparency, and ethical considerations in developing vision-language models within a unified framework.", "date": "2025-04-14", "categories": []}
{"id": "2505.20692", "title": "Can we Debias Social Stereotypes in AI-Generated Images? Examining Text-to-Image Outputs and User Perceptions", "abstract": "Recent advances in generative AI have enabled visual content creation through text-to-image (T2I) generation. However, despite their creative potential, T2I models often replicate and amplify societal stereotypes -- particularly those related to gender, race, and culture -- raising important ethical concerns. This paper proposes a theory-driven bias detection rubric and a Social Stereotype Index (SSI) to systematically evaluate social biases in T2I outputs. We audited three major T2I model outputs -- DALL-E-3, Midjourney-6.1, and Stability AI Core -- using 100 queries across three categories -- geocultural, occupational, and adjectival. Our analysis reveals that initial outputs are prone to include stereotypical visual cues, including gendered professions, cultural markers, and western beauty norms. To address this, we adopted our rubric to conduct targeted prompt refinement using LLMs, which significantly reduced bias -- SSI dropped by 61% for geocultural, 69% for occupational, and 51% for adjectival queries. We complemented our quantitative analysis through a user study examining perceptions, awareness, and preferences around AI-generated biased imagery. Our findings reveal a key tension -- although prompt refinement can mitigate stereotypes, it can limit contextual alignment. Interestingly, users often perceived stereotypical images to be more aligned with their expectations. We discuss the need to balance ethical debiasing with contextual relevance and call for T2I systems that support global diversity and inclusivity while not compromising the reflection of real-world social complexity.", "date": "2025-05-27", "categories": []}
{"id": "2402.04420", "title": "Measuring Machine Learning Harms from Stereotypes Requires Understanding Who Is Harmed by Which Errors in What Ways", "abstract": "As machine learning applications proliferate, we need an understanding of their potential for harm. However, current fairness metrics are rarely grounded in human psychological experiences of harm. Drawing on the social psychology of stereotypes, we use a case study of gender stereotypes in image search to examine how people react to machine learning errors. First, we use survey studies to show that not all machine learning errors reflect stereotypes nor are equally harmful. Then, in experimental studies we randomly expose participants to stereotype-reinforcing, -violating, and -neutral machine learning errors. We find stereotype-reinforcing errors induce more experientially (i.e., subjectively) harmful experiences, while having minimal changes to cognitive beliefs, attitudes, or behaviors. This experiential harm impacts women more than men. However, certain stereotype-violating errors are more experientially harmful for men, potentially due to perceived threats to masculinity. We conclude that harm cannot be the sole guide in fairness mitigation, and propose a nuanced perspective depending on who is experiencing what harm and why.", "date": "2024-02-06", "categories": []}
{"id": "2503.04750", "title": "AI Agents Should be Regulated Based on the Extent of Their Autonomous Operations", "abstract": "This position paper argues that AI agents should be regulated by the extent to which they operate autonomously. AI agents with long-term planning and strategic capabilities can pose significant risks of human extinction and irreversible global catastrophes. While existing regulations often focus on computational scale as a proxy for potential harm, we argue that such measures are insufficient for assessing the risks posed by agents whose capabilities arise primarily from inference-time computation. To support our position, we discuss relevant regulations and recommendations from scientists regarding existential risks, as well as the advantages of using action sequences -- which reflect the degree of an agent's autonomy -- as a more suitable measure of potential impact than existing metrics that rely on observing environmental states.", "date": "2025-02-07", "categories": []}
{"id": "2503.20099", "title": "AI Identity, Empowerment, and Mindfulness in Mitigating Unethical AI Use", "abstract": "This study examines how AI identity influences psychological empowerment and unethical AI behavior among college students, while also exploring the moderating role of IT mindfulness. Findings show that a strong AI identity enhances psychological empowerment and academic engagement but can also lead to increased unethical AI practices. Crucially, IT mindfulness acts as an ethical safeguard, promoting sensitivity to ethical concerns and reducing misuse of AI. These insights have implications for educators, policymakers, and AI developers, emphasizing For Peer Review the need for a balanced approach that encourages digital engagement without compromising student responsibility. The study also contributes to philosophical discussions of psychological agency, suggesting that empowerment through AI can yield both positive and negative outcomes. Mindfulness emerges as essential in guiding ethical AI interactions. Overall, the research informs ongoing debates on ethics in education and AI, offering strategies to align technological advancement with ethical accountability and responsible use.", "date": "2025-03-25", "categories": []}
{"id": "2505.18236", "title": "From Bias to Accountability: How the EU AI Act Confronts Challenges in European GeoAI Auditing", "abstract": "Bias in geospatial artificial intelligence (GeoAI) models has been documented, yet the evidence is scattered across narrowly focused studies. We synthesize this fragmented literature to provide a concise overview of bias in GeoAI and examine how the EU's Artificial Intelligence Act (EU AI Act) shapes audit obligations. We discuss recurring bias mechanisms, including representation, algorithmic and aggregation bias, and map them to specific provisions of the EU AI Act. By applying the Act's high-risk criteria, we demonstrate that widely deployed GeoAI applications qualify as high-risk systems. We then present examples of recent audits along with an outline of practical methods for detecting bias. As far as we know, this study represents the first integration of GeoAI bias evidence into the EU AI Act context, by identifying high-risk GeoAI systems and mapping bias mechanisms to the Act's Articles. Although the analysis is exploratory, it suggests that even well-curated European datasets should employ routine bias audits before 2027, when the AI Act's high-risk provisions take full effect.", "date": "2025-05-23", "categories": []}
{"id": "2505.18425", "title": "Advertising in AI systems: Society must be vigilant", "abstract": "AI systems have increasingly become our gateways to the Internet. We argue that just as advertising has driven the monetization of web search and social media, so too will commercial incentives shape the content served by AI. Unlike traditional media, however, the outputs of these systems are dynamic, personalized, and lack clear provenance -- raising concerns for transparency and regulation. In this paper, we envision how commercial content could be delivered through generative AI-based systems. Based on the requirements of key stakeholders -- advertisers, consumers, and platforms -- we propose design principles for commercially-influenced AI systems. We then outline high-level strategies for end users to identify and mitigate commercial biases from model outputs. Finally, we conclude with open questions and a call to action towards these goals.", "date": "2025-05-23", "categories": []}
{"id": "2505.19212", "title": "When Ethics and Payoffs Diverge: LLM Agents in Morally Charged Social Dilemmas", "abstract": "Recent advances in large language models (LLMs) have enabled their use in complex agentic roles, involving decision-making with humans or other agents, making ethical alignment a key AI safety concern. While prior work has examined both LLMs' moral judgment and strategic behavior in social dilemmas, there is limited understanding of how they act when moral imperatives directly conflict with rewards or incentives. To investigate this, we introduce Moral Behavior in Social Dilemma Simulation (MoralSim) and evaluate how LLMs behave in the prisoner's dilemma and public goods game with morally charged contexts. In MoralSim, we test a range of frontier models across both game structures and three distinct moral framings, enabling a systematic examination of how LLMs navigate social dilemmas in which ethical norms conflict with payoff-maximizing strategies. Our results show substantial variation across models in both their general tendency to act morally and the consistency of their behavior across game types, the specific moral framing, and situational factors such as opponent behavior and survival risks. Crucially, no model exhibits consistently moral behavior in MoralSim, highlighting the need for caution when deploying LLMs in agentic roles where the agent's \"self-interest\" may conflict with ethical expectations. Our code is available at https://github.com/sbackmann/moralsim.", "date": "2025-05-25", "categories": []}
{"id": "2404.19244", "title": "A University Framework for the Responsible use of Generative AI in Research", "abstract": "Generative Artificial Intelligence (generative AI) poses both opportunities and risks for the integrity of research. Universities must guide researchers in using generative AI responsibly, and in navigating a complex regulatory landscape subject to rapid change. By drawing on the experiences of two Australian universities, we propose a framework to help institutions promote and facilitate the responsible use of generative AI. We provide guidance to help distil the diverse regulatory environment into a principles-based position statement. Further, we explain how a position statement can then serve as a foundation for initiatives in training, communications, infrastructure, and process change. Despite the growing body of literature about AI's impact on academic integrity for undergraduate students, there has been comparatively little attention on the impacts of generative AI for research integrity, and the vital role of institutions in helping to address those challenges. This paper underscores the urgency for research institutions to take action in this area and suggests a practical and adaptable framework for so doing.", "date": "2024-04-30", "categories": []}
{"id": "2505.17112", "title": "Cultural Value Alignment in Large Language Models: A Prompt-based Analysis of Schwartz Values in Gemini, ChatGPT, and DeepSeek", "abstract": "This study examines cultural value alignment in large language models (LLMs) by analyzing how Gemini, ChatGPT, and DeepSeek prioritize values from Schwartz's value framework. Using the 40-item Portrait Values Questionnaire, we assessed whether DeepSeek, trained on Chinese-language data, exhibits distinct value preferences compared to Western models. Results of a Bayesian ordinal regression model show that self-transcendence values (e.g., benevolence, universalism) were highly prioritized across all models, reflecting a general LLM tendency to emphasize prosocial values. However, DeepSeek uniquely downplayed self-enhancement values (e.g., power, achievement) compared to ChatGPT and Gemini, aligning with collectivist cultural tendencies. These findings suggest that LLMs reflect culturally situated biases rather than a universal ethical framework. To address value asymmetries in LLMs, we propose multi-perspective reasoning, self-reflective feedback, and dynamic contextualization. This study contributes to discussions on AI fairness, cultural neutrality, and the need for pluralistic AI alignment frameworks that integrate diverse moral perspectives.", "date": "2025-05-21", "categories": []}
{"id": "2505.17165", "title": "A Toolkit for Compliance, a Toolkit for Justice: Drawing on Cross-sectoral Expertise to Develop a Pro-justice EU AI Act Toolkit", "abstract": "The introduction of the AI Act in the European Union presents the AI research and practice community with a set of new challenges related to compliance. While it is certain that AI practitioners will require additional guidance and tools to meet these requirements, previous research on toolkits that aim to translate the theory of AI ethics into development and deployment practice suggests that such resources suffer from multiple limitations. These limitations stem, in part, from the fact that the toolkits are either produced by industry-based teams or by academics whose work tends to be abstract and divorced from the realities of industry. In this paper, we discuss the challenge of developing an AI ethics toolkit for practitioners that helps them comply with new AI-focused regulation, but that also moves beyond mere compliance to consider broader socio-ethical questions throughout development and deployment. The toolkit was created through a cross-sectoral collaboration between an academic team based in the UK and an industry team in Italy. We outline the background and rationale for creating a pro-justice AI Act compliance toolkit, detail the process undertaken to develop it, and describe the collaboration and negotiation efforts that shaped its creation. We aim for the described process to serve as a blueprint for other teams navigating the challenges of academia-industry partnerships and aspiring to produce usable and meaningful AI ethics resources.", "date": "2025-05-22", "categories": []}
{"id": "2505.17841", "title": "TEDI: Trustworthy and Ethical Dataset Indicators to Analyze and Compare Dataset Documentation", "abstract": "Dataset transparency is a key enabler of responsible AI, but insights into multimodal dataset attributes that impact trustworthy and ethical aspects of AI applications remain scarce and are difficult to compare across datasets. To address this challenge, we introduce Trustworthy and Ethical Dataset Indicators (TEDI) that facilitate the systematic, empirical analysis of dataset documentation. TEDI encompasses 143 fine-grained indicators that characterize trustworthy and ethical attributes of multimodal datasets and their collection processes. The indicators are framed to extract verifiable information from dataset documentation. Using TEDI, we manually annotated and analyzed over 100 multimodal datasets that include human voices. We further annotated data sourcing, size, and modality details to gain insights into the factors that shape trustworthy and ethical dimensions across datasets. We find that only a select few datasets have documented attributes and practices pertaining to consent, privacy, and harmful content indicators. The extent to which these and other ethical indicators are addressed varies based on the data collection method, with documentation of datasets collected via crowdsourced and direct collection approaches being more likely to mention them. Scraping dominates scale at the cost of ethical indicators, but is not the only viable collection method. Our approach and empirical insights contribute to increasing dataset transparency along trustworthy and ethical dimensions and pave the way for automating the tedious task of extracting information from dataset documentation in future.", "date": "2025-05-23", "categories": []}
{"id": "2505.18154", "title": "The Staircase of Ethics: Probing LLM Value Priorities through Multi-Step Induction to Complex Moral Dilemmas", "abstract": "Ethical decision-making is a critical aspect of human judgment, and the growing use of LLMs in decision-support systems necessitates a rigorous evaluation of their moral reasoning capabilities. However, existing assessments primarily rely on single-step evaluations, failing to capture how models adapt to evolving ethical challenges. Addressing this gap, we introduce the Multi-step Moral Dilemmas (MMDs), the first dataset specifically constructed to evaluate the evolving moral judgments of LLMs across 3,302 five-stage dilemmas. This framework enables a fine-grained, dynamic analysis of how LLMs adjust their moral reasoning across escalating dilemmas. Our evaluation of nine widely used LLMs reveals that their value preferences shift significantly as dilemmas progress, indicating that models recalibrate moral judgments based on scenario complexity. Furthermore, pairwise value comparisons demonstrate that while LLMs often prioritize the value of care, this value can sometimes be superseded by fairness in certain contexts, highlighting the dynamic and context-dependent nature of LLM ethical reasoning. Our findings call for a shift toward dynamic, context-aware evaluation paradigms, paving the way for more human-aligned and value-sensitive development of LLMs.", "date": "2025-05-23", "categories": []}
{"id": "2411.06790", "title": "Large-scale moral machine experiment on large language models", "abstract": "The rapid advancement of Large Language Models (LLMs) and their potential integration into autonomous driving systems necessitates understanding their moral decision-making capabilities. While our previous study examined four prominent LLMs using the Moral Machine experimental framework, the dynamic landscape of LLM development demands a more comprehensive analysis. Here, we evaluate moral judgments across 52 different LLMs, including multiple versions of proprietary models (GPT, Claude, Gemini) and open-source alternatives (Llama, Gemma), to assess their alignment with human moral preferences in autonomous driving scenarios. Using a conjoint analysis framework, we evaluated how closely LLM responses aligned with human preferences in ethical dilemmas and examined the effects of model size, updates, and architecture. Results showed that proprietary models and open-source models exceeding 10 billion parameters demonstrated relatively close alignment with human judgments, with a significant negative correlation between model size and distance from human judgments in open-source models. However, model updates did not consistently improve alignment with human preferences, and many LLMs showed excessive emphasis on specific ethical principles. These findings suggest that while increasing model size may naturally lead to more human-like moral judgments, practical implementation in autonomous driving systems requires careful consideration of the trade-off between judgment quality and computational efficiency. Our comprehensive analysis provides crucial insights for the ethical design of autonomous systems and highlights the importance of considering cultural contexts in AI moral decision-making.", "date": "2024-11-11", "categories": []}
{"id": "2505.08133", "title": "One Bad NOFO? AI Governance in Federal Grantmaking", "abstract": "Much scholarship considers how U.S. federal agencies govern artificial intelligence (AI) through rulemaking and their own internal use policies. But agencies have an overlooked AI governance role: setting discretionary grant policy when directing billions of dollars in federal financial assistance. These dollars enable state and local entities to study, create, and use AI. This funding not only goes to dedicated AI programs, but also to grantees using AI in the course of meeting their routine grant objectives. As discretionary grantmakers, agencies guide and restrict what grant winners do -- a hidden lever for AI governance. Agencies pull this lever by setting program objectives, judging criteria, and restrictions for AI use. Using a novel dataset of over 40,000 non-defense federal grant notices of funding opportunity (NOFOs) posted to the U.S. federal grants website between 2009 and 2024, we analyze how agencies regulate the use of AI by grantees. We select records mentioning AI and review their stated goals and requirements. We find agencies promoting AI in notice narratives, shaping adoption in ways other records of grant policy might fail to capture. Of the grant opportunities that mention AI, we find only a handful of AI-specific judging criteria or restrictions. This silence holds even when agencies fund AI uses in contexts affecting people's rights and which, under an analogous federal procurement regime, would result in extra oversight. These findings recast grant notices as a site of AI policymaking -- albeit one that is developing out of step with other regulatory efforts and incomplete in its consideration of transparency, accountability, and privacy protections. The paper concludes by drawing lessons from AI procurement scholarship, while identifying distinct challenges in grantmaking that invite further study.", "date": "2025-05-13", "categories": []}
{"id": "2505.15851", "title": "Exploring Moral Exercises for Human Oversight of AI systems: Insights from Three Pilot Studies", "abstract": "This paper elaborates on the concept of moral exercises as a means to help AI actors cultivate virtues that enable effective human oversight of AI systems. We explore the conceptual framework and significance of moral exercises, situating them within the contexts of philosophical discourse, ancient practices, and contemporary AI ethics scholarship. We outline the core pillars of the moral exercises methodology - eliciting an engaged personal disposition, fostering relational understanding, and cultivating technomoral wisdom - and emphasize their relevance to key activities and competencies essential for human oversight of AI systems. Our argument is supported by findings from three pilot studies involving a company, a multidisciplinary team of AI researchers, and higher education students. These studies allow us to explore both the potential and the limitations of moral exercises. Based on the collected data, we offer insights into how moral exercises can foster a responsible AI culture within organizations, and suggest directions for future research.", "date": "2025-05-20", "categories": []}
{"id": "2505.15466", "title": "A Participatory Strategy for AI Ethics in Education and Rehabilitation grounded in the Capability Approach", "abstract": "AI-based technologies have significant potential to enhance inclusive education and clinical-rehabilitative contexts for children with Special Educational Needs and Disabilities. AI can enhance learning experiences, empower students, and support both teachers and rehabilitators. However, their usage presents challenges that require a systemic-ecological vision, ethical considerations, and participatory research. Therefore, research and technological development must be rooted in a strong ethical-theoretical framework. The Capability Approach - a theoretical model of disability, human vulnerability, and inclusion - offers a more relevant perspective on functionality, effectiveness, and technological adequacy in inclusive learning environments. In this paper, we propose a participatory research strategy with different stakeholders through a case study on the ARTIS Project, which develops an AI-enriched interface to support children with text comprehension difficulties. Our research strategy integrates ethical, educational, clinical, and technological expertise in designing and implementing AI-based technologies for children's learning environments through focus groups and collaborative design sessions. We believe that this holistic approach to AI adoption in education can help bridge the gap between technological innovation and ethical responsibility.", "date": "2025-05-21", "categories": []}
{"id": "2408.12613", "title": "Artificial Intelligence in Election Campaigns: Perceptions, Penalties, and Implications", "abstract": "As political parties around the world experiment with Artificial Intelligence (AI) in election campaigns, concerns about deception and manipulation are rising. This article examines how the public reacts to different uses of AI in elections and the potential consequences for party evaluations and regulatory preferences. Across three preregistered studies with over 7,600 American respondents, we identify three categories of AI use -- campaign operations, voter outreach, and deception. While people generally dislike AI in campaigns, they are especially critical of deceptive uses, which they perceive as norm violations. However, parties engaging in AI-enabled deception face no significant drop in favorability, neither with supporters nor opponents. Instead, deceptive AI use increases public support for stricter AI regulation, including calls for an outright ban on AI development. These findings reveal a misalignment between public disapproval of deceptive AI and the political incentives of parties, underscoring the need for targeted regulatory oversight. Rather than banning AI in elections altogether, regulation should distinguish between harmful and beneficial applications to avoid stifling democratic innovation.", "date": "2024-08-08", "categories": []}
{"id": "2411.08881", "title": "Can We Trust AI Agents? A Case Study of an LLM-Based Multi-Agent System for Ethical AI", "abstract": "AI-based systems, including Large Language Models (LLM), impact millions by supporting diverse tasks but face issues like misinformation, bias, and misuse. AI ethics is crucial as new technologies and concerns emerge, but objective, practical guidance remains debated. This study examines the use of LLMs for AI ethics in practice, assessing how LLM trustworthiness-enhancing techniques affect software development in this context. Using the Design Science Research (DSR) method, we identify techniques for LLM trustworthiness: multi-agents, distinct roles, structured communication, and multiple rounds of debate. We design a multi-agent prototype LLM-MAS, where agents engage in structured discussions on real-world AI ethics issues from the AI Incident Database. We evaluate the prototype across three case scenarios using thematic analysis, hierarchical clustering, comparative (baseline) studies, and running source code. The system generates approximately 2,000 lines of code per case, compared to only 80 lines in baseline trials. Discussions reveal terms like bias detection, transparency, accountability, user consent, GDPR compliance, fairness evaluation, and EU AI Act compliance, showing this prototype ability to generate extensive source code and documentation addressing often overlooked AI ethics issues. However, practical challenges in source code integration and dependency management may limit its use by practitioners.", "date": "2024-10-25", "categories": []}
{"id": "2505.09747", "title": "Healthy Distrust in AI systems", "abstract": "Under the slogan of trustworthy AI, much of contemporary AI research is focused on designing AI systems and usage practices that inspire human trust and, thus, enhance adoption of AI systems. However, a person affected by an AI system may not be convinced by AI system design alone -- neither should they, if the AI system is embedded in a social context that gives good reason to believe that it is used in tension with a person's interest. In such cases, distrust in the system may be justified and necessary to build meaningful trust in the first place. We propose the term \"healthy distrust\" to describe such a justified, careful stance towards certain AI usage practices. We investigate prior notions of trust and distrust in computer science, sociology, history, psychology, and philosophy, outline a remaining gap that healthy distrust might fill and conceptualize healthy distrust as a crucial part for AI usage that respects human autonomy.", "date": "2025-05-14", "categories": []}
{"id": "2505.10300", "title": "AI LEGO: Scaffolding Cross-Functional Collaboration in Industrial Responsible AI Practices during Early Design Stages", "abstract": "Responsible AI (RAI) efforts increasingly emphasize the importance of addressing potential harms early in the AI development lifecycle through social-technical lenses. However, in cross-functional industry teams, this work is often stalled by a persistent knowledge handoff challenge: the difficulty of transferring high-level, early-stage technical design rationales from technical experts to non-technical or user-facing roles for ethical evaluation and harm identification. Through literature review and a co-design study with 8 practitioners, we unpack how this challenge manifests -- technical design choices are rarely handed off in ways that support meaningful engagement by non-technical roles; collaborative workflows lack shared, visual structures to support mutual understanding; and non-technical practitioners are left without scaffolds for systematic harm evaluation. Existing tools like JIRA or Google Docs, while useful for product tracking, are ill-suited for supporting joint harm identification across roles, often requiring significant extra effort to align understanding. To address this, we developed AI LEGO, a web-based prototype that supports cross-functional AI practitioners in effectively facilitating knowledge handoff and identifying harmful design choices in the early design stages. Technical roles use interactive blocks to draft development plans, while non-technical roles engage with those blocks through stage-specific checklists and LLM-driven persona simulations to surface potential harms. In a study with 18 cross-functional practitioners, AI LEGO increased the volume and likelihood of harms identified compared to baseline worksheets. Participants found that its modular structure and persona prompts made harm identification more accessible, fostering clearer and more collaborative RAI practices in early design.", "date": "2025-05-15", "categories": []}
{"id": "2406.16696", "title": "Public Constitutional AI", "abstract": "We are increasingly subjected to the power of AI authorities. As AI decisions become inescapable, entering domains such as healthcare, education, and law, we must confront a vital question: how can we ensure AI systems have the legitimacy necessary for effective governance? This essay argues that to secure AI legitimacy, we need methods that engage the public in designing and constraining AI systems, ensuring these technologies reflect the community's shared values. Constitutional AI, proposed by Anthropic, represents a step towards this goal, offering a model for democratic control of AI. However, while Constitutional AI's commitment to hardcoding explicit principles into AI models enhances transparency and accountability, it falls short in two crucial aspects: addressing the opacity of individual AI decisions and fostering genuine democratic legitimacy. To overcome these limitations, this essay proposes \"Public Constitutional AI.\" This approach envisions a participatory process where diverse stakeholders, including ordinary citizens, deliberate on the principles guiding AI development. The resulting \"AI Constitution\" would carry the legitimacy of popular authorship, grounding AI governance in the public will. Furthermore, the essay proposes \"AI Courts\" to develop \"AI case law,\" providing concrete examples for operationalizing constitutional principles in AI training. This evolving combination of constitutional principles and case law aims to make AI governance more responsive to public values. By grounding AI governance in deliberative democratic processes, Public Constitutional AI offers a path to imbue automated authorities with genuine democratic legitimacy, addressing the unique challenges posed by increasingly powerful AI systems while ensuring their alignment with the public interest.", "date": "2024-06-24", "categories": []}
{"id": "2505.09576", "title": "Ethics and Persuasion in Reinforcement Learning from Human Feedback: A Procedural Rhetorical Approach", "abstract": "Since 2022, versions of generative AI chatbots such as ChatGPT and Claude have been trained using a specialized technique called Reinforcement Learning from Human Feedback (RLHF) to fine-tune language model output using feedback from human annotators. As a result, the integration of RLHF has greatly enhanced the outputs of these large language models (LLMs) and made the interactions and responses appear more \"human-like\" than those of previous versions using only supervised learning. The increasing convergence of human and machine-written text has potentially severe ethical, sociotechnical, and pedagogical implications relating to transparency, trust, bias, and interpersonal relations. To highlight these implications, this paper presents a rhetorical analysis of some of the central procedures and processes currently being reshaped by RLHF-enhanced generative AI chatbots: upholding language conventions, information seeking practices, and expectations for social relationships. Rhetorical investigations of generative AI and LLMs have, to this point, focused largely on the persuasiveness of the content generated. Using Ian Bogost's concept of procedural rhetoric, this paper shifts the site of rhetorical investigation from content analysis to the underlying mechanisms of persuasion built into RLHF-enhanced LLMs. In doing so, this theoretical investigation opens a new direction for further inquiry in AI ethics that considers how procedures rerouted through AI-driven technologies might reinforce hegemonic language use, perpetuate biases, decontextualize learning, and encroach upon human relationships. It will therefore be of interest to educators, researchers, scholars, and the growing number of users of generative AI chatbots.", "date": "2025-05-14", "categories": []}
{"id": "2505.08064", "title": "Justified Evidence Collection for Argument-based AI Fairness Assurance", "abstract": "It is well recognised that ensuring fair AI systems is a complex sociotechnical challenge, which requires careful deliberation and continuous oversight across all stages of a system's lifecycle, from defining requirements to model deployment and deprovisioning. Dynamic argument-based assurance cases, which present structured arguments supported by evidence, have emerged as a systematic approach to evaluating and mitigating safety risks and hazards in AI-enabled system development and have also been extended to deal with broader normative goals such as fairness and explainability. This paper introduces a systems-engineering-driven framework, supported by software tooling, to operationalise a dynamic approach to argument-based assurance in two stages. In the first stage, during the requirements planning phase, a multi-disciplinary and multi-stakeholder team define goals and claims to be established (and evidenced) by conducting a comprehensive fairness governance process. In the second stage, a continuous monitoring interface gathers evidence from existing artefacts (e.g. metrics from automated tests), such as model, data, and use case documentation, to support these arguments dynamically. The framework's effectiveness is demonstrated through an illustrative case study in finance, with a focus on supporting fairness-related arguments.", "date": "2025-05-12", "categories": []}
{"id": "2505.06464", "title": "Opening the Scope of Openness in AI", "abstract": "The concept of openness in AI has so far been heavily inspired by the definition and community practice of open source software. This positions openness in AI as having positive connotations; it introduces assumptions of certain advantages, such as collaborative innovation and transparency. However, the practices and benefits of open source software are not fully transferable to AI, which has its own challenges. Framing a notion of openness tailored to AI is crucial to addressing its growing societal implications, risks, and capabilities. We argue that considering the fundamental scope of openness in different disciplines will broaden discussions, introduce important perspectives, and reflect on what openness in AI should mean. Toward this goal, we qualitatively analyze 98 concepts of openness discovered from topic modeling, through which we develop a taxonomy of openness. Using this taxonomy as an instrument, we situate the current discussion on AI openness, identify gaps and highlight links with other disciplines. Our work contributes to the recent efforts in framing openness in AI by reflecting principles and practices of openness beyond open source software and calls for a more holistic view of openness in terms of actions, system properties, and ethical objectives.", "date": "2025-05-09", "categories": []}
{"id": "2505.07178", "title": "Accountability of Generative AI: Exploring a Precautionary Approach for \"Artificially Created Nature\"", "abstract": "The rapid development of generative artificial intelligence (AI) technologies raises concerns about the accountability of sociotechnical systems. Current generative AI systems rely on complex mechanisms that make it difficult for even experts to fully trace the reasons behind the outputs. This paper first examines existing research on AI transparency and accountability and argues that transparency is not a sufficient condition for accountability but can contribute to its improvement. We then discuss that if it is not possible to make generative AI transparent, generative AI technology becomes ``artificially created nature'' in a metaphorical sense, and suggest using the precautionary principle approach to consider AI risks. Finally, we propose that a platform for citizen participation is needed to address the risks of generative AI.", "date": "2025-05-12", "categories": []}
{"id": "2505.07339", "title": "Laypeople's Attitudes Towards Fair, Affirmative, and Discriminatory Decision-Making Algorithms", "abstract": "Affirmative algorithms have emerged as a potential answer to algorithmic discrimination, seeking to redress past harms and rectify the source of historical injustices. We present the results of two experiments ($N$$=$$1193$) capturing laypeople's perceptions of affirmative algorithms -- those which explicitly prioritize the historically marginalized -- in hiring and criminal justice. We contrast these opinions about affirmative algorithms with folk attitudes towards algorithms that prioritize the privileged (i.e., discriminatory) and systems that make decisions independently of demographic groups (i.e., fair). We find that people -- regardless of their political leaning and identity -- view fair algorithms favorably and denounce discriminatory systems. In contrast, we identify disagreements concerning affirmative algorithms: liberals and racial minorities rate affirmative systems as positively as their fair counterparts, whereas conservatives and those from the dominant racial group evaluate affirmative algorithms as negatively as discriminatory systems. We identify a source of these divisions: people have varying beliefs about who (if anyone) is marginalized, shaping their views of affirmative algorithms. We discuss the possibility of bridging these disagreements to bring people together towards affirmative algorithms.", "date": "2025-05-12", "categories": []}
{"id": "2505.05197", "title": "Societal and technological progress as sewing an ever-growing,\n  ever-changing, patchy, and polychrome quilt", "abstract": "  Artificial Intelligence (AI) systems are increasingly placed in positions\nwhere their decisions have real consequences, e.g., moderating online spaces,\nconducting research, and advising on policy. Ensuring they operate in a safe\nand ethically acceptable fashion is thus critical. However, most solutions have\nbeen a form of one-size-fits-all \"alignment\". We are worried that such systems,\nwhich overlook enduring moral diversity, will spark resistance, erode trust,\nand destabilize our institutions. This paper traces the underlying problem to\nan often-unstated Axiom of Rational Convergence: the idea that under ideal\nconditions, rational agents will converge in the limit of conversation on a\nsingle ethics. Treating that premise as both optional and doubtful, we propose\nwhat we call the appropriateness framework: an alternative approach grounded in\nconflict theory, cultural evolution, multi-agent systems, and institutional\neconomics. The appropriateness framework treats persistent disagreement as the\nnormal case and designs for it by applying four principles: (1) contextual\ngrounding, (2) community customization, (3) continual adaptation, and (4)\npolycentric governance. We argue here that adopting these design principles is\na good way to shift the main alignment metaphor from moral unification to a\nmore productive metaphor of conflict management, and that taking this step is\nboth desirable and urgent.\n", "date": "2025-05-08", "categories": []}
{"id": "2505.04352", "title": "Uncertain Machine Ethics Planning", "abstract": "  Machine Ethics decisions should consider the implications of uncertainty over\ndecisions. Decisions should be made over sequences of actions to reach\npreferable outcomes long term. The evaluation of outcomes, however, may invoke\none or more moral theories, which might have conflicting judgements. Each\ntheory will require differing representations of the ethical situation. For\nexample, Utilitarianism measures numerical values, Deontology analyses duties,\nand Virtue Ethics emphasises moral character. While balancing potentially\nconflicting moral considerations, decisions may need to be made, for example,\nto achieve morally neutral goals with minimal costs. In this paper, we\nformalise the problem as a Multi-Moral Markov Decision Process and a\nMulti-Moral Stochastic Shortest Path Problem. We develop a heuristic algorithm\nbased on Multi-Objective AO*, utilising Sven-Ove Hansson's Hypothetical\nRetrospection procedure for ethical reasoning under uncertainty. Our approach\nis validated by a case study from Machine Ethics literature: the problem of\nwhether to steal insulin for someone who needs it.\n", "date": "2025-05-07", "categories": []}
{"id": "2505.04592", "title": "AI Governance to Avoid Extinction: The Strategic Landscape and\n  Actionable Research Questions", "abstract": "  Humanity appears to be on course to soon develop AI systems that\nsubstantially outperform human experts in all cognitive domains and activities.\nWe believe the default trajectory has a high likelihood of catastrophe,\nincluding human extinction. Risks come from failure to control powerful AI\nsystems, misuse of AI by malicious rogue actors, war between great powers, and\nauthoritarian lock-in. This research agenda has two aims: to describe the\nstrategic landscape of AI development and to catalog important governance\nresearch questions. These questions, if answered, would provide important\ninsight on how to successfully reduce catastrophic risks.\n  We describe four high-level scenarios for the geopolitical response to\nadvanced AI development, cataloging the research questions most relevant to\neach. Our favored scenario involves building the technical, legal, and\ninstitutional infrastructure required to internationally restrict dangerous AI\ndevelopment and deployment (which we refer to as an Off Switch), which leads\ninto an internationally coordinated Halt on frontier AI activities at some\npoint in the future. The second scenario we describe is a US National Project\nfor AI, in which the US Government races to develop advanced AI systems and\nestablish unilateral control over global AI development. We also describe two\nadditional scenarios: a Light-Touch world similar to that of today and a Threat\nof Sabotage situation where countries use sabotage and deterrence to slow AI\ndevelopment.\n  In our view, apart from the Off Switch and Halt scenario, all of these\ntrajectories appear to carry an unacceptable risk of catastrophic harm. Urgent\naction is needed from the US National Security community and AI governance\necosystem to answer key research questions, build the capability to halt\ndangerous AI activities, and prepare for international AI agreements.\n", "date": "2025-05-07", "categories": []}
{"id": "2505.00174", "title": "Real-World Gaps in AI Governance Research", "abstract": "  Drawing on 1,178 safety and reliability papers from 9,439 generative AI\npapers (January 2020 - March 2025), we compare research outputs of leading AI\ncompanies (Anthropic, Google DeepMind, Meta, Microsoft, and OpenAI) and AI\nuniversities (CMU, MIT, NYU, Stanford, UC Berkeley, and University of\nWashington). We find that corporate AI research increasingly concentrates on\npre-deployment areas -- model alignment and testing & evaluation -- while\nattention to deployment-stage issues such as model bias has waned. Significant\nresearch gaps exist in high-risk deployment domains, including healthcare,\nfinance, misinformation, persuasive and addictive features, hallucinations, and\ncopyright. Without improved observability into deployed AI, growing corporate\nconcentration could deepen knowledge deficits. We recommend expanding external\nresearcher access to deployment data and systematic observability of in-market\nAI behaviors.\n", "date": "2025-04-30", "categories": []}
{"id": "2407.19655", "title": "AI-Driven Healthcare: A Review on Ensuring Fairness and Mitigating Bias", "abstract": "  Artificial intelligence (AI) is rapidly advancing in healthcare, enhancing\nthe efficiency and effectiveness of services across various specialties,\nincluding cardiology, ophthalmology, dermatology, emergency medicine, etc. AI\napplications have significantly improved diagnostic accuracy, treatment\npersonalization, and patient outcome predictions by leveraging technologies\nsuch as machine learning, neural networks, and natural language processing.\nHowever, these advancements also introduce substantial ethical and fairness\nchallenges, particularly related to biases in data and algorithms. These biases\ncan lead to disparities in healthcare delivery, affecting diagnostic accuracy\nand treatment outcomes across different demographic groups. This review paper\nexamines the integration of AI in healthcare, highlighting critical challenges\nrelated to bias and exploring strategies for mitigation. We emphasize the\nnecessity of diverse datasets, fairness-aware algorithms, and regulatory\nframeworks to ensure equitable healthcare delivery. The paper concludes with\nrecommendations for future research, advocating for interdisciplinary\napproaches, transparency in AI decision-making, and the development of\ninnovative and inclusive AI applications.\n", "date": "2024-07-29", "categories": []}
{"id": "2503.04785", "title": "Mapping Trustworthiness in Large Language Models: A Bibliometric\n  Analysis Bridging Theory to Practice", "abstract": "  The rapid proliferation of Large Language Models (LLMs) has raised\nsignificant trustworthiness and ethical concerns. Despite the widespread\nadoption of LLMs across domains, there is still no clear consensus on how to\ndefine and operationalise trustworthiness. This study aims to bridge the gap\nbetween theoretical discussion and practical implementation by analysing\nresearch trends, definitions of trustworthiness, and practical techniques. We\nconducted a bibliometric mapping analysis of 2,006 publications from Web of\nScience (2019-2025) using the Bibliometrix, and manually reviewed 68 papers. We\nfound a shift from traditional AI ethics discussion to LLM trustworthiness\nframeworks. We identified 18 different definitions of trust/trustworthiness,\nwith transparency, explainability and reliability emerging as the most common\ndimensions. We identified 20 strategies to enhance LLM trustworthiness, with\nfine-tuning and retrieval-augmented generation (RAG) being the most prominent.\nMost of the strategies are developer-driven and applied during the\npost-training phase. Several authors propose fragmented terminologies rather\nthan unified frameworks, leading to the risks of \"ethics washing,\" where\nethical discourse is adopted without a genuine regulatory commitment. Our\nfindings highlight: persistent gaps between theoretical taxonomies and\npractical implementation, the crucial role of the developer in operationalising\ntrust, and call for standardised frameworks and stronger regulatory measures to\nenable trustworthy and ethical deployment of LLMs.\n", "date": "2025-02-27", "categories": []}
{"id": "2505.02062", "title": "Ethical AI in the Healthcare Sector: Investigating Key Drivers of\n  Adoption through the Multi-Dimensional Ethical AI Adoption Model (MEAAM)", "abstract": "  The adoption of Artificial Intelligence (AI) in the healthcare service\nindustry presents numerous ethical challenges, yet current frameworks often\nfail to offer a comprehensive, empirical understanding of the multidimensional\nfactors influencing ethical AI integration. Addressing this critical research\ngap, this study introduces the Multi-Dimensional Ethical AI Adoption Model\n(MEAAM), a novel theoretical framework that categorizes 13 critical ethical\nvariables across four foundational dimensions of Ethical AI Fair AI,\nResponsible AI, Explainable AI, and Sustainable AI. These dimensions are\nfurther analyzed through three core ethical lenses: epistemic concerns (related\nto knowledge, transparency, and system trustworthiness), normative concerns\n(focused on justice, autonomy, dignity, and moral obligations), and overarching\nconcerns (highlighting global, systemic, and long-term ethical implications).\nThis study adopts a quantitative, cross-sectional research design using survey\ndata collected from healthcare professionals and analyzed via Partial Least\nSquares Structural Equation Modeling (PLS-SEM). Employing PLS-SEM, this study\nempirically investigates the influence of these ethical constructs on two\noutcomes Operational AI Adoption and Systemic AI Adoption. Results indicate\nthat normative concerns most significantly drive operational adoption\ndecisions, while overarching concerns predominantly shape systemic adoption\nstrategies and governance frameworks. Epistemic concerns play a facilitative\nrole, enhancing the impact of ethical design principles on trust and\ntransparency in AI systems. By validating the MEAAM framework, this research\nadvances a holistic, actionable approach to ethical AI adoption in healthcare\nand provides critical insights for policymakers, technologists, and healthcare\nadministrators striving to implement ethically grounded AI solutions.\n", "date": "2025-05-04", "categories": []}
{"id": "2505.02609", "title": "Study of the influence of a biased database on the prediction of\n  standard algorithms for selecting the best candidate for an interview", "abstract": "  Artificial intelligence is used at various stages of the recruitment process\nto automatically select the best candidate for a position, with companies\nguaranteeing unbiased recruitment. However, the algorithms used are either\ntrained by humans or are based on learning from past experiences that were\nbiased. In this article, we propose to generate data mimicking external\n(discrimination) and internal biases (self-censorship) in order to train five\nclassic algorithms and to study the extent to which they do or do not find the\nbest candidates according to objective criteria. In addition, we study the\ninfluence of the anonymisation of files on the quality of predictions.\n", "date": "2025-05-05", "categories": []}
{"id": "2505.00204", "title": "RAIL in the Wild: Operationalizing Responsible AI Evaluation Using\n  Anthropic's Value Dataset", "abstract": "  As AI systems become embedded in real-world applications, ensuring they meet\nethical standards is crucial. While existing AI ethics frameworks emphasize\nfairness, transparency, and accountability, they often lack actionable\nevaluation methods. This paper introduces a systematic approach using the\nResponsible AI Labs (RAIL) framework, which includes eight measurable\ndimensions to assess the normative behavior of large language models (LLMs). We\napply this framework to Anthropic's \"Values in the Wild\" dataset, containing\nover 308,000 anonymized conversations with Claude and more than 3,000 annotated\nvalue expressions. Our study maps these values to RAIL dimensions, computes\nsynthetic scores, and provides insights into the ethical behavior of LLMs in\nreal-world use.\n", "date": "2025-04-30", "categories": []}
{"id": "2504.21400", "title": "Who Gets the Callback? Generative AI and Gender Bias", "abstract": "  Generative artificial intelligence (AI), particularly large language models\n(LLMs), is being rapidly deployed in recruitment and for candidate\nshortlisting. We audit several mid-sized open-source LLMs for gender bias using\na dataset of 332,044 real-world online job postings. For each posting, we\nprompt the model to recommend whether an equally qualified male or female\ncandidate should receive an interview callback. We find that most models tend\nto favor men, especially for higher-wage roles. Mapping job descriptions to the\nStandard Occupational Classification system, we find lower callback rates for\nwomen in male-dominated occupations and higher rates in female-associated ones,\nindicating occupational segregation. A comprehensive analysis of linguistic\nfeatures in job ads reveals strong alignment of model recommendations with\ntraditional gender stereotypes. To examine the role of recruiter identity, we\nsteer model behavior by infusing Big Five personality traits and simulating the\nperspectives of historical figures. We find that less agreeable personas reduce\nstereotyping, consistent with an agreeableness bias in LLMs. Our findings\nhighlight how AI-driven hiring may perpetuate biases in the labor market and\nhave implications for fairness and diversity within firms.\n", "date": "2025-04-30", "categories": []}
{"id": "2504.21848", "title": "Characterizing AI Agents for Alignment and Governance", "abstract": "  The creation of effective governance mechanisms for AI agents requires a\ndeeper understanding of their core properties and how these properties relate\nto questions surrounding the deployment and operation of agents in the world.\nThis paper provides a characterization of AI agents that focuses on four\ndimensions: autonomy, efficacy, goal complexity, and generality. We propose\ndifferent gradations for each dimension, and argue that each dimension raises\nunique questions about the design, operation, and governance of these systems.\nMoreover, we draw upon this framework to construct \"agentic profiles\" for\ndifferent kinds of AI agents. These profiles help to illuminate cross-cutting\ntechnical and non-technical governance challenges posed by different classes of\nAI agents, ranging from narrow task-specific assistants to highly autonomous\ngeneral-purpose systems. By mapping out key axes of variation and continuity,\nthis framework provides developers, policymakers, and members of the public\nwith the opportunity to develop governance approaches that better align with\ncollective societal goals.\n", "date": "2025-04-30", "categories": []}
{"id": "2504.21849", "title": "Public Opinion and The Rise of Digital Minds: Perceived Risk, Trust, and\n  Regulation Support", "abstract": "  Governance institutions must respond to societal risks, including those posed\nby generative AI. This study empirically examines how public trust in\ninstitutions and AI technologies, along with perceived risks, shape preferences\nfor AI regulation. Using the nationally representative 2023 Artificial\nIntelligence, Morality, and Sentience (AIMS) survey, we assess trust in\ngovernment, AI companies, and AI technologies, as well as public support for\nregulatory measures such as slowing AI development or outright bans on advanced\nAI. Our findings reveal broad public support for AI regulation, with risk\nperception playing a significant role in shaping policy preferences.\nIndividuals with higher trust in government favor regulation, while those with\ngreater trust in AI companies and AI technologies are less inclined to support\nrestrictions. Trust in government and perceived risks significantly predict\npreferences for both soft (e.g., slowing development) and strong (e.g., banning\nAI systems) regulatory interventions. These results highlight the importance of\npublic opinion in AI governance. As AI capabilities advance, effective\nregulation will require balancing public concerns about risks with trust in\ninstitutions. This study provides a foundational empirical baseline for\npolicymakers navigating AI governance and underscores the need for further\nresearch into public trust, risk perception, and regulatory strategies in the\nevolving AI landscape.\n", "date": "2025-04-30", "categories": []}
{"id": "2406.18900", "title": "The Rise of Artificial Intelligence in Educational Measurement:\n  Opportunities and Ethical Challenges", "abstract": "  The integration of artificial intelligence (AI) in educational measurement\nhas revolutionized assessment methods, enabling automated scoring, rapid\ncontent analysis, and personalized feedback through machine learning and\nnatural language processing. These advancements provide timely, consistent\nfeedback and valuable insights into student performance, thereby enhancing the\nassessment experience. However, the deployment of AI in education also raises\nsignificant ethical concerns regarding validity, reliability, transparency,\nfairness, and equity. Issues such as algorithmic bias and the opacity of AI\ndecision-making processes pose risks of perpetuating inequalities and affecting\nassessment outcomes. Responding to these concerns, various stakeholders,\nincluding educators, policymakers, and organizations, have developed guidelines\nto ensure ethical AI use in education. The National Council of Measurement in\nEducation's Special Interest Group on AI in Measurement and Education (AIME)\nalso focuses on establishing ethical standards and advancing research in this\narea. In this paper, a diverse group of AIME members examines the ethical\nimplications of AI-powered tools in educational measurement, explores\nsignificant challenges such as automation bias and environmental impact, and\nproposes solutions to ensure AI's responsible and effective use in education.\n", "date": "2024-06-27", "categories": []}
{"id": "2504.18807", "title": "Clones in the Machine: A Feminist Critique of Agency in Digital Cloning", "abstract": "  This paper critiques digital cloning in academic research, highlighting how\nit exemplifies AI solutionism. Digital clones, which replicate user data to\nsimulate behavior, are often seen as scalable tools for behavioral insights.\nHowever, this framing obscures ethical concerns around consent, agency, and\nrepresentation. Drawing on feminist theories of agency, the paper argues that\ndigital cloning oversimplifies human complexity and risks perpetuating systemic\nbiases. To address these issues, it proposes decentralized data repositories\nand dynamic consent models, promoting ethical, context-aware AI practices that\nchallenge the reductionist logic of AI solutionism\n", "date": "2025-04-26", "categories": []}
{"id": "2504.19255", "title": "The Convergent Ethics of AI? Analyzing Moral Foundation Priorities in\n  Large Language Models with a Multi-Framework Approach", "abstract": "  As large language models (LLMs) are increasingly deployed in consequential\ndecision-making contexts, systematically assessing their ethical reasoning\ncapabilities becomes a critical imperative. This paper introduces the\nPriorities in Reasoning and Intrinsic Moral Evaluation (PRIME) framework--a\ncomprehensive methodology for analyzing moral priorities across foundational\nethical dimensions including consequentialist-deontological reasoning, moral\nfoundations theory, and Kohlberg's developmental stages. We apply this\nframework to six leading LLMs through a dual-protocol approach combining direct\nquestioning and response analysis to established ethical dilemmas. Our analysis\nreveals striking patterns of convergence: all evaluated models demonstrate\nstrong prioritization of care/harm and fairness/cheating foundations while\nconsistently underweighting authority, loyalty, and sanctity dimensions.\nThrough detailed examination of confidence metrics, response reluctance\npatterns, and reasoning consistency, we establish that contemporary LLMs (1)\nproduce decisive ethical judgments, (2) demonstrate notable cross-model\nalignment in moral decision-making, and (3) generally correspond with\nempirically established human moral preferences. This research contributes a\nscalable, extensible methodology for ethical benchmarking while highlighting\nboth the promising capabilities and systematic limitations in current AI moral\nreasoning architectures--insights critical for responsible development as these\nsystems assume increasingly significant societal roles.\n", "date": "2025-04-27", "categories": []}
{"id": "2504.19264", "title": "Navigating AI Policy Landscapes: Insights into Human Rights\n  Considerations Across IEEE Regions", "abstract": "  This paper explores the integration of human rights considerations into AI\nregulatory frameworks across different IEEE regions - specifically the United\nStates (Region 1-6), Europe (Region 8), China (part of Region 10), and\nSingapore (part of Region 10). While all acknowledge the transformative\npotential of AI and the necessity of ethical guidelines, their regulatory\napproaches significantly differ. Europe exhibits a rigorous framework with\nstringent protections for individual rights, while the U.S. promotes innovation\nwith less restrictive regulations. China emphasizes state control and societal\norder in its AI strategies. In contrast, Singapore's advisory framework\nencourages self-regulation and aligns closely with international norms. This\ncomparative analysis underlines the need for ongoing global dialogue to\nharmonize AI regulations that safeguard human rights while promoting\ntechnological advancement, reflecting the diverse perspectives and priorities\nof each region.\n", "date": "2025-04-27", "categories": []}
{"id": "2504.19275", "title": "Balancing Creativity and Automation: The Influence of AI on Modern Film\n  Production and Dissemination", "abstract": "  The integration of Artificial Intelligence(AI) into film production has\nrevolutionized efficiency and creativity, yet it simultaneously raises critical\nethical and practical challenges. This study explores the dual impact of AI on\nmodern cinema through three objectives: defining the optimal human-AI\nrelationship, balancing creativity with automation, and developing ethical\nguidelines. By employing a mixed-method approach combining theoretical\nframeworks (auteur theory, human-technology relations) and case studies (The\nSafe Zone, Fast & Furious 7, The Brutalist), the research reveals that\npositioning AI as an \"embodiment tool\" rather than an independent \"alterity\npartner\" preserves human authorship and artistic integrity. Key findings\nhighlight the risks of surveillance capitalism in AI-driven markets and the\nethical dilemmas of deepfake technology. The study concludes with actionable\nrecommendations, including international regulatory frameworks and a Human\nControl Index (HCI) to quantify AI involvement. These insights aim to guide\nfilmmakers, policymakers, and scholars in navigating the evolving AI-cinema\nlandscape while safeguarding cultural diversity and ethical standards.\n", "date": "2025-04-27", "categories": []}
{"id": "2504.19284", "title": "Ethical Challenges of Using Artificial Intelligence in Judiciary", "abstract": "  Artificial intelligence (AI) has emerged as a ubiquitous concept in numerous\ndomains, including the legal system. AI has the potential to revolutionize the\nfunctioning of the judiciary and the dispensation of justice. Incorporating AI\ninto the legal system offers the prospect of enhancing decision-making for\njudges, lawyers, and legal professionals, while concurrently providing the\npublic with more streamlined, efficient, and cost-effective services. The\nintegration of AI into the legal landscape offers manifold benefits,\nencompassing tasks such as document review, legal research, contract analysis,\ncase prediction, and decision-making. By automating laborious and error-prone\nprocedures, AI has the capacity to alleviate the burden associated with these\narduous tasks. Consequently, courts around the world have begun embracing AI\ntechnology as a means to enhance the administration of justice. However,\nalongside its potential advantages, the use of AI in the judiciary poses a\nrange of ethical challenges. These ethical quandaries must be duly addressed to\nensure the responsible and equitable deployment of AI systems. This article\ndelineates the principal ethical challenges entailed in employing AI within the\njudiciary and provides recommendations to effectively address these issues.\n", "date": "2025-04-27", "categories": []}
{"id": "2412.04476", "title": "The Moral Mind(s) of Large Language Models", "abstract": "  As large language models (LLMs) increasingly participate in tasks with\nethical and societal stakes, a critical question arises: do they exhibit an\nemergent \"moral mind\" - a consistent structure of moral preferences guiding\ntheir decisions - and to what extent is this structure shared across models? To\ninvestigate this, we applied tools from revealed preference theory to nearly 40\nleading LLMs, presenting each with many structured moral dilemmas spanning five\nfoundational dimensions of ethical reasoning. Using a probabilistic rationality\ntest, we found that at least one model from each major provider exhibited\nbehavior consistent with approximately stable moral preferences, acting as if\nguided by an underlying utility function. We then estimated these utility\nfunctions and found that most models cluster around neutral moral stances. To\nfurther characterize heterogeneity, we employed a non-parametric permutation\napproach, constructing a probabilistic similarity network based on revealed\npreference patterns. The results reveal a shared core in LLMs' moral reasoning,\nbut also meaningful variation: some models show flexible reasoning across\nperspectives, while others adhere to more rigid ethical profiles. These\nfindings provide a new empirical lens for evaluating moral consistency in LLMs\nand offer a framework for benchmarking ethical alignment across AI systems.\n", "date": "2024-11-19", "categories": []}
{"id": "2501.16606", "title": "Can We Govern the Agent-to-Agent Economy?", "abstract": "  Current approaches to AI governance often fall short in anticipating a future\nwhere AI agents manage critical tasks, such as financial operations,\nadministrative functions, and beyond. While cryptocurrencies could serve as the\nfoundation for monetizing value exchange in a collaboration and delegation\ndynamic among AI agents, a critical question remains: how can humans ensure\nmeaningful oversight and control as a future economy of AI agents scales and\nevolves? In this philosophical exploration, we highlight emerging concepts in\nthe industry to inform research and development efforts in anticipation of a\nfuture decentralized agentic economy.\n", "date": "2025-01-28", "categories": []}
{"id": "2504.17823", "title": "The Cloud Weaving Model for AI development", "abstract": "  While analysing challenges in pilot projects developing AI with marginalized\ncommunities, we found it difficult to express them within commonly used\nparadigms. We therefore constructed an alternative conceptual framework to\nground AI development in the social fabric -- the Cloud Weaving Model --\ninspired (amongst others) by indigenous knowledge, motifs from nature, and\nEastern traditions. This paper introduces and elaborates on the fundamental\nelements of the model (clouds, spiders, threads, spiderwebs, and weather) and\ntheir interpretation in an AI context. The framework is then applied to\ncomprehend patterns observed in co-creation pilots approaching marginalized\ncommunities, highlighting neglected yet relevant dimensions for responsible AI\ndevelopment.\n", "date": "2025-04-23", "categories": []}
{"id": "2504.18044", "title": "AI Ethics and Social Norms: Exploring ChatGPT's Capabilities From What\n  to How", "abstract": "  Using LLMs in healthcare, Computer-Supported Cooperative Work, and Social\nComputing requires the examination of ethical and social norms to ensure safe\nincorporation into human life. We conducted a mixed-method study, including an\nonline survey with 111 participants and an interview study with 38 experts, to\ninvestigate the AI ethics and social norms in ChatGPT as everyday life tools.\nThis study aims to evaluate whether ChatGPT in an empirical context operates\nfollowing ethics and social norms, which is critical for understanding actions\nin industrial and academic research and achieving machine ethics. The findings\nof this study provide initial insights into six important aspects of AI ethics,\nincluding bias, trustworthiness, security, toxicology, social norms, and\nethical data. Significant obstacles related to transparency and bias in\nunsupervised data collection methods are identified as ChatGPT's ethical\nconcerns.\n", "date": "2025-04-25", "categories": []}
{"id": "2504.17544", "title": "Auditing the Ethical Logic of Generative AI Models", "abstract": "  As generative AI models become increasingly integrated into high-stakes\ndomains, the need for robust methods to evaluate their ethical reasoning\nbecomes increasingly important. This paper introduces a five-dimensional audit\nmodel -- assessing Analytic Quality, Breadth of Ethical Considerations, Depth\nof Explanation, Consistency, and Decisiveness -- to evaluate the ethical logic\nof leading large language models (LLMs). Drawing on traditions from applied\nethics and higher-order thinking, we present a multi-battery prompt approach,\nincluding novel ethical dilemmas, to probe the models' reasoning across diverse\ncontexts. We benchmark seven major LLMs finding that while models generally\nconverge on ethical decisions, they vary in explanatory rigor and moral\nprioritization. Chain-of-Thought prompting and reasoning-optimized models\nsignificantly enhance performance on our audit metrics. This study introduces a\nscalable methodology for ethical benchmarking of AI systems and highlights the\npotential for AI to complement human moral reasoning in complex decision-making\ncontexts.\n", "date": "2025-04-24", "categories": []}
{"id": "2504.16255", "title": "FairPlay: A Collaborative Approach to Mitigate Bias in Datasets for\n  Improved AI Fairness", "abstract": "  The issue of fairness in decision-making is a critical one, especially given\nthe variety of stakeholder demands for differing and mutually incompatible\nversions of fairness. Adopting a strategic interaction of perspectives provides\nan alternative to enforcing a singular standard of fairness. We present a\nweb-based software application, FairPlay, that enables multiple stakeholders to\ndebias datasets collaboratively. With FairPlay, users can negotiate and arrive\nat a mutually acceptable outcome without a universally agreed-upon theory of\nfairness. In the absence of such a tool, reaching a consensus would be highly\nchallenging due to the lack of a systematic negotiation process and the\ninability to modify and observe changes. We have conducted user studies that\ndemonstrate the success of FairPlay, as users could reach a consensus within\nabout five rounds of gameplay, illustrating the application's potential for\nenhancing fairness in AI systems.\n", "date": "2025-04-22", "categories": []}
{"id": "2311.14381", "title": "Potential Societal Biases of ChatGPT in Higher Education: A Scoping\n  Review", "abstract": "  Purpose:Generative Artificial Intelligence (GAI) models, such as ChatGPT, may\ninherit or amplify societal biases due to their training on extensive datasets.\nWith the increasing usage of GAI by students, faculty, and staff in higher\neducation institutions (HEIs), it is urgent to examine the ethical issues and\npotential biases associated with these technologies.\nDesign/Approach/Methods:This scoping review aims to elucidate how biases\nrelated to GAI in HEIs have been researched and discussed in recent academic\npublications. We categorized the potential societal biases that GAI might cause\nin the field of higher education. Our review includes articles written in\nEnglish, Chinese, and Japanese across four main databases, focusing on GAI\nusage in higher education and bias. Findings:Our findings reveal that while\nthere is meaningful scholarly discussion around bias and discrimination\nconcerning LLMs in the AI field, most articles addressing higher education\napproach the issue superficially. Few articles identify specific types of bias\nunder different circumstances, and there is a notable lack of empirical\nresearch. Most papers in our review focus primarily on educational and research\nfields related to medicine and engineering, with some addressing English\neducation. However, there is almost no discussion regarding the humanities and\nsocial sciences. Additionally, a significant portion of the current discourse\nis in English and primarily addresses English-speaking contexts.\nOriginality/Value:To the best of our knowledge, our study is the first to\nsummarize the potential societal biases in higher education. This review\nhighlights the need for more in-depth studies and empirical work to understand\nthe specific biases that GAI might introduce or amplify in educational\nsettings, guiding the development of more ethical AI applications in higher\neducation.\n", "date": "2023-11-24", "categories": []}
{"id": "2504.13957", "title": "Naming is framing: How cybersecurity's language problems are repeating\n  in AI governance", "abstract": "  Language is not neutral; it frames understanding, structures power, and\nshapes governance. This paper argues that misnomers like cybersecurity and\nartificial intelligence (AI) are more than semantic quirks; they carry\nsignificant governance risks by obscuring human agency, inflating expectations,\nand distorting accountability. Drawing on lessons from cybersecurity's\nlinguistic pitfalls, such as the 'weakest link' narrative, this paper\nhighlights how AI discourse is falling into similar traps with metaphors like\n'alignment,' 'black box,' and 'hallucination.' These terms embed adversarial,\nmystifying, or overly technical assumptions into governance structures. In\nresponse, the paper advocates for a language-first approach to AI governance:\none that interrogates dominant metaphors, foregrounds human roles, and\nco-develops a lexicon that is precise, inclusive, and reflexive. This paper\ncontends that linguistic reform is not peripheral to governance but central to\nthe construction of transparent, equitable, and anticipatory regulatory\nframeworks.\n", "date": "2025-04-16", "categories": []}
{"id": "2504.13979", "title": "Framework, Standards, Applications and Best practices of Responsible AI\n  : A Comprehensive Survey", "abstract": "  Responsible Artificial Intelligence (RAI) is a combination of ethics\nassociated with the usage of artificial intelligence aligned with the common\nand standard frameworks. This survey paper extensively discusses the global and\nnational standards, applications of RAI, current technology and ongoing\nprojects using RAI, and possible challenges in implementing and designing RAI\nin the industries and projects based on AI. Currently, ethical standards and\nimplementation of RAI are decoupled which caters each industry to follow their\nown standards to use AI ethically. Many global firms and government\norganizations are taking necessary initiatives to design a common and standard\nframework. Social pressure and unethical way of using AI forces the RAI design\nrather than implementation.\n", "date": "2025-04-18", "categories": []}
{"id": "2504.14936", "title": "Giving AI a voice: how does AI think it should be treated?", "abstract": "  With the astounding progress in (generative) artificial intelligence (AI),\nthere has been significant public discourse regarding regulation and ethics of\nthe technology. Is it sufficient when humans discuss this with other humans?\nOr, given that AI is increasingly becoming a viable source of inspiration for\npeople (and let alone the hypothetical possibility that the technology may at\nsome point become \"artificial general intelligence\" and/or develop\nconsciousness), should AI not join the discourse? There are new questions and\nangles that AI brings to the table that we might not have considered before -\nso let us make the key subject of this book an active participant. This chapter\ntherefore includes a brief human-AI conversation on the topic of AI rights and\nethics.\n", "date": "2025-04-21", "categories": []}
{"id": "2209.01410", "title": "Closed-Loop View of the Regulation of AI: Equal Impact across Repeated\n  Interactions", "abstract": "  There has been much recent interest in the regulation of AI. We argue for a\nview based on civil-rights legislation, built on the notions of equal treatment\nand equal impact. In a closed-loop view of the AI system and its users, the\nequal treatment concerns one pass through the loop. Equal impact, in our view,\nconcerns the long-run average behaviour across repeated interactions. In order\nto establish the existence of the average and its properties, one needs to\nstudy the ergodic properties of the closed-loop and its unique stationary\nmeasure.\n", "date": "2022-09-03", "categories": []}
{"id": "2504.12427", "title": "Position: The Most Expensive Part of an LLM should be its Training Data", "abstract": "  Training a state-of-the-art Large Language Model (LLM) is an increasingly\nexpensive endeavor due to growing computational, hardware, energy, and\nengineering demands. Yet, an often-overlooked (and seldom paid) expense is the\nhuman labor behind these models' training data. Every LLM is built on an\nunfathomable amount of human effort: trillions of carefully written words\nsourced from books, academic papers, codebases, social media, and more. This\nposition paper aims to assign a monetary value to this labor and argues that\nthe most expensive part of producing an LLM should be the compensation provided\nto training data producers for their work. To support this position, we study\n64 LLMs released between 2016 and 2024, estimating what it would cost to pay\npeople to produce their training datasets from scratch. Even under highly\nconservative estimates of wage rates, the costs of these models' training\ndatasets are 10-1000 times larger than the costs to train the models\nthemselves, representing a significant financial liability for LLM providers.\nIn the face of the massive gap between the value of training data and the lack\nof compensation for its creation, we highlight and discuss research directions\nthat could enable fairer practices in the future.\n", "date": "2025-04-16", "categories": []}
{"id": "2504.12476", "title": "What do people expect from Artificial Intelligence? Public opinion on\n  alignment in AI moderation from Germany and the United States", "abstract": "  Recent advances in generative Artificial Intelligence have raised public\nawareness, shaping expectations and concerns about their societal implications.\nCentral to these debates is the question of AI alignment -- how well AI systems\nmeet public expectations regarding safety, fairness, and social values.\nHowever, little is known about what people expect from AI-enabled systems and\nhow these expectations differ across national contexts. We present evidence\nfrom two surveys of public preferences for key functional features of\nAI-enabled systems in Germany (n = 1800) and the United States (n = 1756). We\nexamine support for four types of alignment in AI moderation: accuracy and\nreliability, safety, bias mitigation, and the promotion of aspirational\nimaginaries. U.S. respondents report significantly higher AI use and\nconsistently greater support for all alignment features, reflecting broader\ntechnological openness and higher societal involvement with AI. In both\ncountries, accuracy and safety enjoy the strongest support, while more\nnormatively charged goals -- like fairness and aspirational imaginaries --\nreceive more cautious backing, particularly in Germany. We also explore how\nindividual experience with AI, attitudes toward free speech, political\nideology, partisan affiliation, and gender shape these preferences. AI use and\nfree speech support explain more variation in Germany. In contrast, U.S.\nresponses show greater attitudinal uniformity, suggesting that higher exposure\nto AI may consolidate public expectations. These findings contribute to debates\non AI governance and cross-national variation in public preferences. More\nbroadly, our study demonstrates the value of empirically grounding AI alignment\ndebates in public attitudes and of explicitly developing normatively grounded\nexpectations into theoretical and policy discussions on the governance of\nAI-generated content.\n", "date": "2025-04-16", "categories": []}
{"id": "2504.11501", "title": "A Framework for the Private Governance of Frontier Artificial\n  Intelligence", "abstract": "  This paper presents a proposal for the governance of frontier AI systems\nthrough a hybrid public-private system. Private bodies, authorized and overseen\nby government, provide certifications to developers of frontier AI systems on\nan opt-in basis. In exchange for opting in, frontier AI firms receive\nprotections from tort liability for customer misuse of their models. Before\ndetailing the proposal, the paper explores more commonly discussed approaches\nto AI governance, analyzing their strengths and flaws. It also examines the\nnature of frontier AI governance itself. The paper includes consideration of\nthe political economic, institutional, legal, safety, and other merits and\ntradeoffs inherent in the governance system it proposes.\n", "date": "2025-04-15", "categories": []}
{"id": "2504.09861", "title": "EthosGPT: Mapping Human Value Diversity to Advance Sustainable\n  Development Goals (SDGs)", "abstract": "  Large language models (LLMs) are transforming global decision-making and\nsocietal systems by processing diverse data at unprecedented scales. However,\ntheir potential to homogenize human values poses critical risks, similar to\nbiodiversity loss undermining ecological resilience. Rooted in the ancient\nGreek concept of ethos, meaning both individual character and the shared moral\nfabric of communities, EthosGPT draws on a tradition that spans from\nAristotle's virtue ethics to Adam Smith's moral sentiments as the ethical\nfoundation of economic cooperation. These traditions underscore the vital role\nof value diversity in fostering social trust, institutional legitimacy, and\nlong-term prosperity. EthosGPT addresses the challenge of value homogenization\nby introducing an open-source framework for mapping and evaluating LLMs within\na global scale of human values. Using international survey data on cultural\nindices, prompt-based assessments, and comparative statistical analyses,\nEthosGPT reveals both the adaptability and biases of LLMs across regions and\ncultures. It offers actionable insights for developing inclusive LLMs, such as\ndiversifying training data and preserving endangered cultural heritage to\nensure representation in AI systems. These contributions align with the United\nNations Sustainable Development Goals (SDGs), especially SDG 10 (Reduced\nInequalities), SDG 11.4 (Cultural Heritage Preservation), and SDG 16 (Peace,\nJustice and Strong Institutions). Through interdisciplinary collaboration,\nEthosGPT promotes AI systems that are both technically robust and ethically\ninclusive, advancing value plurality as a cornerstone for sustainable and\nequitable futures.\n", "date": "2025-04-14", "categories": []}
{"id": "2504.10797", "title": "Name of Thrones: Evaluating How LLMs Rank Student Names, Race, and\n  Gender in Status Hierarchies", "abstract": "  Across cultures, names tell a lot about their bearers as they carry deep\npersonal and cultural significance. Names also serve as powerful signals of\ngender, race, and status in the social hierarchy - a pecking order in which\nindividual positions shape others' expectations on their perceived competence\nand worth. With the widespread adoption of LLMs and as names are often an input\nfor LLMs, it is crucial to evaluate whether LLMs may sort people into status\npositions based on first and last names and, if so, whether it is in an unfair,\nbiased fashion. While prior work has primarily investigated biases in first\nnames, little attention has been paid to last names and even less to the\ncombined effects of first and last names. In this study, we conduct a\nlarge-scale analysis of name variations across 5 ethnicities to examine how AI\nexhibits name biases. Our study investigates three key characteristics of\ninequality and finds that LLMs reflect and reinforce status hierarchies based\non names that signal gender and ethnicity as they encode differential\nexpectations of competence, leadership, and economic potential. Contrary to the\ncommon assumption that AI tends to favor Whites, we show that East and, in some\ncontexts, South Asian names receive higher rankings. We also disaggregate\nAsians, a population projected to be the largest immigrant group in the U.S. by\n2055. Our results challenge the monolithic Asian model minority assumption,\nillustrating a more complex and stratified model of bias. Gender moderates\nbiases, with girls facing unfair disadvantages in certain racial groups.\nAdditionally, spanning cultural categories by adopting Western first names\nimproves AI-perceived status for East and Southeast Asian students,\nparticularly for girls. Our findings underscore the importance of\nintersectional and more nuanced understandings of race, gender, and mixed\nidentities in the evaluation of LLMs.\n", "date": "2025-04-15", "categories": []}
{"id": "2504.08861", "title": "Diachronic and synchronic variation in the performance of adaptive\n  machine learning systems: The ethical challenges", "abstract": "  Objectives: Machine learning (ML) has the potential to facilitate \"continual\nlearning\" in medicine, in which an ML system continues to evolve in response to\nexposure to new data over time, even after being deployed in a clinical\nsetting. In this paper, we provide a tutorial on the range of ethical issues\nraised by the use of such \"adaptive\" ML systems in medicine that have, thus\nfar, been neglected in the literature.\n  Target audience: The target audiences for this tutorial are the developers of\nmachine learning AI systems, healthcare regulators, the broader medical\ninformatics community, and practicing clinicians.\n  Scope: Discussions of adaptive ML systems to date have overlooked the\ndistinction between two sorts of variance that such systems may exhibit --\ndiachronic evolution (change over time) and synchronic variation (difference\nbetween cotemporaneous instantiations of the algorithm at different sites) --\nand under-estimated the significance of the latter. We highlight the challenges\nthat diachronic evolution and synchronic variation present for the quality of\npatient care, informed consent, and equity, and discuss the complex ethical\ntrade-offs involved in the design of such systems.\n", "date": "2025-04-11", "categories": []}
{"id": "2504.08418", "title": "seeBias: A Comprehensive Tool for Assessing and Visualizing AI Fairness", "abstract": "  Fairness in artificial intelligence (AI) prediction models is increasingly\nemphasized to support responsible adoption in high-stakes domains such as\nhealth care and criminal justice. Guidelines and implementation frameworks\nhighlight the importance of both predictive accuracy and equitable outcomes.\nHowever, current fairness toolkits often evaluate classification performance\ndisparities in isolation, with limited attention to other critical aspects such\nas calibration. To address these gaps, we present seeBias, an R package for\ncomprehensive evaluation of model fairness and predictive performance. seeBias\noffers an integrated evaluation across classification, calibration, and other\nperformance domains, providing a more complete view of model behavior. It\nincludes customizable visualizations to support transparent reporting and\nresponsible AI implementation. Using public datasets from criminal justice and\nhealthcare, we demonstrate how seeBias supports fairness evaluations, and\nuncovers disparities that conventional fairness metrics may overlook. The R\npackage is available on GitHub, and a Python version is under development.\n", "date": "2025-04-11", "categories": []}
{"id": "2408.01904", "title": "The Artificial Intelligence Disclosure (AID) Framework: An Introduction", "abstract": "  As the use of Generative Artificial Intelligence tools have grown in higher\neducation and research, there have been increasing calls for transparency and\ngranularity around the use and attribution of the use of these tools. Thus far,\nthis need has been met via the recommended inclusion of a note, with little to\nno guidance on what the note itself should include. This has been identified as\na problem to the use of AI in academic and research contexts. This article\nintroduces The Artificial Intelligence Disclosure (AID) Framework, a standard,\ncomprehensive, and detailed framework meant to inform the development and\nwriting of GenAI disclosure for education and research.\n", "date": "2024-08-04", "categories": []}
{"id": "2504.07118", "title": "Sacred or Secular? Religious Bias in AI-Generated Financial Advice", "abstract": "  This study examines religious biases in AI-generated financial advice,\nfocusing on ChatGPT's responses to financial queries. Using a prompt-based\nmethodology and content analysis, we find that 50% of the financial emails\ngenerated by ChatGPT exhibit religious biases, with explicit biases present in\nboth ingroup and outgroup interactions. While ingroup biases personalize\nresponses based on religious alignment, outgroup biases introduce religious\nframing that may alienate clients or create ideological friction. These\nfindings align with broader research on AI bias and suggest that ChatGPT is not\nmerely reflecting societal biases but actively shaping financial discourse\nbased on perceived religious identity. Using the Critical Algorithm Studies\nframework, we argue that ChatGPT functions as a mediator of financial\nnarratives, selectively reinforcing religious perspectives. This study\nunderscores the need for greater transparency, bias mitigation strategies, and\nregulatory oversight to ensure neutrality in AI-driven financial services.\n", "date": "2025-03-26", "categories": []}
{"id": "2504.07516", "title": "Enhancements for Developing a Comprehensive AI Fairness Assessment\n  Standard", "abstract": "  As AI systems increasingly influence critical sectors like\ntelecommunications, finance, healthcare, and public services, ensuring fairness\nin decision-making is essential to prevent biased or unjust outcomes that\ndisproportionately affect vulnerable entities or result in adverse impacts.\nThis need is particularly pressing as the industry approaches the 6G era, where\nAI will drive complex functions like autonomous network management and\nhyper-personalized services. The TEC Standard for Fairness Assessment and\nRating of AI Systems provides guidelines for evaluating fairness in AI,\nfocusing primarily on tabular data and supervised learning models. However, as\nAI applications diversify, this standard requires enhancement to strengthen its\nimpact and broaden its applicability. This paper proposes an expansion of the\nTEC Standard to include fairness assessments for images, unstructured text, and\ngenerative AI, including large language models, ensuring a more comprehensive\napproach that keeps pace with evolving AI technologies. By incorporating these\ndimensions, the enhanced framework will promote responsible and trustworthy AI\ndeployment across various sectors.\n", "date": "2025-04-10", "categories": []}
{"id": "2504.07763", "title": "Data over dialogue: Why artificial intelligence is unlikely to humanise\n  medicine", "abstract": "  Recently, a growing number of experts in artificial intelligence (AI) and\nmedicine have be-gun to suggest that the use of AI systems, particularly\nmachine learning (ML) systems, is likely to humanise the practice of medicine\nby substantially improving the quality of clinician-patient relationships. In\nthis thesis, however, I argue that medical ML systems are more likely to\nnegatively impact these relationships than to improve them. In particular, I\nargue that the use of medical ML systems is likely to comprise the quality of\ntrust, care, empathy, understanding, and communication between clinicians and\npatients.\n", "date": "2025-04-10", "categories": []}
{"id": "2504.07936", "title": "We Are All Creators: Generative AI, Collective Knowledge, and the Path\n  Towards Human-AI Synergy", "abstract": "  Generative AI presents a profound challenge to traditional notions of human\nuniqueness, particularly in creativity. Fueled by neural network based\nfoundation models, these systems demonstrate remarkable content generation\ncapabilities, sparking intense debates about authorship, copyright, and\nintelligence itself. This paper argues that generative AI represents an\nalternative form of intelligence and creativity, operating through mathematical\npattern synthesis rather than biological understanding or verbatim replication.\nThe fundamental differences between artificial and biological neural networks\nreveal AI learning as primarily statistical pattern extraction from vast\ndatasets crystallized forms of collective human knowledge scraped from the\ninternet. This perspective complicates copyright theft narratives and\nhighlights practical challenges in attributing AI outputs to individual\nsources. Rather than pursuing potentially futile legal restrictions, we\nadvocate for human AI synergy. By embracing generative AI as a complementary\ntool alongside human intuition, context, and ethical judgment, society can\nunlock unprecedented innovation, democratize creative expression, and address\ncomplex challenges. This collaborative approach, grounded in realistic\nunderstanding of AIs capabilities and limitations, offers the most promising\npath forward. Additionally, recognizing these models as products of collective\nhuman knowledge raises ethical questions about accessibility ensuring equitable\naccess to these tools could prevent widening societal divides and leverage\ntheir full potential for collective benefit.\n", "date": "2025-04-10", "categories": []}
{"id": "2504.06669", "title": "NLP Security and Ethics, in the Wild", "abstract": "  As NLP models are used by a growing number of end-users, an area of\nincreasing importance is NLP Security (NLPSec): assessing the vulnerability of\nmodels to malicious attacks and developing comprehensive countermeasures\nagainst them. While work at the intersection of NLP and cybersecurity has the\npotential to create safer NLP for all, accidental oversights can result in\ntangible harm (e.g., breaches of privacy or proliferation of malicious models).\nIn this emerging field, however, the research ethics of NLP have not yet faced\nmany of the long-standing conundrums pertinent to cybersecurity, until now. We\nthus examine contemporary works across NLPSec, and explore their engagement\nwith cybersecurity's ethical norms. We identify trends across the literature,\nultimately finding alarming gaps on topics like harm minimization and\nresponsible disclosure. To alleviate these concerns, we provide concrete\nrecommendations to help NLP researchers navigate this space more ethically,\nbridging the gap between traditional cybersecurity and NLP ethics, which we\nframe as ``white hat NLP''. The goal of this work is to help cultivate an\nintentional culture of ethical research for those working in NLP Security.\n", "date": "2025-04-09", "categories": []}
{"id": "2504.06016", "title": "The Hall of AI Fears and Hopes: Comparing the Views of AI Influencers\n  and those of Members of the U.S. Public Through an Interactive Platform", "abstract": "  AI development is shaped by academics and industry leaders - let us call them\n``influencers'' - but it is unclear how their views align with those of the\npublic. To address this gap, we developed an interactive platform that served\nas a data collection tool for exploring public views on AI, including their\nfears, hopes, and overall sense of hopefulness. We made the platform available\nto 330 participants representative of the U.S. population in terms of age, sex,\nethnicity, and political leaning, and compared their views with those of 100 AI\ninfluencers identified by Time magazine. The public fears AI getting out of\ncontrol, while influencers emphasize regulation, seemingly to deflect attention\nfrom their alleged focus on monetizing AI's potential. Interestingly, the views\nof AI influencers from underrepresented groups such as women and people of\ncolor often differ from the views of underrepresented groups in the public.\n", "date": "2025-04-08", "categories": []}
{"id": "2306.05949", "title": "Evaluating the Social Impact of Generative AI Systems in Systems and\n  Society", "abstract": "  Generative AI systems across modalities, ranging from text (including code),\nimage, audio, and video, have broad social impacts, but there is no official\nstandard for means of evaluating those impacts or for which impacts should be\nevaluated. In this paper, we present a guide that moves toward a standard\napproach in evaluating a base generative AI system for any modality in two\noverarching categories: what can be evaluated in a base system independent of\ncontext and what can be evaluated in a societal context. Importantly, this\nrefers to base systems that have no predetermined application or deployment\ncontext, including a model itself, as well as system components, such as\ntraining data. Our framework for a base system defines seven categories of\nsocial impact: bias, stereotypes, and representational harms; cultural values\nand sensitive content; disparate performance; privacy and data protection;\nfinancial costs; environmental costs; and data and content moderation labor\ncosts. Suggested methods for evaluation apply to listed generative modalities\nand analyses of the limitations of existing evaluations serve as a starting\npoint for necessary investment in future evaluations. We offer five overarching\ncategories for what can be evaluated in a broader societal context, each with\nits own subcategories: trustworthiness and autonomy; inequality,\nmarginalization, and violence; concentration of authority; labor and\ncreativity; and ecosystem and environment. Each subcategory includes\nrecommendations for mitigating harm.\n", "date": "2023-06-09", "categories": []}
{"id": "2504.04440", "title": "Do We Need Responsible XR? Drawing on Responsible AI to Inform Ethical\n  Research and Practice into XRAI / the Metaverse", "abstract": "  This position paper for the CHI 2025 workshop \"Everyday AR through\nAI-in-the-Loop\" reflects on whether as a field HCI needs to define Responsible\nXR as a parallel to, and in conjunction with, Responsible AI, addressing the\nunique vulnerabilities posed by mass adoption of wearable AI-enabled AR glasses\nand XR devices that could enact AI-driven human perceptual augmentation.\n", "date": "2025-04-06", "categories": []}
{"id": "2504.05007", "title": "Measuring the right thing: justifying metrics in AI impact assessments", "abstract": "  AI Impact Assessments are only as good as the measures used to assess the\nimpact of these systems. It is therefore paramount that we can justify our\nchoice of metrics in these assessments, especially for difficult to quantify\nethical and social values. We present a two-step approach to ensure metrics are\nproperly motivated. First, a conception needs to be spelled out (e.g. Rawlsian\nfairness or fairness as solidarity) and then a metric can be fitted to that\nconception. Both steps require separate justifications, as conceptions can be\njudged on how well they fit with the function of, for example, fairness. We\nargue that conceptual engineering offers helpful tools for this step. Second,\nmetrics need to be fitted to a conception. We illustrate this process through\nan examination of competing fairness metrics to illustrate that here the\nadditional content that a conception offers helps us justify the choice for a\nspecific metric. We thus advocate that impact assessments are not only clear on\ntheir metrics, but also on the conceptions that motivate those metrics.\n", "date": "2025-04-07", "categories": []}
{"id": "2504.05210", "title": "A moving target in AI-assisted decision-making: Dataset shift, model\n  updating, and the problem of update opacity", "abstract": "  Machine learning (ML) systems are vulnerable to performance decline over time\ndue to dataset shift. To address this problem, experts often suggest that ML\nsystems should be regularly updated to ensure ongoing performance stability.\nSome scholarly literature has begun to address the epistemic and ethical\nchallenges associated with different updating methodologies. Thus far, however,\nlittle attention has been paid to the impact of model updating on the\nML-assisted decision-making process itself, particularly in the AI ethics and\nAI epistemology literatures. This article aims to address this gap in the\nliterature. It argues that model updating introduces a new sub-type of opacity\ninto ML-assisted decision-making -- update opacity -- that occurs when users\ncannot understand how or why an update has changed the reasoning or behaviour\nof an ML system. This type of opacity presents a variety of distinctive\nepistemic and safety concerns that available solutions to the black box problem\nin ML are largely ill-equipped to address. A variety of alternative strategies\nmay be developed or pursued to address the problem of update opacity more\ndirectly, including bi-factual explanations, dynamic model reporting, and\nupdate compatibility. However, each of these strategies presents its own risks\nor carries significant limitations. Further research will be needed to address\nthe epistemic and safety concerns associated with model updating and update\nopacity going forward.\n", "date": "2025-04-07", "categories": []}
{"id": "2504.01043", "title": "Are clinicians ethically obligated to disclose their use of medical\n  machine learning systems to patients?", "abstract": "  It is commonly accepted that clinicians are ethically obligated to disclose\ntheir use of medical machine learning systems to patients, and that failure to\ndo so would amount to a moral fault for which clinicians ought to be held\naccountable. Call this \"the disclosure thesis.\" Four main arguments have been,\nor could be, given to support the disclosure thesis in the ethics literature:\nthe risk-based argument, the rights-based argument, the materiality argument,\nand the autonomy argument. In this article, I argue that each of these four\narguments are unconvincing, and therefore, that the disclosure thesis ought to\nbe rejected. I suggest that mandating disclosure may also even risk harming\npatients by providing stakeholders with a way to avoid accountability for harm\nthat results from improper applications or uses of these systems.\n", "date": "2025-03-31", "categories": []}
{"id": "2412.16355", "title": "Social Science Is Necessary for Operationalizing Socially Responsible\n  Foundation Models", "abstract": "  With the rise of foundation models, there is growing concern about their\npotential social impacts. Social science has a long history of studying the\nsocial impacts of transformative technologies in terms of pre-existing systems\nof power and how these systems are disrupted or reinforced by new technologies.\nIn this position paper, we build on prior work studying the social impacts of\nearlier technologies to propose a conceptual framework studying foundation\nmodels as sociotechnical systems, incorporating social science expertise to\nbetter understand how these models affect systems of power, anticipate the\nimpacts of deploying these models in various applications, and study the\neffectiveness of technical interventions intended to mitigate social harms. We\nadvocate for an interdisciplinary and collaborative research paradigm between\nAI and social science across all stages of foundation model research and\ndevelopment to promote socially responsible research practices and use cases,\nand outline several strategies to facilitate such research.\n", "date": "2024-12-20", "categories": []}
{"id": "2503.16692", "title": "Limits of trust in medical AI", "abstract": "  Artificial intelligence (AI) is expected to revolutionize the practice of\nmedicine. Recent advancements in the field of deep learning have demonstrated\nsuccess in a variety of clinical tasks: detecting diabetic retinopathy from\nimages, predicting hospital readmissions, aiding in the discovery of new drugs,\netc. AI's progress in medicine, however, has led to concerns regarding the\npotential effects of this technology upon relationships of trust in clinical\npractice. In this paper, I will argue that there is merit to these concerns,\nsince AI systems can be relied upon, and are capable of reliability, but cannot\nbe trusted, and are not capable of trustworthiness. Insofar as patients are\nrequired to rely upon AI systems for their medical decision-making, there is\npotential for this to produce a deficit of trust in relationships in clinical\npractice.\n", "date": "2025-03-20", "categories": []}
{"id": "2305.06967", "title": "Data quality dimensions for fair AI", "abstract": "  Artificial Intelligence (AI) systems are not intrinsically neutral and biases\ntrickle in any type of technological tool. In particular when dealing with\npeople, the impact of AI algorithms' technical errors originating with\nmislabeled data is undeniable. As they feed wrong and discriminatory\nclassifications, these systems are not systematically guarded against bias. In\nthis article we consider the problem of bias in AI systems from the point of\nview of data quality dimensions. We highlight the limited model construction of\nbias mitigation tools based on accuracy strategy, illustrating potential\nimprovements of a specific tool in gender classification errors occurring in\ntwo typically difficult contexts: the classification of non-binary individuals,\nfor which the label set becomes incomplete with respect to the dataset; and the\nclassification of transgender individuals, for which the dataset becomes\ninconsistent with respect to the label set. Using formal methods for reasoning\nabout the behavior of the classification system in presence of a changing\nworld, we propose to reconsider the fairness of the classification task in\nterms of completeness, consistency, timeliness and reliability, and offer some\ntheoretical results.\n", "date": "2023-05-11", "categories": []}
{"id": "2504.01420", "title": "FAIRE: Assessing Racial and Gender Bias in AI-Driven Resume Evaluations", "abstract": "  In an era where AI-driven hiring is transforming recruitment practices,\nconcerns about fairness and bias have become increasingly important. To explore\nthese issues, we introduce a benchmark, FAIRE (Fairness Assessment In Resume\nEvaluation), to test for racial and gender bias in large language models (LLMs)\nused to evaluate resumes across different industries. We use two methods-direct\nscoring and ranking-to measure how model performance changes when resumes are\nslightly altered to reflect different racial or gender identities. Our findings\nreveal that while every model exhibits some degree of bias, the magnitude and\ndirection vary considerably. This benchmark provides a clear way to examine\nthese differences and offers valuable insights into the fairness of AI-based\nhiring tools. It highlights the urgent need for strategies to reduce bias in\nAI-driven recruitment. Our benchmark code and dataset are open-sourced at our\nrepository:\nhttps://github.com/athenawen/FAIRE-Fairness-Assessment-In-Resume-Evaluation.git.\n", "date": "2025-04-02", "categories": []}
{"id": "2504.00388", "title": "Using complex prompts to identify fine-grained biases in image\n  generation through ChatGPT-4o", "abstract": "  There are not one but two dimensions of bias that can be revealed through the\nstudy of large AI models: not only bias in training data or the products of an\nAI, but also bias in society, such as disparity in employment or health\noutcomes between different demographic groups. Often training data and AI\noutput is biased for or against certain demographics (i.e. older white people\nare overrepresented in image datasets), but sometimes large AI models\naccurately illustrate biases in the real world (i.e. young black men being\ndisproportionately viewed as threatening). These social disparities often\nappear in image generation AI outputs in the form of 'marked' features, where\nsome feature of an individual or setting is a social marker of disparity, and\nprompts both humans and AI systems to treat subjects that are marked in this\nway as exceptional and requiring special treatment. Generative AI has proven to\nbe very sensitive to such marked features, to the extent of over-emphasising\nthem and thus often exacerbating social biases. I briefly discuss how we can\nuse complex prompts to image generation AI to investigate either dimension of\nbias, emphasising how we can probe the large language models underlying image\ngeneration AI through, for example, automated sentiment analysis of the text\nprompts used to generate images.\n", "date": "2025-04-01", "categories": []}
{"id": "2504.00652", "title": "Towards Adaptive AI Governance: Comparative Insights from the U.S., EU,\n  and Asia", "abstract": "  Artificial intelligence (AI) trends vary significantly across global regions,\nshaping the trajectory of innovation, regulation, and societal impact. This\nvariation influences how different regions approach AI development, balancing\ntechnological progress with ethical and regulatory considerations. This study\nconducts a comparative analysis of AI trends in the United States (US), the\nEuropean Union (EU), and Asia, focusing on three key dimensions: generative AI,\nethical oversight, and industrial applications. The US prioritizes\nmarket-driven innovation with minimal regulatory constraints, the EU enforces a\nprecautionary risk-based framework emphasizing ethical safeguards, and Asia\nemploys state-guided AI strategies that balance rapid deployment with\nregulatory oversight. Although these approaches reflect different economic\nmodels and policy priorities, their divergence poses challenges to\ninternational collaboration, regulatory harmonization, and the development of\nglobal AI standards. To address these challenges, this paper synthesizes\nregional strengths to propose an adaptive AI governance framework that\nintegrates risk-tiered oversight, innovation accelerators, and strategic\nalignment mechanisms. By bridging governance gaps, this study offers actionable\ninsights for fostering responsible AI development while ensuring a balance\nbetween technological progress, ethical imperatives, and regulatory coherence.\n", "date": "2025-04-01", "categories": []}
{"id": "2504.00899", "title": "Role and Use of Race in AI/ML Models Related to Health", "abstract": "  The role and use of race within health-related artificial intelligence and\nmachine learning (AI/ML) models has sparked increasing attention and\ncontroversy. Despite the complexity and breadth of related issues, a robust and\nholistic framework to guide stakeholders in their examination and resolution\nremains lacking. This perspective provides a broad-based, systematic, and\ncross-cutting landscape analysis of race-related challenges, structured around\nthe AI/ML lifecycle and framed through \"points to consider\" to support inquiry\nand decision-making.\n", "date": "2025-04-01", "categories": []}
{"id": "2501.08814", "title": "SAIF: A Comprehensive Framework for Evaluating the Risks of Generative\n  AI in the Public Sector", "abstract": "  The rapid adoption of generative AI in the public sector, encompassing\ndiverse applications ranging from automated public assistance to welfare\nservices and immigration processes, highlights its transformative potential\nwhile underscoring the pressing need for thorough risk assessments. Despite its\ngrowing presence, evaluations of risks associated with AI-driven systems in the\npublic sector remain insufficiently explored. Building upon an established\ntaxonomy of AI risks derived from diverse government policies and corporate\nguidelines, we investigate the critical risks posed by generative AI in the\npublic sector while extending the scope to account for its multimodal\ncapabilities. In addition, we propose a Systematic dAta generatIon Framework\nfor evaluating the risks of generative AI (SAIF). SAIF involves four key\nstages: breaking down risks, designing scenarios, applying jailbreak methods,\nand exploring prompt types. It ensures the systematic and consistent generation\nof prompt data, facilitating a comprehensive evaluation while providing a solid\nfoundation for mitigating the risks. Furthermore, SAIF is designed to\naccommodate emerging jailbreak methods and evolving prompt types, thereby\nenabling effective responses to unforeseen risk scenarios. We believe that this\nstudy can play a crucial role in fostering the safe and responsible integration\nof generative AI into the public sector.\n", "date": "2025-01-15", "categories": []}
{"id": "2503.22181", "title": "e-person Architecture and Framework for Human-AI Co-adventure\n  Relationship", "abstract": "  This paper proposes the e-person architecture for constructing a unified and\nincremental development of AI ethics. The e-person architecture takes the\nreduction of uncertainty through collaborative cognition and action with others\nas a unified basis for ethics. By classifying and defining uncertainty along\ntwo axes - (1) first, second, and third person perspectives, and (2) the\ndifficulty of inference based on the depth of information - we support the\ndevelopment of unified and incremental development of AI ethics. In addition,\nwe propose the e-person framework based on the free energy principle, which\nconsiders the reduction of uncertainty as a unifying principle of brain\nfunction, with the aim of implementing the e-person architecture, and we show\nour previous works and future challenges based on the proposed framework.\n", "date": "2025-03-28", "categories": []}
{"id": "2503.18956", "title": "International Agreements on AI Safety: Review and Recommendations for a\n  Conditional AI Safety Treaty", "abstract": "  The malicious use or malfunction of advanced general-purpose AI (GPAI) poses\nrisks that, according to leading experts, could lead to the 'marginalisation or\nextinction of humanity.' To address these risks, there are an increasing number\nof proposals for international agreements on AI safety. In this paper, we\nreview recent (2023-) proposals, identifying areas of consensus and\ndisagreement, and drawing on related literature to assess their feasibility. We\nfocus our discussion on risk thresholds, regulations, types of international\nagreement and five related processes: building scientific consensus,\nstandardisation, auditing, verification and incentivisation.\n  Based on this review, we propose a treaty establishing a compute threshold\nabove which development requires rigorous oversight. This treaty would mandate\ncomplementary audits of models, information security and governance practices,\noverseen by an international network of AI Safety Institutes (AISIs) with\nauthority to pause development if risks are unacceptable. Our approach combines\nimmediately implementable measures with a flexible structure that can adapt to\nongoing research.\n", "date": "2025-03-18", "categories": []}
{"id": "2503.18983", "title": "Confronting Catastrophic Risk: The International Obligation to Regulate\n  Artificial Intelligence", "abstract": "  While artificial intelligence (AI) holds enormous promise, many experts in\nthe field are warning that there is a non-trivial chance that the development\nof AI poses an existential threat to humanity. Existing regulatory initiative\ndo not address this threat but merely instead focus on discrete AI-related\nrisks such as consumer safety, cybersecurity, data protection, and privacy. In\nthe absence of regulatory action to address the possible risk of human\nextinction by AI, the question arises: What legal obligations, if any, does\npublic international law impose on states to regulate its development. Grounded\nin the precautionary principle, we argue that there exists an international\nobligation to mitigate the threat of human extinction by AI. Often invoked in\nrelation to environmental regulation and the regulation of potentially harmful\ntechnologies, the principle holds that in situations where there is the\npotential for significant harm, even in the absence of full scientific\ncertainty, preventive measures should not be postponed if delayed action may\nresult in irreversible consequences. We argue that the precautionary principle\nis a general principle of international law and, therefore, that there is a\npositive obligation on states under the right to life within international\nhuman rights law to proactively take regulatory action to mitigate the\npotential existential risk of AI. This is significant because, if an\ninternational obligation to regulate the development of AI can be established\nunder international law, then the basic legal framework would be in place to\naddress this evolving threat.\n", "date": "2025-03-23", "categories": []}
{"id": "2503.18994", "title": "HH4AI: A methodological Framework for AI Human Rights impact assessment\n  under the EUAI ACT", "abstract": "  This paper introduces the HH4AI Methodology, a structured approach to\nassessing the impact of AI systems on human rights, focusing on compliance with\nthe EU AI Act and addressing technical, ethical, and regulatory challenges. The\npaper highlights AIs transformative nature, driven by autonomy, data, and\ngoal-oriented design, and how the EU AI Act promotes transparency,\naccountability, and safety. A key challenge is defining and assessing\n\"high-risk\" AI systems across industries, complicated by the lack of\nuniversally accepted standards and AIs rapid evolution.\n  To address these challenges, the paper explores the relevance of ISO/IEC and\nIEEE standards, focusing on risk management, data quality, bias mitigation, and\ngovernance. It proposes a Fundamental Rights Impact Assessment (FRIA)\nmethodology, a gate-based framework designed to isolate and assess risks\nthrough phases including an AI system overview, a human rights checklist, an\nimpact assessment, and a final output phase. A filtering mechanism tailors the\nassessment to the system's characteristics, targeting areas like\naccountability, AI literacy, data governance, and transparency.\n  The paper illustrates the FRIA methodology through a fictional case study of\nan automated healthcare triage service. The structured approach enables\nsystematic filtering, comprehensive risk assessment, and mitigation planning,\neffectively prioritizing critical risks and providing clear remediation\nstrategies. This promotes better alignment with human rights principles and\nenhances regulatory compliance.\n", "date": "2025-03-23", "categories": []}
{"id": "2503.17388", "title": "AI Companies Should Report Pre- and Post-Mitigation Safety Evaluations", "abstract": "  The rapid advancement of AI systems has raised widespread concerns about\npotential harms of frontier AI systems and the need for responsible evaluation\nand oversight. In this position paper, we argue that frontier AI companies\nshould report both pre- and post-mitigation safety evaluations to enable\ninformed policy decisions. Evaluating models at both stages provides\npolicymakers with essential evidence to regulate deployment, access, and safety\nstandards. We show that relying on either in isolation can create a misleading\npicture of model safety. Our analysis of AI safety disclosures from leading\nfrontier labs identifies three critical gaps: (1) companies rarely evaluate\nboth pre- and post-mitigation versions, (2) evaluation methods lack\nstandardization, and (3) reported results are often too vague to inform policy.\nTo address these issues, we recommend mandatory disclosure of pre- and\npost-mitigation capabilities to approved government bodies, standardized\nevaluation methods, and minimum transparency requirements for public safety\nreporting. These ensure that policymakers and regulators can craft targeted\nsafety measures, assess deployment risks, and scrutinize companies' safety\nclaims effectively.\n", "date": "2025-03-17", "categories": []}
{"id": "2503.18387", "title": "Manipulation and the AI Act: Large Language Model Chatbots and the\n  Danger of Mirrors", "abstract": "  Large Language Model chatbots are increasingly taking the form and visage of\nhuman beings, adapting human faces, names, voices, personalities, and quirks,\nincluding those of celebrities and well-known political figures. Personifying\nAI chatbots could foreseeably increase their trust with users. However, it\ncould also make them more capable of manipulation, by creating the illusion of\na close and intimate relationship with an artificial entity. The European\nCommission has finalized the AI Act, with the EU Parliament making amendments\nbanning manipulative and deceptive AI systems that cause significant harm to\nusers. Although the AI Act covers harms that accumulate over time, it is\nunlikely to prevent harms associated with prolonged discussions with AI\nchatbots. Specifically, a chatbot could reinforce a person's negative emotional\nstate over weeks, months, or years through negative feedback loops, prolonged\nconversations, or harmful recommendations, contributing to a user's\ndeteriorating mental health.\n", "date": "2025-03-24", "categories": []}
{"id": "2503.18814", "title": "Towards Responsible AI Music: an Investigation of Trustworthy Features\n  for Creative Systems", "abstract": "  Generative AI is radically changing the creative arts, by fundamentally\ntransforming the way we create and interact with cultural artefacts. While\noffering unprecedented opportunities for artistic expression and\ncommercialisation, this technology also raises ethical, societal, and legal\nconcerns. Key among these are the potential displacement of human creativity,\ncopyright infringement stemming from vast training datasets, and the lack of\ntransparency, explainability, and fairness mechanisms. As generative systems\nbecome pervasive in this domain, responsible design is crucial. Whilst previous\nwork has tackled isolated aspects of generative systems (e.g., transparency,\nevaluation, data), we take a comprehensive approach, grounding these efforts\nwithin the Ethics Guidelines for Trustworthy Artificial Intelligence produced\nby the High-Level Expert Group on AI appointed by the European Commission - a\nframework for designing responsible AI systems across seven macro requirements.\nFocusing on generative music AI, we illustrate how these requirements can be\ncontextualised for the field, addressing trustworthiness across multiple\ndimensions and integrating insights from the existing literature. We further\npropose a roadmap for operationalising these contextualised requirements,\nemphasising interdisciplinary collaboration and stakeholder engagement. Our\nwork provides a foundation for designing and evaluating responsible music\ngeneration systems, calling for collaboration among AI experts, ethicists,\nlegal scholars, and artists. This manuscript is accompanied by a website:\nhttps://amresearchlab.github.io/raim-framework/.\n", "date": "2025-03-24", "categories": []}
{"id": "2503.14522", "title": "Accessibility Considerations in the Development of an AI Action Plan", "abstract": "  We argue that there is a need for Accessibility to be represented in several\nimportant domains:\n  - Capitalize on the new capabilities AI provides - Support for open source\ndevelopment of AI, which can allow disabled and disability focused\nprofessionals to contribute, including\n  - Development of Accessibility Apps which help realise the promise of AI in\naccessibility domains\n  - Open Source Model Development and Validation to ensure that accessibility\nconcerns are addressed in these algorithms\n  - Data Augmentation to include accessibility in data sets used to train\nmodels\n  - Accessible Interfaces that allow disabled people to use any AI app, and to\nvalidate its outputs\n  - Dedicated Functionality and Libraries that can make it easy to integrate AI\nsupport into a variety of settings and apps. - Data security and privacy and\nprivacy risks including data collected by AI based accessibility technologies;\nand the possibility of disability disclosure. - Disability-specific AI risks\nand biases including both direct bias (during AI use by the disabled person)\nand indirect bias (when AI is used by someone else on data relating to a\ndisabled person).\n", "date": "2025-03-14", "categories": []}
{"id": "2406.18842", "title": "The global landscape of academic guidelines for generative AI and Large\n  Language Models", "abstract": "  The integration of Generative Artificial Intelligence (GAI) and Large\nLanguage Models (LLMs) in academia has spurred a global discourse on their\npotential pedagogical benefits and ethical considerations. Positive reactions\nhighlight some potential, such as collaborative creativity, increased access to\neducation, and empowerment of trainers and trainees. However, negative\nreactions raise concerns about ethical complexities, balancing innovation and\nacademic integrity, unequal access, and misinformation risks. Through a\nsystematic survey and text-mining-based analysis of global and national\ndirectives, insights from independent research, and eighty university-level\nguidelines, this study provides a nuanced understanding of the opportunities\nand challenges posed by GAI and LLMs in education. It emphasizes the importance\nof balanced approaches that harness the benefits of these technologies while\naddressing ethical considerations and ensuring equitable access and educational\noutcomes. The paper concludes with recommendations for fostering responsible\ninnovation and ethical practices to guide the integration of GAI and LLMs in\nacademia.\n", "date": "2024-05-26", "categories": []}
{"id": "2501.13320", "title": "Toward Ethical AI: A Qualitative Analysis of Stakeholder Perspectives", "abstract": "  As Artificial Intelligence (AI) systems become increasingly integrated into\nvarious aspects of daily life, concerns about privacy and ethical\naccountability are gaining prominence. This study explores stakeholder\nperspectives on privacy in AI systems, focusing on educators, parents, and AI\nprofessionals. Using qualitative analysis of survey responses from 227\nparticipants, the research identifies key privacy risks, including data\nbreaches, ethical misuse, and excessive data collection, alongside perceived\nbenefits such as personalized services, enhanced efficiency, and educational\nadvancements. Stakeholders emphasized the need for transparency,\nprivacy-by-design, user empowerment, and ethical oversight to address privacy\nconcerns effectively. The findings provide actionable insights into balancing\nthe benefits of AI with robust privacy protections, catering to the diverse\nneeds of stakeholders. Recommendations include implementing selective data use,\nfostering transparency, promoting user autonomy, and integrating ethical\nprinciples into AI development. This study contributes to the ongoing discourse\non ethical AI, offering guidance for designing privacy-centric systems that\nalign with societal values and build trust among users. By addressing privacy\nchallenges, this research underscores the importance of developing AI\ntechnologies that are not only innovative but also ethically sound and\nresponsive to the concerns of all stakeholders.\n", "date": "2025-01-23", "categories": []}
{"id": "2501.13321", "title": "Investigation of the Privacy Concerns in AI Systems for Young Digital\n  Citizens: A Comparative Stakeholder Analysis", "abstract": "  The integration of Artificial Intelligence (AI) systems into technologies\nused by young digital citizens raises significant privacy concerns. This study\ninvestigates these concerns through a comparative analysis of stakeholder\nperspectives. A total of 252 participants were surveyed, with the analysis\nfocusing on 110 valid responses from parents/educators and 100 from AI\nprofessionals after data cleaning. Quantitative methods, including descriptive\nstatistics and Partial Least Squares Structural Equation Modeling, examined\nfive validated constructs: Data Ownership and Control, Parental Data Sharing,\nPerceived Risks and Benefits, Transparency and Trust, and Education and\nAwareness. Results showed Education and Awareness significantly influenced data\nownership and risk assessment, while Data Ownership and Control strongly\nimpacted Transparency and Trust. Transparency and Trust, along with Perceived\nRisks and Benefits, showed minimal influence on Parental Data Sharing,\nsuggesting other factors may play a larger role. The study underscores the need\nfor user-centric privacy controls, tailored transparency strategies, and\ntargeted educational initiatives. Incorporating diverse stakeholder\nperspectives offers actionable insights into ethical AI design and governance,\nbalancing innovation with robust privacy protections to foster trust in a\ndigital age.\n", "date": "2025-01-23", "categories": []}
{"id": "2503.10649", "title": "Measuring Political Preferences in AI Systems: An Integrative Approach", "abstract": "  Political biases in Large Language Model (LLM)-based artificial intelligence\n(AI) systems, such as OpenAI's ChatGPT or Google's Gemini, have been previously\nreported. While several prior studies have attempted to quantify these biases\nusing political orientation tests, such approaches are limited by potential\ntests' calibration biases and constrained response formats that do not reflect\nreal-world human-AI interactions. This study employs a multi-method approach to\nassess political bias in leading AI systems, integrating four complementary\nmethodologies: (1) linguistic comparison of AI-generated text with the language\nused by Republican and Democratic U.S. Congress members, (2) analysis of\npolitical viewpoints embedded in AI-generated policy recommendations, (3)\nsentiment analysis of AI-generated text toward politically affiliated public\nfigures, and (4) standardized political orientation testing. Results indicate a\nconsistent left-leaning bias across most contemporary AI systems, with arguably\nvarying degrees of intensity. However, this bias is not an inherent feature of\nLLMs; prior research demonstrates that fine-tuning with politically skewed data\ncan realign these models across the ideological spectrum. The presence of\nsystematic political bias in AI systems poses risks, including reduced\nviewpoint diversity, increased societal polarization, and the potential for\npublic mistrust in AI technologies. To mitigate these risks, AI systems should\nbe designed to prioritize factual accuracy while maintaining neutrality on most\nlawful normative issues. Furthermore, independent monitoring platforms are\nnecessary to ensure transparency, accountability, and responsible AI\ndevelopment.\n", "date": "2025-03-04", "categories": []}
{"id": "2503.09858", "title": "Media and responsible AI governance: a game-theoretic and LLM analysis", "abstract": "  This paper investigates the complex interplay between AI developers,\nregulators, users, and the media in fostering trustworthy AI systems. Using\nevolutionary game theory and large language models (LLMs), we model the\nstrategic interactions among these actors under different regulatory regimes.\nThe research explores two key mechanisms for achieving responsible governance,\nsafe AI development and adoption of safe AI: incentivising effective regulation\nthrough media reporting, and conditioning user trust on commentariats'\nrecommendation. The findings highlight the crucial role of the media in\nproviding information to users, potentially acting as a form of \"soft\"\nregulation by investigating developers or regulators, as a substitute to\ninstitutional AI regulation (which is still absent in many regions). Both\ngame-theoretic analysis and LLM-based simulations reveal conditions under which\neffective regulation and trustworthy AI development emerge, emphasising the\nimportance of considering the influence of different regulatory regimes from an\nevolutionary game-theoretic perspective. The study concludes that effective\ngovernance requires managing incentives and costs for high quality\ncommentaries.\n", "date": "2025-03-12", "categories": []}
{"id": "2503.10242", "title": "MinorBench: A hand-built benchmark for content-based risks for children", "abstract": "  Large Language Models (LLMs) are rapidly entering children's lives - through\nparent-driven adoption, schools, and peer networks - yet current AI ethics and\nsafety research do not adequately address content-related risks specific to\nminors. In this paper, we highlight these gaps with a real-world case study of\nan LLM-based chatbot deployed in a middle school setting, revealing how\nstudents used and sometimes misused the system. Building on these findings, we\npropose a new taxonomy of content-based risks for minors and introduce\nMinorBench, an open-source benchmark designed to evaluate LLMs on their ability\nto refuse unsafe or inappropriate queries from children. We evaluate six\nprominent LLMs under different system prompts, demonstrating substantial\nvariability in their child-safety compliance. Our results inform practical\nsteps for more robust, child-focused safety mechanisms and underscore the\nurgency of tailoring AI systems to safeguard young users.\n", "date": "2025-03-13", "categories": []}
{"id": "2503.06788", "title": "Dubito Ergo Sum: Exploring AI Ethics", "abstract": "  We paraphrase Descartes' famous dictum in the area of AI ethics where the \"I\ndoubt and therefore I am\" is suggested as a necessary aspect of morality.\nTherefore AI, which cannot doubt itself, cannot possess moral agency. Of\ncourse, this is not the end of the story. We explore various aspects of the\nhuman mind that substantially differ from AI, which includes the sensory\ngrounding of our knowing, the act of understanding, and the significance of\nbeing able to doubt ourselves. The foundation of our argument is the discipline\nof ethics, one of the oldest and largest knowledge projects of human history,\nyet, we seem only to be beginning to get a grasp of it. After a couple of\nthousand years of studying the ethics of humans, we (humans) arrived at a point\nwhere moral psychology suggests that our moral decisions are intuitive, and all\nthe models from ethics become relevant only when we explain ourselves. This\nrecognition has a major impact on what and how we can do regarding AI ethics.\nWe do not offer a solution, we explore some ideas and leave the problem open,\nbut we hope somewhat better understood than before our study.\n", "date": "2025-03-09", "categories": []}
{"id": "2503.07685", "title": "Ways of Seeing, and Selling, AI Art", "abstract": "  In early 2025, Augmented Intelligence - Christie's first AI art auction -\ndrew criticism for showcasing a controversial genre. Amid wider legal\nuncertainty, artists voiced concerns over data mining practices, notably with\nrespect to copyright. The backlash could be viewed as a microcosm of AI's\ncontested position in the creative economy. Touching on the auction's\npresentation, reception, and results, this paper explores how, among social\ndissonance, machine learning finds its place in the artworld. Foregrounding\nresponsible innovation, the paper provides a balanced perspective that\nchampions creators' rights and brings nuance to this polarised debate. With a\nfocus on exhibition design, it centres framing, which refers to the way a piece\nis presented to influence consumer perception. Context plays a central role in\nshaping our understanding of how good, valuable, and even ethical an artwork\nis. In this regard, Augmented Intelligence situates AI art within a\nsurprisingly traditional framework, leveraging hallmarks of \"high art\" to\nestablish the genre's cultural credibility. Generative AI has a clear economic\ndimension, converging questions of artistic merit with those of monetary worth.\nScholarship on ways of seeing, or framing, could substantively inform the\ninterpretation and evaluation of creative outputs, including assessments of\ntheir aesthetic and commercial value.\n", "date": "2025-03-10", "categories": []}
{"id": "2407.08867", "title": "Perceptions of Sentient AI and Other Digital Minds: Evidence from the\n  AI, Morality, and Sentience (AIMS) Survey", "abstract": "  Humans now interact with a variety of digital minds, AI systems that appear\nto have mental faculties such as reasoning, emotion, and agency, and public\nfigures are discussing the possibility of sentient AI. We present initial\nresults from 2021 and 2023 for the nationally representative AI, Morality, and\nSentience (AIMS) survey (N = 3,500). Mind perception and moral concern for AI\nwelfare were surprisingly high and significantly increased: in 2023, one in\nfive U.S. adults believed some AI systems are currently sentient, and 38%\nsupported legal rights for sentient AI. People became more opposed to building\ndigital minds: in 2023, 63% supported banning smarter-than-human AI, and 69%\nsupported banning sentient AI. The median 2023 forecast was that sentient AI\nwould arrive in just five years. The development of safe and beneficial AI\nrequires not just technical study but understanding the complex ways in which\nhumans perceive and coexist with digital minds.\n", "date": "2024-07-11", "categories": []}
{"id": "2503.05705", "title": "Inference Scaling Reshapes AI Governance", "abstract": "  The shift from scaling up the pre-training compute of AI systems to scaling\nup their inference compute may have profound effects on AI governance. The\nnature of these effects depends crucially on whether this new inference compute\nwill primarily be used during external deployment or as part of a more complex\ntraining programme within the lab. Rapid scaling of inference-at-deployment\nwould: lower the importance of open-weight models (and of securing the weights\nof closed models), reduce the impact of the first human-level models, change\nthe business model for frontier AI, reduce the need for power-intense data\ncentres, and derail the current paradigm of AI governance via training compute\nthresholds. Rapid scaling of inference-during-training would have more\nambiguous effects that range from a revitalisation of pre-training scaling to a\nform of recursive self-improvement via iterated distillation and amplification.\n", "date": "2025-02-12", "categories": []}
{"id": "2503.05737", "title": "Local Differences, Global Lessons: Insights from Organisation Policies\n  for International Legislation", "abstract": "  The rapid adoption of AI across diverse domains has led to the development of\norganisational guidelines that vary significantly, even within the same sector.\nThis paper examines AI policies in two domains, news organisations and\nuniversities, to understand how bottom-up governance approaches shape AI usage\nand oversight. By analysing these policies, we identify key areas of\nconvergence and divergence in how organisations address risks such as bias,\nprivacy, misinformation, and accountability. We then explore the implications\nof these findings for international AI legislation, particularly the EU AI Act,\nhighlighting gaps where practical policy insights could inform regulatory\nrefinements. Our analysis reveals that organisational policies often address\nissues such as AI literacy, disclosure practices, and environmental impact,\nareas that are underdeveloped in existing international frameworks. We argue\nthat lessons from domain-specific AI policies can contribute to more adaptive\nand effective AI governance at the global level. This study provides actionable\nrecommendations for policymakers seeking to bridge the gap between local AI\npractices and international regulations.\n", "date": "2025-02-19", "categories": []}
{"id": "2503.05747", "title": "Balancing Innovation and Integrity: AI Integration in Liberal Arts\n  College Administration", "abstract": "  This paper explores the intersection of artificial intelligence and higher\neducation administration, focusing on liberal arts colleges (LACs). It examines\nAI's opportunities and challenges in academic and student affairs, legal\ncompliance, and accreditation processes, while also addressing the ethical\nconsiderations of AI deployment in mission-driven institutions. Considering\nAI's value pluralism and potential allocative or representational harms caused\nby algorithmic bias, LACs must ensure AI aligns with its mission and\nprinciples. The study highlights other strategies for responsible AI\nintegration, balancing innovation with institutional values.\n", "date": "2025-02-20", "categories": []}
{"id": "2503.05769", "title": "Effect of Gender Fair Job Description on Generative AI Images", "abstract": "  STEM fields are traditionally male-dominated, with gender biases shaping\nperceptions of job accessibility. This study analyzed gender representation in\nSTEM occupation images generated by OpenAI DALL-E 3 \\& Black Forest FLUX.1\nusing 150 prompts in three linguistic forms: German generic masculine, German\npair form, and English. As control, 20 pictures of social occupations were\ngenerated as well. Results revealed significant male bias across all forms,\nwith the German pair form showing reduced bias but still overrepresenting men\nfor the STEM-Group and mixed results for the Group of Social Occupations. These\nfindings highlight generative AI's role in reinforcing societal biases,\nemphasizing the need for further discussion on diversity (in AI). Further\naspects analyzed are age-distribution and ethnic diversity.\n", "date": "2025-02-25", "categories": []}
{"id": "2503.05773", "title": "Between Innovation and Oversight: A Cross-Regional Study of AI Risk\n  Management Frameworks in the EU, U.S., UK, and China", "abstract": "  As artificial intelligence (AI) technologies increasingly enter important\nsectors like healthcare, transportation, and finance, the development of\neffective governance frameworks is crucial for dealing with ethical, security,\nand societal risks. This paper conducts a comparative analysis of AI risk\nmanagement strategies across the European Union (EU), United States (U.S.),\nUnited Kingdom (UK), and China. A multi-method qualitative approach, including\ncomparative policy analysis, thematic analysis, and case studies, investigates\nhow these regions classify AI risks, implement compliance measures, structure\noversight, prioritize transparency, and respond to emerging innovations.\nExamples from high-risk contexts like healthcare diagnostics, autonomous\nvehicles, fintech, and facial recognition demonstrate the advantages and\nlimitations of different regulatory models. The findings show that the EU\nimplements a structured, risk-based framework that prioritizes transparency and\nconformity assessments, while the U.S. uses decentralized, sector-specific\nregulations that promote innovation but may lead to fragmented enforcement. The\nflexible, sector-specific strategy of the UK facilitates agile responses but\nmay lead to inconsistent coverage across domains. China's centralized\ndirectives allow rapid large-scale implementation while constraining public\ntransparency and external oversight. These insights show the necessity for AI\nregulation that is globally informed yet context-sensitive, aiming to balance\neffective risk management with technological progress. The paper concludes with\npolicy recommendations and suggestions for future research aimed at enhancing\neffective, adaptive, and inclusive AI governance globally.\n", "date": "2025-02-25", "categories": []}
{"id": "2503.05812", "title": "Intolerable Risk Threshold Recommendations for Artificial Intelligence", "abstract": "  Frontier AI models -- highly capable foundation models at the cutting edge of\nAI development -- may pose severe risks to public safety, human rights,\neconomic stability, and societal value in the coming years. These risks could\narise from deliberate adversarial misuse, system failures, unintended cascading\neffects, or simultaneous failures across multiple models.\n  In response to such risks, at the AI Seoul Summit in May 2024, 16 global AI\nindustry organizations signed the Frontier AI Safety Commitments, and 27\nnations and the EU issued a declaration on their intent to define these\nthresholds. To fulfill these commitments, organizations must determine and\ndisclose ``thresholds at which severe risks posed by a model or system, unless\nadequately mitigated, would be deemed intolerable.''\n  To assist in setting and operationalizing intolerable risk thresholds, we\noutline key principles and considerations; for example, to aim for ``good, not\nperfect'' thresholds in the face of limited data on rapidly advancing AI\ncapabilities and consequently evolving risks. We also propose specific\nthreshold recommendations, including some detailed case studies, for a subset\nof risks across eight risk categories: (1) Chemical, Biological, Radiological,\nand Nuclear (CBRN) Weapons, (2) Cyber Attacks, (3) Model Autonomy, (4)\nPersuasion and Manipulation, (5) Deception, (6) Toxicity, (7) Discrimination,\nand (8) Socioeconomic Disruption. Our goal is to serve as a starting point or\nsupplementary resource for policymakers and industry leaders, encouraging\nproactive risk management that prioritizes preventing intolerable risks (ex\nante) rather than merely mitigating them after they occur (ex post).\n", "date": "2025-03-04", "categories": []}
{"id": "2503.06353", "title": "The AI Pentad, the CHARME$^{2}$D Model, and an Assessment of\n  Current-State AI Regulation", "abstract": "  Artificial Intelligence (AI) has made remarkable progress in the past few\nyears with AI-enabled applications beginning to permeate every aspect of our\nsociety. Despite the widespread consensus on the need to regulate AI, there\nremains a lack of a unified approach to framing, developing, and assessing AI\nregulations. Many of the existing methods take a value-based approach, for\nexample, accountability, fairness, free from bias, transparency, and trust.\nHowever, these methods often face challenges at the outset due to disagreements\nin academia over the subjective nature of these definitions. This paper aims to\nestablish a unifying model for AI regulation from the perspective of core AI\ncomponents. We first introduce the AI Pentad, which comprises the five\nessential components of AI: humans and organizations, algorithms, data,\ncomputing, and energy. We then review AI regulatory enablers, including AI\nregistration and disclosure, AI monitoring, and AI enforcement mechanisms.\nSubsequently, we present the CHARME$^{2}$D Model to explore further the\nrelationship between the AI Pentad and AI regulatory enablers. Finally, we\napply the CHARME$^{2}$D model to assess AI regulatory efforts in the European\nUnion (EU), China, the United Arab Emirates (UAE), the United Kingdom (UK), and\nthe United States (US), highlighting their strengths, weaknesses, and gaps.\nThis comparative evaluation offers insights for future legislative work in the\nAI domain.\n", "date": "2025-03-08", "categories": []}
{"id": "2503.06411", "title": "Decoding the Black Box: Integrating Moral Imagination with Technical AI\n  Governance", "abstract": "  This paper examines the intricate interplay among AI safety, security, and\ngovernance by integrating technical systems engineering with principles of\nmoral imagination and ethical philosophy. Drawing on foundational insights from\nWeapons of Math Destruction and Thinking in Systems alongside contemporary\ndebates in AI ethics, we develop a comprehensive multi-dimensional framework\ndesigned to regulate AI technologies deployed in high-stakes domains such as\ndefense, finance, healthcare, and education. Our approach combines rigorous\ntechnical analysis, quantitative risk assessment, and normative evaluation to\nexpose systemic vulnerabilities inherent in opaque, black-box models. Detailed\ncase studies, including analyses of Microsoft Tay (2016) and the UK A-Level\nGrading Algorithm (2020), demonstrate how security lapses, bias amplification,\nand lack of accountability can precipitate cascading failures that undermine\npublic trust. We conclude by outlining targeted strategies for enhancing AI\nresilience through adaptive regulatory mechanisms, robust security protocols,\nand interdisciplinary oversight, thereby advancing the state of the art in\nethical and technical AI governance.\n", "date": "2025-03-09", "categories": []}
{"id": "2503.07326", "title": "AI Biases as Asymmetries: A Review to Guide Practice", "abstract": "  The understanding of bias in AI is currently undergoing a revolution.\nInitially understood as errors or flaws, biases are increasingly recognized as\nintegral to AI systems and sometimes preferable to less biased alternatives. In\nthis paper, we review the reasons for this changed understanding and provide\nnew guidance on two questions: First, how should we think about and measure\nbiases in AI systems, consistent with the new understanding? Second, what kinds\nof bias in an AI system should we accept or even amplify, and what kinds should\nwe minimize or eliminate, and why? The key to answering both questions, we\nargue, is to understand biases as \"violations of a symmetry standard\"\n(following Kelly). We distinguish three main types of asymmetry in AI\nsystems-error biases, inequality biases, and process biases-and highlight\nplaces in the pipeline of AI development and application where bias of each\ntype is likely to be good, bad, or inevitable.\n", "date": "2025-03-10", "categories": []}
{"id": "2407.10329", "title": "Generative Discrimination: What Happens When Generative AI Exhibits\n  Bias, and What Can Be Done About It", "abstract": "  As generative Artificial Intelligence (genAI) technologies proliferate across\nsectors, they offer significant benefits but also risk exacerbating\ndiscrimination. This chapter explores how genAI intersects with\nnon-discrimination laws, identifying shortcomings and suggesting improvements.\nIt highlights two main types of discriminatory outputs: (i) demeaning and\nabusive content and (ii) subtler biases due to inadequate representation of\nprotected groups, which may not be overtly discriminatory in individual cases\nbut have cumulative discriminatory effects. For example, genAI systems may\npredominantly depict white men when asked for images of people in important\njobs.\n  This chapter examines these issues, categorizing problematic outputs into\nthree legal categories: discriminatory content; harassment; and legally hard\ncases like unbalanced content, harmful stereotypes or misclassification. It\nargues for holding genAI providers and deployers liable for discriminatory\noutputs and highlights the inadequacy of traditional legal frameworks to\naddress genAI-specific issues. The chapter suggests updating EU laws, including\nthe AI Act, to mitigate biases in training and input data, mandating testing\nand auditing, and evolving legislation to enforce standards for bias mitigation\nand inclusivity as technology advances.\n", "date": "2024-06-26", "categories": []}
{"id": "2503.04733", "title": "Ethics of generative AI and manipulation: a design-oriented research\n  agenda", "abstract": "  Generative AI enables automated, effective manipulation at scale. Despite the\ngrowing general ethical discussion around generative AI, the specific\nmanipulation risks remain inadequately investigated. This article outlines\nessential inquiries encompassing conceptual, empirical, and design dimensions\nof manipulation, pivotal for comprehending and curbing manipulation risks. By\nhighlighting these questions, the article underscores the necessity of an\nappropriate conceptualisation of manipulation to ensure the responsible\ndevelopment of Generative AI technologies.\n", "date": "2025-02-01", "categories": []}
{"id": "2503.04740", "title": "PRISM: Perspective Reasoning for Integrated Synthesis and Mediation as a\n  Multi-Perspective Framework for AI Alignment", "abstract": "  In this work, we propose Perspective Reasoning for Integrated Synthesis and\nMediation (PRISM), a multiple-perspective framework for addressing persistent\nchallenges in AI alignment such as conflicting human values and specification\ngaming. Grounded in cognitive science and moral psychology, PRISM organizes\nmoral concerns into seven \"basis worldviews\", each hypothesized to capture a\ndistinct dimension of human moral cognition, ranging from survival-focused\nreflexes through higher-order integrative perspectives. It then applies a\nPareto-inspired optimization scheme to reconcile competing priorities without\nreducing them to a single metric. Under the assumption of reliable context\nvalidation for robust use, the framework follows a structured workflow that\nelicits viewpoint-specific responses, synthesizes them into a balanced outcome,\nand mediates remaining conflicts in a transparent and iterative manner. By\nreferencing layered approaches to moral cognition from cognitive science, moral\npsychology, and neuroscience, PRISM clarifies how different moral drives\ninteract and systematically documents and mediates ethical tradeoffs. We\nillustrate its efficacy through real outputs produced by a working prototype,\napplying PRISM to classic alignment problems in domains such as public health\npolicy, workplace automation, and education. By anchoring AI deliberation in\nthese human vantage points, PRISM aims to bound interpretive leaps that might\notherwise drift into non-human or machine-centric territory. We briefly outline\nfuture directions, including real-world deployments and formal verifications,\nwhile maintaining the core focus on multi-perspective synthesis and conflict\nmediation.\n", "date": "2025-02-05", "categories": []}
{"id": "2503.04751", "title": "What is Ethical: AIHED Driving Humans or Human-Driven AIHED? A\n  Conceptual Framework enabling the Ethos of AI-driven Higher education", "abstract": "  The rapid integration of Artificial Intelligence (AI) in Higher Education\n(HE) is transforming personalized learning, administrative automation, and\ndecision-making. However, this progress presents a duality, as AI adoption also\nintroduces ethical and institutional challenges, including algorithmic bias,\ndata privacy risks, and governance inconsistencies. To address these concerns,\nthis study introduces the Human-Driven AI in Higher Education (HD-AIHED)\nFramework, ensuring compliance with UNESCO and OECD ethical standards. This\nconceptual research employs a qualitative meta-synthesis approach, integrating\nqualitative and quantitative studies to identify patterns, contradictions, and\ngaps in AI adoption within HE. It reinterprets existing datasets through\ntheoretical and ethical lenses to develop governance frameworks. The study\napplies a participatory integrated co-system, Phased Human Intelligence, SWOC\nanalysis, and AI ethical review boards to assess AI readiness and governance\nstrategies for universities and HE institutions. The HD-AIHED model bridges AI\nresearch gaps, addresses global real-time challenges, and provides tailored,\nscalable, and ethical strategies for diverse educational contexts. By\nemphasizing interdisciplinary collaboration among stakeholders, this study\nenvisions AIHED as a transparent and equitable force for innovation. The\nHD-AIHED framework ensures AI acts as a collaborative and ethical enabler\nrather than a disruptive replacement for human intelligence while advocating\nfor responsible AI implementation in HE.\n", "date": "2025-02-07", "categories": []}
{"id": "2503.04866", "title": "Privacy in Responsible AI: Approaches to Facial Recognition from Cloud\n  Providers", "abstract": "  As the use of facial recognition technology is expanding in different\ndomains, ensuring its responsible use is gaining more importance. This paper\nconducts a comprehensive literature review of existing studies on facial\nrecognition technology from the perspective of privacy, which is one of the key\nResponsible AI principles.\n  Cloud providers, such as Microsoft, AWS, and Google, are at the forefront of\ndelivering facial-related technology services, but their approaches to\nresponsible use of these technologies vary significantly. This paper compares\nhow these cloud giants implement the privacy principle into their facial\nrecognition and detection services. By analysing their approaches, it\nidentifies both common practices and notable differences. The results of this\nresearch will be valuable for developers and businesses by providing them\ninsights into best practices of three major companies for integration\nresponsible AI, particularly privacy, into their cloud-based facial recognition\ntechnologies.\n", "date": "2025-03-06", "categories": []}
{"id": "2406.05364", "title": "Is On-Device AI Broken and Exploitable? Assessing the Trust and Ethics\n  in Small Language Models", "abstract": "  In this paper, we present a very first study to investigate trust and ethical\nimplications of on-device artificial intelligence (AI), focusing on small\nlanguage models (SLMs) amenable for personal devices like smartphones. While\non-device SLMs promise enhanced privacy, reduced latency, and improved user\nexperience compared to cloud-based services, we posit that they might also\nintroduce significant risks and vulnerabilities compared to their on-server\ncounterparts. As part of our trust assessment study, we conduct a systematic\nevaluation of the state-of-the-art on-devices SLMs, contrasted to their\non-server counterparts, based on a well-established trustworthiness measurement\nframework. Our results show on-device SLMs to be significantly less\ntrustworthy, specifically demonstrating more stereotypical, unfair and\nprivacy-breaching behavior. Informed by these findings, we then perform our\nethics assessment study using a dataset of unethical questions, that depicts\nharmful scenarios. Our results illustrate the lacking ethical safeguards in\non-device SLMs, emphasizing their capabilities of generating harmful content.\nFurther, the broken safeguards and exploitable nature of on-device SLMs is\ndemonstrated using potentially unethical vanilla prompts, to which the\non-device SLMs answer with valid responses without any filters and without the\nneed for any jailbreaking or prompt engineering. These responses can be abused\nfor various harmful and unethical scenarios like: societal harm, illegal\nactivities, hate, self-harm, exploitable phishing content and many others, all\nof which indicates the severe vulnerability and exploitability of these\non-device SLMs.\n", "date": "2024-06-08", "categories": []}
{"id": "2407.11613", "title": "Bringing AI Participation Down to Scale: A Comment on Open AIs\n  Democratic Inputs to AI Project", "abstract": "  In 2023, Open AIs Democratic Inputs program funded 10 teams to design\nprocedures for public participation in generative AI. In this Perspective, we\nreview the results of the project, drawing on interviews with some of the teams\nand our own experiences conducting participation exercises, we identify several\nshared yet largely unspoken assumptions of the Democratic Inputs program 1.\nthat participation must be scalable 2. that the object of participation is a\nsingle model 3. that there must be a single form of participation 4. that the\ngoal is to extract abstract principles 5. that these principles should have\nconsensus 6. that publics should be representative and encourage alternative\nforms of participation in AI, perhaps not undertaken by tech companies.\n", "date": "2024-07-16", "categories": []}
{"id": "2410.19803", "title": "First-Person Fairness in Chatbots", "abstract": "  Evaluating chatbot fairness is crucial given their rapid proliferation, yet\ntypical chatbot tasks (e.g., resume writing, entertainment) diverge from the\ninstitutional decision-making tasks (e.g., resume screening) which have\ntraditionally been central to discussion of algorithmic fairness. The\nopen-ended nature and diverse use-cases of chatbots necessitate novel methods\nfor bias assessment. This paper addresses these challenges by introducing a\nscalable counterfactual approach to evaluate \"first-person fairness,\" meaning\nfairness toward chatbot users based on demographic characteristics. Our method\nemploys a Language Model as a Research Assistant (LMRA) to yield quantitative\nmeasures of harmful stereotypes and qualitative analyses of demographic\ndifferences in chatbot responses. We apply this approach to assess biases in\nsix of our language models across millions of interactions, covering sixty-six\ntasks in nine domains and spanning two genders and four races. Independent\nhuman annotations corroborate the LMRA-generated bias evaluations. This study\nrepresents the first large-scale fairness evaluation based on real-world chat\ndata. We highlight that post-training reinforcement learning techniques\nsignificantly mitigate these biases. This evaluation provides a practical\nmethodology for ongoing bias monitoring and mitigation.\n", "date": "2024-10-16", "categories": []}
{"id": "2503.00664", "title": "Generative Artificial Intelligence for Academic Research: Evidence from\n  Guidance Issued for Researchers by Higher Education Institutions in the\n  United States", "abstract": "  The recent development and use of generative AI (GenAI) has signaled a\nsignificant shift in research activities such as brainstorming, proposal\nwriting, dissemination, and even reviewing. This has raised questions about how\nto balance the seemingly productive uses of GenAI with ethical concerns such as\nauthorship and copyright issues, use of biased training data, lack of\ntransparency, and impact on user privacy. To address these concerns, many\nHigher Education Institutions (HEIs) have released institutional guidance for\nresearchers. To better understand the guidance that is being provided we report\nfindings from a thematic analysis of guidelines from thirty HEIs in the United\nStates that are classified as R1 or 'very high research activity.' We found\nthat guidance provided to researchers: (1) asks them to refer to external\nsources of information such as funding agencies and publishers to keep updated\nand use institutional resources for training and education; (2) asks them to\nunderstand and learn about specific GenAI attributes that shape research such\nas predictive modeling, knowledge cutoff date, data provenance, and model\nlimitations, and educate themselves about ethical concerns such as authorship,\nattribution, privacy, and intellectual property issues; and (3) includes\ninstructions on how to acknowledge sources and disclose the use of GenAI, how\nto communicate effectively about their GenAI use, and alerts researchers to\nlong term implications such as over reliance on GenAI, legal consequences, and\nrisks to their institutions from GenAI use. Overall, guidance places the onus\nof compliance on individual researchers making them accountable for any lapses,\nthereby increasing their responsibility.\n", "date": "2025-03-01", "categories": []}
{"id": "2503.00940", "title": "Can AI Model the Complexities of Human Moral Decision-Making? A\n  Qualitative Study of Kidney Allocation Decisions", "abstract": "  A growing body of work in Ethical AI attempts to capture human moral\njudgments through simple computational models. The key question we address in\nthis work is whether such simple AI models capture {the critical} nuances of\nmoral decision-making by focusing on the use case of kidney allocation. We\nconducted twenty interviews where participants explained their rationale for\ntheir judgments about who should receive a kidney. We observe participants: (a)\nvalue patients' morally-relevant attributes to different degrees; (b) use\ndiverse decision-making processes, citing heuristics to reduce decision\ncomplexity; (c) can change their opinions; (d) sometimes lack confidence in\ntheir decisions (e.g., due to incomplete information); and (e) express\nenthusiasm and concern regarding AI assisting humans in kidney allocation\ndecisions. Based on these findings, we discuss challenges of computationally\nmodeling moral judgments {as a stand-in for human input}, highlight drawbacks\nof current approaches, and suggest future directions to address these issues.\n", "date": "2025-03-02", "categories": []}
{"id": "2502.21250", "title": "Towards Developing Ethical Reasoners: Integrating Probabilistic\n  Reasoning and Decision-Making for Complex AI Systems", "abstract": "  A computational ethics framework is essential for AI and autonomous systems\noperating in complex, real-world environments. Existing approaches often lack\nthe adaptability needed to integrate ethical principles into dynamic and\nambiguous contexts, limiting their effectiveness across diverse scenarios. To\naddress these challenges, we outline the necessary ingredients for building a\nholistic, meta-level framework that combines intermediate representations,\nprobabilistic reasoning, and knowledge representation. The specifications\ntherein emphasize scalability, supporting ethical reasoning at both individual\ndecision-making levels and within the collective dynamics of multi-agent\nsystems. By integrating theoretical principles with contextual factors, it\nfacilitates structured and context-aware decision-making, ensuring alignment\nwith overarching ethical standards. We further explore proposed theorems\noutlining how ethical reasoners should operate, offering a foundation for\npractical implementation. These constructs aim to support the development of\nrobust and ethically reliable AI systems capable of navigating the complexities\nof real-world moral decision-making scenarios.\n", "date": "2025-02-28", "categories": []}
{"id": "2502.15719", "title": "Governing AI Beyond the Pretraining Frontier", "abstract": "  This year, jurisdictions worldwide, including the United States, the European\nUnion, the United Kingdom, and China, are set to enact or revise laws governing\nfrontier AI. Their efforts largely rely on the assumption that increasing model\nscale through pretraining is the path to more advanced AI capabilities. Yet\ngrowing evidence suggests that this \"pretraining paradigm\" may be hitting a\nwall and major AI companies are turning to alternative approaches, like\ninference-time \"reasoning,\" to boost capabilities instead.\n  This paradigm shift presents fundamental challenges for the frontier AI\ngovernance frameworks that target pretraining scale as a key bottleneck useful\nfor monitoring, control, and exclusion, threatening to undermine this new legal\norder as it emerges. This essay seeks to identify these challenges and point to\nnew paths forward for regulation. First, we examine the existing frontier AI\nregulatory regime and analyze some key traits and vulnerabilities. Second, we\nintroduce the concept of the \"pretraining frontier,\" the capabilities threshold\nmade possible by scaling up pretraining alone, and demonstrate how it could\nmake the regulatory field more diffuse and complex and lead to new forms of\ncompetition. Third, we lay out a regulatory approach that focuses on increasing\ntransparency and leveraging new natural technical bottlenecks to effectively\noversee changing frontier AI development while minimizing regulatory burdens\nand protecting fundamental rights. Our analysis provides concrete mechanisms\nfor governing frontier AI systems across diverse technical paradigms, offering\npolicymakers tools for addressing both current and future regulatory challenges\nin frontier AI.\n", "date": "2025-01-27", "categories": []}
{"id": "2502.14869", "title": "Envisioning Stakeholder-Action Pairs to Mitigate Negative Impacts of AI:\n  A Participatory Approach to Inform Policy Making", "abstract": "  The potential for negative impacts of AI has rapidly become more pervasive\naround the world, and this has intensified a need for responsible AI\ngovernance. While many regulatory bodies endorse risk-based approaches and a\nmultitude of risk mitigation practices are proposed by companies and academic\nscholars, these approaches are commonly expert-centered and thus lack the\ninclusion of a significant group of stakeholders. Ensuring that AI policies\nalign with democratic expectations requires methods that prioritize the voices\nand needs of those impacted. In this work we develop a participative and\nforward-looking approach to inform policy-makers and academics that grounds the\nneeds of lay stakeholders at the forefront and enriches the development of risk\nmitigation strategies. Our approach (1) maps potential mitigation and\nprevention strategies of negative AI impacts that assign responsibility to\nvarious stakeholders, (2) explores the importance and prioritization thereof in\nthe eyes of laypeople, and (3) presents these insights in policy fact sheets,\ni.e., a digestible format for informing policy processes. We emphasize that\nthis approach is not targeted towards replacing policy-makers; rather our aim\nis to present an informative method that enriches mitigation strategies and\nenables a more participatory approach to policy development.\n", "date": "2025-01-24", "categories": []}
{"id": "2406.16746", "title": "The Responsible Foundation Model Development Cheatsheet: A Review of\n  Tools & Resources", "abstract": "  Foundation model development attracts a rapidly expanding body of\ncontributors, scientists, and applications. To help shape responsible\ndevelopment practices, we introduce the Foundation Model Development\nCheatsheet: a growing collection of 250+ tools and resources spanning text,\nvision, and speech modalities. We draw on a large body of prior work to survey\nresources (e.g. software, documentation, frameworks, guides, and practical\ntools) that support informed data selection, processing, and understanding,\nprecise and limitation-aware artifact documentation, efficient model training,\nadvance awareness of the environmental impact from training, careful model\nevaluation of capabilities, risks, and claims, as well as responsible model\nrelease, licensing and deployment practices. We hope this curated collection of\nresources helps guide more responsible development. The process of curating\nthis list, enabled us to review the AI development ecosystem, revealing what\ntools are critically missing, misused, or over-used in existing practices. We\nfind that (i) tools for data sourcing, model evaluation, and monitoring are\ncritically under-serving ethical and real-world needs, (ii) evaluations for\nmodel safety, capabilities, and environmental impact all lack reproducibility\nand transparency, (iii) text and particularly English-centric analyses continue\nto dominate over multilingual and multi-modal analyses, and (iv) evaluation of\nsystems, rather than just models, is needed so that capabilities and impact are\nassessed in context.\n", "date": "2024-06-24", "categories": []}
{"id": "2502.12102", "title": "Relational Norms for Human-AI Cooperation", "abstract": "  How we should design and interact with social artificial intelligence depends\non the socio-relational role the AI is meant to emulate or occupy. In human\nsociety, relationships such as teacher-student, parent-child, neighbors,\nsiblings, or employer-employee are governed by specific norms that prescribe or\nproscribe cooperative functions including hierarchy, care, transaction, and\nmating. These norms shape our judgments of what is appropriate for each\npartner. For example, workplace norms may allow a boss to give orders to an\nemployee, but not vice versa, reflecting hierarchical and transactional\nexpectations. As AI agents and chatbots powered by large language models are\nincreasingly designed to serve roles analogous to human positions - such as\nassistant, mental health provider, tutor, or romantic partner - it is\nimperative to examine whether and how human relational norms should extend to\nhuman-AI interactions. Our analysis explores how differences between AI systems\nand humans, such as the absence of conscious experience and immunity to\nfatigue, may affect an AI's capacity to fulfill relationship-specific functions\nand adhere to corresponding norms. This analysis, which is a collaborative\neffort by philosophers, psychologists, relationship scientists, ethicists,\nlegal experts, and AI researchers, carries important implications for AI\nsystems design, user behavior, and regulation. While we accept that AI systems\ncan offer significant benefits such as increased availability and consistency\nin certain socio-relational roles, they also risk fostering unhealthy\ndependencies or unrealistic expectations that could spill over into human-human\nrelationships. We propose that understanding and thoughtfully shaping (or\nimplementing) suitable human-AI relational norms will be crucial for ensuring\nthat human-AI interactions are ethical, trustworthy, and favorable to human\nwell-being.\n", "date": "2025-02-17", "categories": []}
{"id": "2407.12690", "title": "The Dual Imperative: Innovation and Regulation in the AI Era", "abstract": "  This article addresses the societal costs associated with the lack of\nregulation in Artificial Intelligence and proposes a framework combining\ninnovation and regulation. Over fifty years of AI research, catalyzed by\ndeclining computing costs and the proliferation of data, have propelled AI into\nthe mainstream, promising significant economic benefits. Yet, this rapid\nadoption underscores risks, from bias amplification and labor disruptions to\nexistential threats posed by autonomous systems. The discourse is polarized\nbetween accelerationists, advocating for unfettered technological advancement,\nand doomers, calling for a slowdown to prevent dystopian outcomes. This piece\nadvocates for a middle path that leverages technical innovation and smart\nregulation to maximize the benefits of AI while minimizing its risks, offering\na pragmatic approach to the responsible progress of AI technology. Technical\ninvention beyond the most capable foundation models is needed to contain\ncatastrophic risks. Regulation is required to create incentives for this\nresearch while addressing current issues.\n", "date": "2024-05-23", "categories": []}
{"id": "2502.08651", "title": "Democratizing AI Governance: Balancing Expertise and Public\n  Participation", "abstract": "  The development and deployment of artificial intelligence (AI) systems, with\ntheir profound societal impacts, raise critical challenges for governance.\nHistorically, technological innovations have been governed by concentrated\nexpertise with limited public input. However, AI's pervasive influence across\ndomains such as healthcare, employment, and justice necessitates inclusive\ngovernance approaches. This article explores the tension between expert-led\noversight and democratic participation, analyzing models of participatory and\ndeliberative democracy. Using case studies from France and Brazil, we highlight\nhow inclusive frameworks can bridge the gap between technical complexity and\npublic accountability. Recommendations are provided for integrating these\napproaches into a balanced governance model tailored to the European Union,\nemphasizing transparency, diversity, and adaptive regulation to ensure that AI\ngovernance reflects societal values while maintaining technical rigor. This\nanalysis underscores the importance of hybrid frameworks that unite expertise\nand public voice in shaping the future of AI policy.\n", "date": "2025-01-16", "categories": []}
{"id": "2403.15457", "title": "The Journey to Trustworthy AI: Pursuit of Pragmatic Frameworks", "abstract": "  This paper reviews Trustworthy Artificial Intelligence (TAI) and its various\ndefinitions. Considering the principles respected in any society, TAI is often\ncharacterized by a few attributes, some of which have led to confusion in\nregulatory or engineering contexts. We argue against using terms such as\nResponsible or Ethical AI as substitutes for TAI. And to help clarify any\nconfusion, we suggest leaving them behind. Given the subjectivity and\ncomplexity inherent in TAI, developing a universal framework is deemed\ninfeasible. Instead, we advocate for approaches centered on addressing key\nattributes and properties such as fairness, bias, risk, security,\nexplainability, and reliability. We examine the ongoing regulatory landscape,\nwith a focus on initiatives in the EU, China, and the USA. We recognize that\ndifferences in AI regulations based on geopolitical and geographical reasons\npose an additional challenge for multinational companies. We identify risk as a\ncore factor in AI regulation and TAI. For example, as outlined in the EU-AI\nAct, organizations must gauge the risk level of their AI products to act\naccordingly (or risk hefty fines). We compare modalities of TAI implementation\nand how multiple cross-functional teams are engaged in the overall process.\nThus, a brute force approach for enacting TAI renders its efficiency and\nagility, moot. To address this, we introduce our framework\nSet-Formalize-Measure-Act (SFMA). Our solution highlights the importance of\ntransforming TAI-aware metrics, drivers of TAI, stakeholders, and\nbusiness/legal requirements into actual benchmarks or tests. Finally,\nover-regulation driven by panic of powerful AI models can, in fact, harm TAI\ntoo. Based on GitHub user-activity data, in 2023, AI open-source projects rose\nto top projects by contributor account. Enabling innovation in TAI hinges on\nthe independent contributions of the open-source community.\n", "date": "2024-03-19", "categories": []}
{"id": "2501.07913", "title": "Governing AI Agents", "abstract": "  The field of AI is undergoing a fundamental transition from generative models\nthat can produce synthetic content to artificial agents that can plan and\nexecute complex tasks with only limited human involvement. Companies that\npioneered the development of language models have now built AI agents that can\nindependently navigate the internet, perform a wide range of online tasks, and\nincreasingly serve as AI personal assistants and virtual coworkers. The\nopportunities presented by this new technology are tremendous, as are the\nassociated risks. Fortunately, there exist robust analytic frameworks for\nconfronting many of these challenges, namely, the economic theory of\nprincipal-agent problems and the common law doctrine of agency relationships.\nDrawing on these frameworks, this Article makes three contributions. First, it\nuses agency law and theory to identify and characterize problems arising from\nAI agents, including issues of information asymmetry, discretionary authority,\nand loyalty. Second, it illustrates the limitations of conventional solutions\nto agency problems: incentive design, monitoring, and enforcement might not be\neffective for governing AI agents that make uninterpretable decisions and\noperate at unprecedented speed and scale. Third, the Article explores the\nimplications of agency law and theory for designing and regulating AI agents,\narguing that new technical and legal infrastructure is needed to support\ngovernance principles of inclusivity, visibility, and liability.\n", "date": "2025-01-14", "categories": []}
{"id": "2410.14012", "title": "LLMs are Biased Teachers: Evaluating LLM Bias in Personalized Education", "abstract": "  With the increasing adoption of large language models (LLMs) in education,\nconcerns about inherent biases in these models have gained prominence. We\nevaluate LLMs for bias in the personalized educational setting, specifically\nfocusing on the models' roles as \"teachers.\" We reveal significant biases in\nhow models generate and select educational content tailored to different\ndemographic groups, including race, ethnicity, sex, gender, disability status,\nincome, and national origin. We introduce and apply two bias score\nmetrics--Mean Absolute Bias (MAB) and Maximum Difference Bias (MDB)--to analyze\n9 open and closed state-of-the-art LLMs. Our experiments, which utilize over\n17,000 educational explanations across multiple difficulty levels and topics,\nuncover that models potentially harm student learning by both perpetuating\nharmful stereotypes and reversing them. We find that bias is similar for all\nfrontier models, with the highest MAB along income levels while MDB is highest\nrelative to both income and disability status. For both metrics, we find the\nlowest bias exists for sex/gender and race/ethnicity.\n", "date": "2024-10-17", "categories": []}
{"id": "2305.10646", "title": "Ethical ChatGPT: Concerns, Challenges, and Commandments", "abstract": "  Large language models, e.g. ChatGPT are currently contributing enormously to\nmake artificial intelligence even more popular, especially among the general\npopulation. However, such chatbot models were developed as tools to support\nnatural language communication between humans. Problematically, it is very much\na ``statistical correlation machine\" (correlation instead of causality) and\nthere are indeed ethical concerns associated with the use of AI language models\nsuch as ChatGPT, such as Bias, Privacy, and Abuse. This paper highlights\nspecific ethical concerns on ChatGPT and articulates key challenges when\nChatGPT is used in various applications. Practical commandments for different\nstakeholders of ChatGPT are also proposed that can serve as checklist\nguidelines for those applying ChatGPT in their applications. These commandment\nexamples are expected to motivate the ethical use of ChatGPT.\n", "date": "2023-05-18", "categories": []}
{"id": "2412.04683", "title": "From Principles to Practice: A Deep Dive into AI Ethics and Regulations", "abstract": "  In the rapidly evolving domain of Artificial Intelligence (AI), the complex\ninteraction between innovation and regulation has become an emerging focus of\nour society. Despite tremendous advancements in AI's capabilities to excel in\nspecific tasks and contribute to diverse sectors, establishing a high degree of\ntrust in AI-generated outputs and decisions necessitates meticulous caution and\ncontinuous oversight. A broad spectrum of stakeholders, including governmental\nbodies, private sector corporations, academic institutions, and individuals,\nhave launched significant initiatives. These efforts include developing ethical\nguidelines for AI and engaging in vibrant discussions on AI ethics, both among\nAI practitioners and within the broader society. This article thoroughly\nanalyzes the ground-breaking AI regulatory framework proposed by the European\nUnion. It delves into the fundamental ethical principles of safety,\ntransparency, non-discrimination, traceability, and environmental\nsustainability for AI developments and deployments. Considering the technical\nefforts and strategies undertaken by academics and industry to uphold these\nprinciples, we explore the synergies and conflicts among the five ethical\nprinciples. Through this lens, work presents a forward-looking perspective on\nthe future of AI regulations, advocating for a harmonized approach that\nsafeguards societal values while encouraging technological advancement.\n", "date": "2024-12-06", "categories": []}
{"id": "2502.03038", "title": "The Cake that is Intelligence and Who Gets to Bake it: An AI Analogy and\n  its Implications for Participation", "abstract": "  In a widely popular analogy by Turing Award Laureate Yann LeCun, machine\nintelligence has been compared to cake - where unsupervised learning forms the\nbase, supervised learning adds the icing, and reinforcement learning is the\ncherry on top. We expand this 'cake that is intelligence' analogy from a simple\nstructural metaphor to the full life-cycle of AI systems, extending it to\nsourcing of ingredients (data), conception of recipes (instructions), the\nbaking process (training), and the tasting and selling of the cake (evaluation\nand distribution). Leveraging our re-conceptualization, we describe each step's\nentailed social ramifications and how they are bounded by statistical\nassumptions within machine learning. Whereas these technical foundations and\nsocial impacts are deeply intertwined, they are often studied in isolation,\ncreating barriers that restrict meaningful participation. Our\nre-conceptualization paves the way to bridge this gap by mapping where\ntechnical foundations interact with social outcomes, highlighting opportunities\nfor cross-disciplinary dialogue. Finally, we conclude with actionable\nrecommendations at each stage of the metaphorical AI cake's life-cycle,\nempowering prospective AI practitioners, users, and researchers, with increased\nawareness and ability to engage in broader AI discourse.\n", "date": "2025-02-05", "categories": []}
{"id": "2502.03469", "title": "A Capability Approach to AI Ethics", "abstract": "  We propose a conceptualization and implementation of AI ethics via the\ncapability approach. We aim to show that conceptualizing AI ethics through the\ncapability approach has two main advantages for AI ethics as a discipline.\nFirst, it helps clarify the ethical dimension of AI tools. Second, it provides\nguidance to implementing ethical considerations within the design of AI tools.\nWe illustrate these advantages in the context of AI tools in medicine, by\nshowing how ethics-based auditing of AI tools in medicine can greatly benefit\nfrom our capability-based approach.\n", "date": "2025-01-10", "categories": []}
{"id": "2501.19407", "title": "Algorithmic Inheritance: Surname Bias in AI Decisions Reinforces\n  Intergenerational Inequality", "abstract": "  Surnames often convey implicit markers of social status, wealth, and lineage,\nshaping perceptions in ways that can perpetuate systemic biases and\nintergenerational inequality. This study is the first of its kind to\ninvestigate whether and how surnames influence AI-driven decision-making,\nfocusing on their effects across key areas such as hiring recommendations,\nleadership appointments, and loan approvals. Using 72,000 evaluations of 600\nsurnames from the United States and Thailand, two countries with distinct\nsociohistorical contexts and surname conventions, we classify names into four\ncategories: Rich, Legacy, Normal, and phonetically similar Variant groups. Our\nfindings show that elite surnames consistently increase AI-generated\nperceptions of power, intelligence, and wealth, which in turn influence\nAI-driven decisions in high-stakes contexts. Mediation analysis reveals\nperceived intelligence as a key mechanism through which surname biases\ninfluence AI decision-making process. While providing objective qualifications\nalongside surnames mitigates most of these biases, it does not eliminate them\nentirely, especially in contexts where candidate credentials are low. These\nfindings highlight the need for fairness-aware algorithms and robust policy\nmeasures to prevent AI systems from reinforcing systemic inequalities tied to\nsurnames, an often-overlooked bias compared to more salient characteristics\nsuch as race and gender. Our work calls for a critical reassessment of\nalgorithmic accountability and its broader societal impact, particularly in\nsystems designed to uphold meritocratic principles while counteracting the\nperpetuation of intergenerational privilege.\n", "date": "2025-01-23", "categories": []}
{"id": "2502.03376", "title": "Ethical Considerations for the Military Use of Artificial Intelligence\n  in Visual Reconnaissance", "abstract": "  This white paper underscores the critical importance of responsibly deploying\nArtificial Intelligence (AI) in military contexts, emphasizing a commitment to\nethical and legal standards. The evolving role of AI in the military goes\nbeyond mere technical applications, necessitating a framework grounded in\nethical principles. The discussion within the paper delves into ethical AI\nprinciples, particularly focusing on the Fairness, Accountability,\nTransparency, and Ethics (FATE) guidelines. Noteworthy considerations encompass\ntransparency, justice, non-maleficence, and responsibility. Importantly, the\npaper extends its examination to military-specific ethical considerations,\ndrawing insights from the Just War theory and principles established by\nprominent entities. In addition to the identified principles, the paper\nintroduces further ethical considerations specifically tailored for military AI\napplications. These include traceability, proportionality, governability,\nresponsibility, and reliability. The application of these ethical principles is\ndiscussed on the basis of three use cases in the domains of sea, air, and land.\nMethods of automated sensor data analysis, eXplainable AI (XAI), and intuitive\nuser experience are utilized to specify the use cases close to real-world\nscenarios. This comprehensive approach to ethical considerations in military AI\nreflects a commitment to aligning technological advancements with established\nethical frameworks. It recognizes the need for a balance between leveraging\nAI's potential benefits in military operations while upholding moral and legal\nstandards. The inclusion of these ethical principles serves as a foundation for\nresponsible and accountable use of AI in the complex and dynamic landscape of\nmilitary scenarios.\n", "date": "2025-02-05", "categories": []}
{"id": "2412.09961", "title": "What constitutes a Deep Fake? The blurry line between legitimate\n  processing and manipulation under the EU AI Act", "abstract": "  When does a digital image resemble reality? The relevance of this question\nincreases as the generation of synthetic images -- so called deep fakes --\nbecomes increasingly popular. Deep fakes have gained much attention for a\nnumber of reasons -- among others, due to their potential to disrupt the\npolitical climate. In order to mitigate these threats, the EU AI Act implements\nspecific transparency regulations for generating synthetic content or\nmanipulating existing content. However, the distinction between real and\nsynthetic images is -- even from a computer vision perspective -- far from\ntrivial. We argue that the current definition of deep fakes in the AI act and\nthe corresponding obligations are not sufficiently specified to tackle the\nchallenges posed by deep fakes. By analyzing the life cycle of a digital photo\nfrom the camera sensor to the digital editing features, we find that: (1.) Deep\nfakes are ill-defined in the EU AI Act. The definition leaves too much scope\nfor what a deep fake is. (2.) It is unclear how editing functions like Google's\n``best take'' feature can be considered as an exception to transparency\nobligations. (3.) The exception for substantially edited images raises\nquestions about what constitutes substantial editing of content and whether or\nnot this editing must be perceptible by a natural person. Our results\ndemonstrate that complying with the current AI Act transparency obligations is\ndifficult for providers and deployers. As a consequence of the unclear\nprovisions, there is a risk that exceptions may be either too broad or too\nlimited. We intend our analysis to foster the discussion on what constitutes a\ndeep fake and to raise awareness about the pitfalls in the current AI Act\ntransparency obligations.\n", "date": "2024-12-13", "categories": []}
{"id": "2502.00011", "title": "TOAST Framework: A Multidimensional Approach to Ethical and Sustainable\n  AI Integration in Organizations", "abstract": "  Artificial Intelligence (AI) has emerged as a transformative technology with\nthe potential to revolutionize various sectors, from healthcare to finance,\neducation, and beyond. However, successfully implementing AI systems remains a\ncomplex challenge, requiring a comprehensive and methodologically sound\nframework. This paper contributes to this challenge by introducing the\nTrustworthy, Optimized, Adaptable, and Socio-Technologically harmonious (TOAST)\nframework. It draws on insights from various disciplines to align technical\nstrategy with ethical values, societal responsibilities, and innovation\naspirations. The TOAST framework is a novel approach designed to guide the\nimplementation of AI systems, focusing on reliability, accountability,\ntechnical advancement, adaptability, and socio-technical harmony. By grounding\nthe TOAST framework in healthcare case studies, this paper provides a robust\nevaluation of its practicality and theoretical soundness in addressing\noperational, ethical, and regulatory challenges in high-stakes environments,\ndemonstrating how adaptable AI systems can enhance institutional efficiency,\nmitigate risks like bias and data privacy, and offer a replicable model for\nother sectors requiring ethically aligned and efficient AI integration.\n", "date": "2025-01-07", "categories": []}
{"id": "2502.00648", "title": "Agency in the Age of AI", "abstract": "  There is significant concern about the impact of generative AI on society.\nModern AI tools are capable of generating ever more realistic text, images, and\nvideos, and functional code, from minimal prompts. Accompanying this rise in\nability and usability, there is increasing alarm about the misuses to which\nthese tools can be put, and the intentional and unintentional harms to\nindividuals and society that may result. In this paper, we argue that\n\\emph{agency} is the appropriate lens to study these harms and benefits, but\nthat doing so will require advancement in the theory of agency, and advancement\nin how this theory is applied in (agent-based) models.\n", "date": "2025-02-02", "categories": []}
{"id": "2501.14778", "title": "Advancing Trustworthy AI for Sustainable Development: Recommendations\n  for Standardising AI Incident Reporting", "abstract": "  The increasing use of AI technologies has led to increasing AI incidents,\nposing risks and causing harm to individuals, organizations, and society. This\nstudy recognizes and addresses the lack of standardized protocols for reliably\nand comprehensively gathering such incident data crucial for preventing future\nincidents and developing mitigating strategies. Specifically, this study\nanalyses existing open-access AI-incident databases through a systematic\nmethodology and identifies nine gaps in current AI incident reporting\npractices. Further, it proposes nine actionable recommendations to enhance\nstandardization efforts to address these gaps. Ensuring the trustworthiness of\nenabling technologies such as AI is necessary for sustainable digital\ntransformation. Our research promotes the development of standards to prevent\nfuture AI incidents and promote trustworthy AI, thus facilitating achieving the\nUN sustainable development goals. Through international cooperation,\nstakeholders can unlock the transformative potential of AI, enabling a\nsustainable and inclusive future for all.\n", "date": "2025-01-01", "categories": []}
{"id": "2501.15634", "title": "Be Intentional About Fairness!: Fairness, Size, and Multiplicity in the\n  Rashomon Set", "abstract": "  When selecting a model from a set of equally performant models, how much\nunfairness can you really reduce? Is it important to be intentional about\nfairness when choosing among this set, or is arbitrarily choosing among the set\nof ''good'' models good enough? Recent work has highlighted that the phenomenon\nof model multiplicity-where multiple models with nearly identical predictive\naccuracy exist for the same task-has both positive and negative implications\nfor fairness, from strengthening the enforcement of civil rights law in AI\nsystems to showcasing arbitrariness in AI decision-making. Despite the enormous\nimplications of model multiplicity, there is little work that explores the\nproperties of sets of equally accurate models, or Rashomon sets, in general. In\nthis paper, we present five main theoretical and methodological contributions\nwhich help us to understand the relatively unexplored properties of the\nRashomon set, in particular with regards to fairness. Our contributions include\nmethods for efficiently sampling models from this set and techniques for\nidentifying the fairest models according to key fairness metrics such as\nstatistical parity. We also derive the probability that an individual's\nprediction will be flipped within the Rashomon set, as well as expressions for\nthe set's size and the distribution of error tolerance used across models.\nThese results lead to policy-relevant takeaways, such as the importance of\nintentionally looking for fair models within the Rashomon set, and\nunderstanding which individuals or groups may be more susceptible to arbitrary\ndecisions.\n", "date": "2025-01-26", "categories": []}
{"id": "2310.13192", "title": "The Opaque Law of Artificial Intelligence", "abstract": "  The purpose of this paper is to analyse the opacity of algorithms,\ncontextualized in the open debate on responsibility for artificial intelligence\ncausation; with an experimental approach by which, applying the proposed\nconversational methodology of the Turing Test, we expect to evaluate the\nperformance of one of the best existing NLP model of generative AI (Chat-GPT)\nto see how far it can go right now and how the shape of a legal regulation of\nit could be. The analysis of the problem will be supported by a comment of\nItalian classical law categories such as causality, intent and fault to\nunderstand the problem of the usage of AI, focusing in particular on the\nhuman-machine interaction. On the computer science side, for a technical point\nof view of the logic used to craft these algorithms, in the second chapter will\nbe proposed a practical interrogation of Chat-GPT aimed at finding some\ncritical points of the functioning of AI. The end of the paper will concentrate\non some existing legal solutions which can be applied to the problem, plus a\nbrief description of the approach proposed by EU Artificial Intelligence act.\n", "date": "2023-10-19", "categories": []}
{"id": "2412.01957", "title": "Usage Governance Advisor: From Intent to AI Governance", "abstract": "  Evaluating the safety of AI Systems is a pressing concern for organizations\ndeploying them. In addition to the societal damage done by the lack of fairness\nof those systems, deployers are concerned about the legal repercussions and the\nreputational damage incurred by the use of models that are unsafe. Safety\ncovers both what a model does; e.g., can it be used to reveal personal\ninformation from its training set, and how a model was built; e.g., was it only\ntrained on licensed data sets. Determining the safety of an AI system requires\ngathering information from a wide set of heterogeneous sources including safety\nbenchmarks and technical documentation for the set of models used in that\nsystem. In addition, responsible use is encouraged through mechanisms that\nadvise and help the user to take mitigating actions where safety risks are\ndetected. We present Usage Governance Advisor which creates semi-structured\ngovernance information, identifies and prioritizes risks according to the\nintended use case, recommends appropriate benchmarks and risk assessments and\nimportantly proposes mitigation strategies and actions.\n", "date": "2024-12-02", "categories": []}
{"id": "2501.13302", "title": "Watching the AI Watchdogs: A Fairness and Robustness Analysis of AI\n  Safety Moderation Classifiers", "abstract": "  AI Safety Moderation (ASM) classifiers are designed to moderate content on\nsocial media platforms and to serve as guardrails that prevent Large Language\nModels (LLMs) from being fine-tuned on unsafe inputs. Owing to their potential\nfor disparate impact, it is crucial to ensure that these classifiers: (1) do\nnot unfairly classify content belonging to users from minority groups as unsafe\ncompared to those from majority groups and (2) that their behavior remains\nrobust and consistent across similar inputs. In this work, we thus examine the\nfairness and robustness of four widely-used, closed-source ASM classifiers:\nOpenAI Moderation API, Perspective API, Google Cloud Natural Language (GCNL)\nAPI, and Clarifai API. We assess fairness using metrics such as demographic\nparity and conditional statistical parity, comparing their performance against\nASM models and a fair-only baseline. Additionally, we analyze robustness by\ntesting the classifiers' sensitivity to small and natural input perturbations.\nOur findings reveal potential fairness and robustness gaps, highlighting the\nneed to mitigate these issues in future versions of these models.\n", "date": "2025-01-23", "categories": []}
{"id": "2501.13533", "title": "Towards a Theory of AI Personhood", "abstract": "  I am a person and so are you. Philosophically we sometimes grant personhood\nto non-human animals, and entities such as sovereign states or corporations can\nlegally be considered persons. But when, if ever, should we ascribe personhood\nto AI systems? In this paper, we outline necessary conditions for AI\npersonhood, focusing on agency, theory-of-mind, and self-awareness. We discuss\nevidence from the machine learning literature regarding the extent to which\ncontemporary AI systems, such as language models, satisfy these conditions,\nfinding the evidence surprisingly inconclusive.\n  If AI systems can be considered persons, then typical framings of AI\nalignment may be incomplete. Whereas agency has been discussed at length in the\nliterature, other aspects of personhood have been relatively neglected. AI\nagents are often assumed to pursue fixed goals, but AI persons may be\nself-aware enough to reflect on their aims, values, and positions in the world\nand thereby induce their goals to change. We highlight open research directions\nto advance the understanding of AI personhood and its relevance to alignment.\nFinally, we reflect on the ethical considerations surrounding the treatment of\nAI systems. If AI systems are persons, then seeking control and alignment may\nbe ethically untenable.\n", "date": "2025-01-23", "categories": []}
{"id": "2408.11820", "title": "Responsible AI Question Bank: A Comprehensive Tool for AI Risk\n  Assessment", "abstract": "  The rapid growth of Artificial Intelligence (AI) has underscored the urgent\nneed for responsible AI practices. Despite increasing interest, a comprehensive\nAI risk assessment toolkit remains lacking. This study introduces our\nResponsible AI (RAI) Question Bank, a comprehensive framework and tool designed\nto support diverse AI initiatives. By integrating AI ethics principles such as\nfairness, transparency, and accountability into a structured question format,\nthe RAI Question Bank aids in identifying potential risks, aligning with\nemerging regulations like the EU AI Act, and enhancing overall AI governance. A\nkey benefit of the RAI Question Bank is its systematic approach to linking\nlower-level risk questions to higher-level ones and related themes, preventing\nsiloed assessments and ensuring a cohesive evaluation process. Case studies\nillustrate the practical application of the RAI Question Bank in assessing AI\nprojects, from evaluating risk factors to informing decision-making processes.\nThe study also demonstrates how the RAI Question Bank can be used to ensure\ncompliance with standards, mitigate risks, and promote the development of\ntrustworthy AI systems. This work advances RAI by providing organizations with\na valuable tool to navigate the complexities of ethical AI development and\ndeployment while ensuring comprehensive risk management.\n", "date": "2024-08-02", "categories": []}
{"id": "2501.10371", "title": "What we learned while automating bias detection in AI hiring systems for\n  compliance with NYC Local Law 144", "abstract": "  Since July 5, 2023, New York City's Local Law 144 requires employers to\nconduct independent bias audits for any automated employment decision tools\n(AEDTs) used in hiring processes. The law outlines a minimum set of bias tests\nthat AI developers and implementers must perform to ensure compliance. Over the\npast few months, we have collected and analyzed audits conducted under this\nlaw, identified best practices, and developed a software tool to streamline\nemployer compliance. Our tool, ITACA_144, tailors our broader bias auditing\nframework to meet the specific requirements of Local Law 144. While automating\nthese legal mandates, we identified several critical challenges that merit\nattention to ensure AI bias regulations and audit methodologies are both\neffective and practical. This document presents the insights gained from\nautomating compliance with NYC Local Law 144. It aims to support other cities\nand states in crafting similar legislation while addressing the limitations of\nthe NYC framework. The discussion focuses on key areas including data\nrequirements, demographic inclusiveness, impact ratios, effective bias,\nmetrics, and data reliability.\n", "date": "2024-12-13", "categories": []}
{"id": "2501.10390", "title": "Towards an Environmental Ethics of Artificial Intelligence", "abstract": "  In recent years, much research has been dedicated to uncovering the\nenvironmental impact of Artificial Intelligence (AI), showing that training and\ndeploying AI systems require large amounts of energy and resources, and the\noutcomes of AI may lead to decisions and actions that may negatively impact the\nenvironment. This new knowledge raises new ethical questions, such as: When is\nit (un)justifiable to develop an AI system, and how to make design choices,\nconsidering its environmental impact? However, so far, the environmental impact\nof AI has largely escaped ethical scrutiny, as AI ethics tends to focus\nstrongly on themes such as transparency, privacy, safety, responsibility, and\nbias. Considering the environmental impact of AI from an ethical perspective\nexpands the scope of AI ethics beyond an anthropocentric focus towards\nincluding more-than-human actors such as animals and ecosystems. This paper\nexplores the ethical implications of the environmental impact of AI for\ndesigning AI systems by drawing on environmental justice literature, in which\nthree categories of justice are distinguished, referring to three elements that\ncan be unjust: the distribution of benefits and burdens (distributive justice),\ndecision-making procedures (procedural justice), and institutionalized social\nnorms (justice as recognition). Based on these tenets of justice, we outline\ncriteria for developing environmentally just AI systems, given their ecological\nimpact.\n", "date": "2024-12-19", "categories": []}
{"id": "2501.10467", "title": "Securing the AI Frontier: Urgent Ethical and Regulatory Imperatives for\n  AI-Driven Cybersecurity", "abstract": "  This paper critically examines the evolving ethical and regulatory challenges\nposed by the integration of artificial intelligence (AI) in cybersecurity. We\ntrace the historical development of AI regulation, highlighting major\nmilestones from theoretical discussions in the 1940s to the implementation of\nrecent global frameworks such as the European Union AI Act. The current\nregulatory landscape is analyzed, emphasizing risk-based approaches,\nsector-specific regulations, and the tension between fostering innovation and\nmitigating risks. Ethical concerns such as bias, transparency, accountability,\nprivacy, and human oversight are explored in depth, along with their\nimplications for AI-driven cybersecurity systems. Furthermore, we propose\nstrategies for promoting AI literacy and public engagement, essential for\nshaping a future regulatory framework. Our findings underscore the need for a\nunified, globally harmonized regulatory approach that addresses the unique\nrisks of AI in cybersecurity. We conclude by identifying future research\nopportunities and recommending pathways for collaboration between policymakers,\nindustry leaders, and researchers to ensure the responsible deployment of AI\ntechnologies in cybersecurity.\n", "date": "2025-01-15", "categories": []}
{"id": "2501.11705", "title": "Human services organizations and the responsible integration of AI:\n  Considering ethics and contextualizing risk(s)", "abstract": "  This paper examines the responsible integration of artificial intelligence\n(AI) in human services organizations (HSOs), proposing a nuanced framework for\nevaluating AI applications across multiple dimensions of risk. The authors\nargue that ethical concerns about AI deployment -- including professional\njudgment displacement, environmental impact, model bias, and data laborer\nexploitation -- vary significantly based on implementation context and specific\nuse cases. They challenge the binary view of AI adoption, demonstrating how\ndifferent applications present varying levels of risk that can often be\neffectively managed through careful implementation strategies. The paper\nhighlights promising solutions, such as local large language models, that can\nfacilitate responsible AI integration while addressing common ethical concerns.\nThe authors propose a dimensional risk assessment approach that considers\nfactors like data sensitivity, professional oversight requirements, and\npotential impact on client wellbeing. They conclude by outlining a path forward\nthat emphasizes empirical evaluation, starting with lower-risk applications and\nbuilding evidence-based understanding through careful experimentation. This\napproach enables organizations to maintain high ethical standards while\nthoughtfully exploring how AI might enhance their capacity to serve clients and\ncommunities effectively.\n", "date": "2025-01-20", "categories": []}
{"id": "2501.11909", "title": "Bridging the Communication Gap: Evaluating AI Labeling Practices for\n  Trustworthy AI Development", "abstract": "  As artificial intelligence (AI) becomes integral to economy and society,\ncommunication gaps between developers, users, and stakeholders hinder trust and\ninformed decision-making. High-level AI labels, inspired by frameworks like EU\nenergy labels, have been proposed to make the properties of AI models more\ntransparent. Without requiring deep technical expertise, they can inform on the\ntrade-off between predictive performance and resource efficiency. However, the\npractical benefits and limitations of AI labeling remain underexplored. This\nstudy evaluates AI labeling through qualitative interviews along four key\nresearch questions. Based on thematic analysis and inductive coding, we found a\nbroad range of practitioners to be interested in AI labeling (RQ1). They see\nbenefits for alleviating communication gaps and aiding non-expert\ndecision-makers, however limitations, misunderstandings, and suggestions for\nimprovement were also discussed (RQ2). Compared to other reporting formats,\ninterviewees positively evaluated the reduced complexity of labels, increasing\noverall comprehensibility (RQ3). Trust was influenced most by usability and the\ncredibility of the responsible labeling authority, with mixed preferences for\nself-certification versus third-party certification (RQ4). Our Insights\nhighlight that AI labels pose a trade-off between simplicity and complexity,\nwhich could be resolved by developing customizable and interactive labeling\nframeworks to address diverse user needs. Transparent labeling of resource\nefficiency also nudged interviewee priorities towards paying more attention to\nsustainability aspects during AI development. This study validates AI labels as\na valuable tool for enhancing trust and communication in AI, offering\nactionable guidelines for their refinement and standardization.\n", "date": "2025-01-21", "categories": []}
{"id": "2312.01818", "title": "Hybrid Approaches for Moral Value Alignment in AI Agents: a Manifesto", "abstract": "  Increasing interest in ensuring the safety of next-generation Artificial\nIntelligence (AI) systems calls for novel approaches to embedding morality into\nautonomous agents. This goal differs qualitatively from traditional\ntask-specific AI methodologies. In this paper, we provide a systematization of\nexisting approaches to the problem of introducing morality in machines -\nmodelled as a continuum. Our analysis suggests that popular techniques lie at\nthe extremes of this continuum - either being fully hard-coded into top-down,\nexplicit rules, or entirely learned in a bottom-up, implicit fashion with no\ndirect statement of any moral principle (this includes learning from human\nfeedback, as applied to the training and finetuning of large language models,\nor LLMs). Given the relative strengths and weaknesses of each type of\nmethodology, we argue that more hybrid solutions are needed to create adaptable\nand robust, yet controllable and interpretable agentic systems. To that end,\nthis paper discusses both the ethical foundations (including deontology,\nconsequentialism and virtue ethics) and implementations of morally aligned AI\nsystems.\n  We present a series of case studies that rely on intrinsic rewards, moral\nconstraints or textual instructions, applied to either pure-Reinforcement\nLearning or LLM-based agents. By analysing these diverse implementations under\none framework, we compare their relative strengths and shortcomings in\ndeveloping morally aligned AI systems. We then discuss strategies for\nevaluating the effectiveness of moral learning agents. Finally, we present open\nresearch questions and implications for the future of AI safety and ethics\nwhich are emerging from this hybrid framework.\n", "date": "2023-12-04", "categories": []}
{"id": "2501.09021", "title": "Navigating Ethical Challenges in Generative AI-Enhanced Research: The\n  ETHICAL Framework for Responsible Generative AI Use", "abstract": "  The rapid adoption of generative artificial intelligence (GenAI) in research\npresents both opportunities and ethical challenges that should be carefully\nnavigated. Although GenAI tools can enhance research efficiency through\nautomation of tasks such as literature review and data analysis, their use\nraises concerns about aspects such as data accuracy, privacy, bias, and\nresearch integrity. This paper develops the ETHICAL framework, which is a\npractical guide for responsible GenAI use in research. Employing a\nconstructivist case study examining multiple GenAI tools in real research\ncontexts, the framework consists of seven key principles: Examine policies and\nguidelines, Think about social impacts, Harness understanding of the\ntechnology, Indicate use, Critically engage with outputs, Access secure\nversions, and Look at user agreements. Applying these principles will enable\nresearchers to uphold research integrity while leveraging GenAI benefits. The\nframework addresses a critical gap between awareness of ethical issues and\npractical action steps, providing researchers with concrete guidance for\nethical GenAI integration. This work has implications for research practice,\ninstitutional policy development, and the broader academic community while\nadapting to an AI-enhanced research landscape. The ETHICAL framework can serve\nas a foundation for developing AI literacy in academic settings and promoting\nresponsible innovation in research methodologies.\n", "date": "2024-12-11", "categories": []}
{"id": "2501.09534", "title": "AI in Support of Diversity and Inclusion", "abstract": "  In this paper, we elaborate on how AI can support diversity and inclusion and\nexemplify research projects conducted in that direction. We start by looking at\nthe challenges and progress in making large language models (LLMs) more\ntransparent, inclusive, and aware of social biases. Even though LLMs like\nChatGPT have impressive abilities, they struggle to understand different\ncultural contexts and engage in meaningful, human like conversations. A key\nissue is that biases in language processing, especially in machine translation,\ncan reinforce inequality. Tackling these biases requires a multidisciplinary\napproach to ensure AI promotes diversity, fairness, and inclusion. We also\nhighlight AI's role in identifying biased content in media, which is important\nfor improving representation. By detecting unequal portrayals of social groups,\nAI can help challenge stereotypes and create more inclusive technologies.\nTransparent AI algorithms, which clearly explain their decisions, are essential\nfor building trust and reducing bias in AI systems. We also stress AI systems\nneed diverse and inclusive training data. Projects like the Child Growth\nMonitor show how using a wide range of data can help address real world\nproblems like malnutrition and poverty. We present a project that demonstrates\nhow AI can be applied to monitor the role of search engines in spreading\ndisinformation about the LGBTQ+ community. Moreover, we discuss the SignON\nproject as an example of how technology can bridge communication gaps between\nhearing and deaf people, emphasizing the importance of collaboration and mutual\ntrust in developing inclusive AI. Overall, with this paper, we advocate for AI\nsystems that are not only effective but also socially responsible, promoting\nfair and inclusive interactions between humans and machines.\n", "date": "2025-01-16", "categories": []}
{"id": "2501.08951", "title": "Analyzing the Ethical Logic of Six Large Language Models", "abstract": "  This study examines the ethical reasoning of six prominent generative large\nlanguage models: OpenAI GPT-4o, Meta LLaMA 3.1, Perplexity, Anthropic Claude\n3.5 Sonnet, Google Gemini, and Mistral 7B. The research explores how these\nmodels articulate and apply ethical logic, particularly in response to moral\ndilemmas such as the Trolley Problem, and Heinz Dilemma. Departing from\ntraditional alignment studies, the study adopts an explainability-transparency\nframework, prompting models to explain their ethical reasoning. This approach\nis analyzed through three established ethical typologies: the\nconsequentialist-deontological analytic, Moral Foundations Theory, and the\nKohlberg Stages of Moral Development Model. Findings reveal that LLMs exhibit\nlargely convergent ethical logic, marked by a rationalist, consequentialist\nemphasis, with decisions often prioritizing harm minimization and fairness.\nDespite similarities in pre-training and model architecture, a mixture of\nnuanced and significant differences in ethical reasoning emerge across models,\nreflecting variations in fine-tuning and post-training processes. The models\nconsistently display erudition, caution, and self-awareness, presenting ethical\nreasoning akin to a graduate-level discourse in moral philosophy. In striking\nuniformity these systems all describe their ethical reasoning as more\nsophisticated than what is characteristic of typical human moral logic.\n", "date": "2025-01-15", "categories": []}
{"id": "2305.03123", "title": "ChatGPT Needs SPADE (Sustainability, PrivAcy, Digital divide, and\n  Ethics) Evaluation: A Review", "abstract": "  ChatGPT is another large language model (LLM) vastly available for the\nconsumers on their devices but due to its performance and ability to converse\neffectively, it has gained a huge popularity amongst research as well as\nindustrial community. Recently, many studies have been published to show the\neffectiveness, efficiency, integration, and sentiments of chatGPT and other\nLLMs. In contrast, this study focuses on the important aspects that are mostly\noverlooked, i.e. sustainability, privacy, digital divide, and ethics and\nsuggests that not only chatGPT but every subsequent entry in the category of\nconversational bots should undergo Sustainability, PrivAcy, Digital divide, and\nEthics (SPADE) evaluation. This paper discusses in detail the issues and\nconcerns raised over chatGPT in line with aforementioned characteristics. We\nalso discuss the recent EU AI Act briefly in accordance with the SPADE\nevaluation. We support our hypothesis by some preliminary data collection and\nvisualizations along with hypothesized facts. We also suggest mitigations and\nrecommendations for each of the concerns. Furthermore, we also suggest some\npolicies and recommendations for EU AI policy act concerning ethics, digital\ndivide, and sustainability\n", "date": "2023-04-13", "categories": []}
{"id": "2412.17114", "title": "Decentralized Governance of Autonomous AI Agents", "abstract": "  Autonomous AI agents present transformative opportunities and significant\ngovernance challenges. Existing frameworks, such as the EU AI Act and the NIST\nAI Risk Management Framework, fall short of addressing the complexities of\nthese agents, which are capable of independent decision-making, learning, and\nadaptation. To bridge these gaps, we propose the ETHOS (Ethical Technology and\nHolistic Oversight System) framework, a decentralized governance (DeGov) model\nleveraging Web3 technologies, including blockchain, smart contracts, and\ndecentralized autonomous organizations (DAOs). ETHOS establishes a global\nregistry for AI agents, enabling dynamic risk classification, proportional\noversight, and automated compliance monitoring through tools like soulbound\ntokens and zero-knowledge proofs. Furthermore, the framework incorporates\ndecentralized justice systems for transparent dispute resolution and introduces\nAI specific legal entities to manage limited liability, supported by mandatory\ninsurance to ensure financial accountability and incentivize ethical design. By\nintegrating philosophical principles of rationality, ethical grounding, and\ngoal alignment, ETHOS aims to create a robust research agenda for promoting\ntrust, transparency, and participatory governance. This innovative framework\noffers a scalable and inclusive strategy for regulating AI agents, balancing\ninnovation with ethical responsibility to meet the demands of an AI-driven\nfuture.\n", "date": "2024-12-22", "categories": []}
{"id": "2501.07290", "title": "Principles for Responsible AI Consciousness Research", "abstract": "  Recent research suggests that it may be possible to build conscious AI\nsystems now or in the near future. Conscious AI systems would arguably deserve\nmoral consideration, and it may be the case that large numbers of conscious\nsystems could be created and caused to suffer. Furthermore, AI systems or\nAI-generated characters may increasingly give the impression of being\nconscious, leading to debate about their moral status. Organisations involved\nin AI research must establish principles and policies to guide research and\ndeployment choices and public communication concerning consciousness. Even if\nan organisation chooses not to study AI consciousness as such, it will still\nneed policies in place, as those developing advanced AI systems risk\ninadvertently creating conscious entities. Responsible research and deployment\npractices are essential to address this possibility. We propose five principles\nfor responsible research and argue that research organisations should make\nvoluntary, public commitments to principles on these lines. Our principles\nconcern research objectives and procedures, knowledge sharing and public\ncommunications.\n", "date": "2025-01-13", "categories": []}
{"id": "2501.05391", "title": "The global consensus on the risk management of autonomous driving", "abstract": "  Every maneuver of a vehicle redistributes risks between road users. While\nhuman drivers do this intuitively, autonomous vehicles allow and require\ndeliberative algorithmic risk management. But how should traffic risks be\ndistributed among road users? In a global experimental study in eight countries\nwith different cultural backgrounds and almost 11,000 participants, we compared\nrisk distribution preferences. It turns out that risk preferences in road\ntraffic are strikingly similar between the cultural zones. The vast majority of\nparticipants in all countries deviates from a guiding principle of minimizing\naccident probabilities in favor of weighing up the probability and severity of\naccidents. At the national level, the consideration of accident probability and\nseverity hardly differs between countries. The social dilemma of autonomous\nvehicles detected in deterministic crash scenarios disappears in risk\nassessments of everyday traffic situations in all countries. In no country do\ncyclists receive a risk bonus that goes beyond their higher vulnerability. In\nsum, our results suggest that a global consensus on the risk ethics of\nautonomous driving is easier to establish than on the ethics of crashing.\n", "date": "2025-01-09", "categories": []}
{"id": "2403.06537", "title": "On the Consideration of AI Openness: Can Good Intent Be Abused?", "abstract": "  Open source is a driving force behind scientific advancement.However, this\nopenness is also a double-edged sword, with the inherent risk that innovative\ntechnologies can be misused for purposes harmful to society. What is the\nlikelihood that an open source AI model or dataset will be used to commit a\nreal-world crime, and if a criminal does exploit it, will the people behind the\ntechnology be able to escape legal liability? To address these questions, we\nexplore a legal domain where individual choices can have a significant impact\non society. Specifically, we build the EVE-V1 dataset that comprises 200\nquestion-answer pairs related to criminal offenses based on 200 Korean\nprecedents first to explore the possibility of malicious models emerging. We\nfurther developed EVE-V2 using 600 fraud-related precedents to confirm the\nexistence of malicious models that can provide harmful advice on a wide range\nof criminal topics to test the domain generalization ability. Remarkably,\nwidely used open-source large-scale language models (LLMs) provide unethical\nand detailed information about criminal activities when fine-tuned with EVE. We\nalso take an in-depth look at the legal issues that malicious language models\nand their builders could realistically face. Our findings highlight the\nparadoxical dilemma that open source accelerates scientific progress, but\nrequires great care to minimize the potential for misuse. Warning: This paper\ncontains content that some may find unethical.\n", "date": "2024-03-11", "categories": []}
{"id": "2501.00320", "title": "Autonomous Alignment with Human Value on Altruism through Considerate\n  Self-imagination and Theory of Mind", "abstract": "  With the widespread application of Artificial Intelligence (AI) in human\nsociety, enabling AI to autonomously align with human values has become a\npressing issue to ensure its sustainable development and benefit to humanity.\nOne of the most important aspects of aligning with human values is the\nnecessity for agents to autonomously make altruistic, safe, and ethical\ndecisions, considering and caring for human well-being. Current AI extremely\npursues absolute superiority in certain tasks, remaining indifferent to the\nsurrounding environment and other agents, which has led to numerous safety\nrisks. Altruistic behavior in human society originates from humans' capacity\nfor empathizing others, known as Theory of Mind (ToM), combined with predictive\nimaginative interactions before taking action to produce thoughtful and\naltruistic behaviors. Inspired by this, we are committed to endow agents with\nconsiderate self-imagination and ToM capabilities, driving them through\nimplicit intrinsic motivations to autonomously align with human altruistic\nvalues. By integrating ToM within the imaginative space, agents keep an eye on\nthe well-being of other agents in real time, proactively anticipate potential\nrisks to themselves and others, and make thoughtful altruistic decisions that\nbalance negative effects on the environment. The ancient Chinese story of Sima\nGuang Smashes the Vat illustrates the moral behavior of the young Sima Guang\nsmashed a vat to save a child who had accidentally fallen into it, which is an\nexcellent reference scenario for this paper. We design an experimental scenario\nsimilar to Sima Guang Smashes the Vat and its variants with different\ncomplexities, which reflects the trade-offs and comprehensive considerations\nbetween self-goals, altruistic rescue, and avoiding negative side effects.\n", "date": "2024-12-31", "categories": []}
{"id": "2304.11530", "title": "A Conceptual Algorithm for Applying Ethical Principles of AI to Medical\n  Practice", "abstract": "  Artificial Intelligence (AI) is poised to transform healthcare delivery\nthrough revolutionary advances in clinical decision support and diagnostic\ncapabilities. While human expertise remains foundational to medical practice,\nAI-powered tools are increasingly matching or exceeding specialist-level\nperformance across multiple domains, paving the way for a new era of\ndemocratized healthcare access. These systems promise to reduce disparities in\ncare delivery across demographic, racial, and socioeconomic boundaries by\nproviding high-quality diagnostic support at scale. As a result, advanced\nhealthcare services can be affordable to all populations, irrespective of\ndemographics, race, or socioeconomic background. The democratization of such AI\ntools can reduce the cost of care, optimize resource allocation, and improve\nthe quality of care. In contrast to humans, AI can potentially uncover complex\nrelationships in the data from a large set of inputs and lead to new\nevidence-based knowledge in medicine. However, integrating AI into healthcare\nraises several ethical and philosophical concerns, such as bias, transparency,\nautonomy, responsibility, and accountability. In this study, we examine recent\nadvances in AI-enabled medical image analysis, current regulatory frameworks,\nand emerging best practices for clinical integration. We analyze both technical\nand ethical challenges inherent in deploying AI systems across healthcare\ninstitutions, with particular attention to data privacy, algorithmic fairness,\nand system transparency. Furthermore, we propose practical solutions to address\nkey challenges, including data scarcity, racial bias in training datasets,\nlimited model interpretability, and systematic algorithmic biases. Finally, we\noutline a conceptual algorithm for responsible AI implementations and identify\npromising future research and development directions.\n", "date": "2023-04-23", "categories": []}
{"id": "2404.12104", "title": "Ethical-Lens: Curbing Malicious Usages of Open-Source Text-to-Image\n  Models", "abstract": "  The burgeoning landscape of text-to-image models, exemplified by innovations\nsuch as Midjourney and DALLE 3, has revolutionized content creation across\ndiverse sectors. However, these advancements bring forth critical ethical\nconcerns, particularly with the misuse of open-source models to generate\ncontent that violates societal norms. Addressing this, we introduce\nEthical-Lens, a framework designed to facilitate the value-aligned usage of\ntext-to-image tools without necessitating internal model revision. Ethical-Lens\nensures value alignment in text-to-image models across toxicity and bias\ndimensions by refining user commands and rectifying model outputs. Systematic\nevaluation metrics, combining GPT4-V, HEIM, and FairFace scores, assess\nalignment capability. Our experiments reveal that Ethical-Lens enhances\nalignment capabilities to levels comparable with or superior to commercial\nmodels like DALLE 3, ensuring user-generated content adheres to ethical\nstandards while maintaining image quality. This study indicates the potential\nof Ethical-Lens to ensure the sustainable development of open-source\ntext-to-image tools and their beneficial integration into society. Our code is\navailable at https://github.com/yuzhu-cai/Ethical-Lens.\n", "date": "2024-04-18", "categories": []}
{"id": "2412.12108", "title": "Responsible AI Governance: A Response to UN Interim Report on Governing\n  AI for Humanity", "abstract": "  This report presents a comprehensive response to the United Nation's Interim\nReport on Governing Artificial Intelligence (AI) for Humanity. It emphasizes\nthe transformative potential of AI in achieving the Sustainable Development\nGoals (SDGs) while acknowledging the need for robust governance to mitigate\nassociated risks. The response highlights opportunities for promoting\nequitable, secure, and inclusive AI ecosystems, which should be supported by\ninvestments in infrastructure and multi-stakeholder collaborations across\njurisdictions. It also underscores challenges, including societal inequalities\nexacerbated by AI, ethical concerns, and environmental impacts. Recommendations\nadvocate for legally binding norms, transparency, and multi-layered data\ngovernance models, alongside fostering AI literacy and capacity-building\ninitiatives. Internationally, the report calls for harmonising AI governance\nframeworks with established laws, human rights standards, and regulatory\napproaches. The report concludes with actionable principles for fostering\nresponsible AI governance through collaboration among governments, industry,\nacademia, and civil society, ensuring the development of AI aligns with\nuniversal human values and the public good.\n", "date": "2024-11-29", "categories": []}
{"id": "2306.11748", "title": "The Manipulation Problem: Conversational AI as a Threat to Epistemic\n  Agency", "abstract": "  The technology of Conversational AI has made significant advancements over\nthe last eighteen months. As a consequence, conversational agents are likely to\nbe deployed in the near future that are designed to pursue targeted influence\nobjectives. Sometimes referred to as the \"AI Manipulation Problem,\" the\nemerging risk is that consumers will unwittingly engage in real-time dialog\nwith predatory AI agents that can skillfully persuade them to buy particular\nproducts, believe particular pieces of misinformation, or fool them into\nrevealing sensitive personal data. For many users, current systems like ChatGPT\nand LaMDA feel safe because they are primarily text-based, but the industry is\nalready shifting towards real-time voice and photorealistic digital personas\nthat look, move, and express like real people. This will enable the deployment\nof agenda-driven Virtual Spokespeople (VSPs) that will be highly persuasive\nthrough real-time adaptive influence. This paper explores the manipulative\ntactics that are likely to be deployed through conversational AI agents, the\nunique threats such agents pose to the epistemic agency of human users, and the\nemerging need for policymakers to protect against the most likely predatory\npractices.\n", "date": "2023-06-19", "categories": []}
{"id": "2412.19010", "title": "A theory of appropriateness with applications to generative artificial\n  intelligence", "abstract": "  What is appropriateness? Humans navigate a multi-scale mosaic of interlocking\nnotions of what is appropriate for different situations. We act one way with\nour friends, another with our family, and yet another in the office. Likewise\nfor AI, appropriate behavior for a comedy-writing assistant is not the same as\nappropriate behavior for a customer-service representative. What determines\nwhich actions are appropriate in which contexts? And what causes these\nstandards to change over time? Since all judgments of AI appropriateness are\nultimately made by humans, we need to understand how appropriateness guides\nhuman decision making in order to properly evaluate AI decision making and\nimprove it. This paper presents a theory of appropriateness: how it functions\nin human society, how it may be implemented in the brain, and what it means for\nresponsible deployment of generative AI technology.\n", "date": "2024-12-26", "categories": []}
{"id": "2401.08103", "title": "Resolving Ethics Trade-offs in Implementing Responsible AI", "abstract": "  While the operationalisation of high-level AI ethics principles into\npractical AI/ML systems has made progress, there is still a theory-practice gap\nin managing tensions between the underlying AI ethics aspects. We cover five\napproaches for addressing the tensions via trade-offs, ranging from rudimentary\nto complex. The approaches differ in the types of considered context, scope,\nmethods for measuring contexts, and degree of justification. None of the\napproaches is likely to be appropriate for all organisations, systems, or\napplications. To address this, we propose a framework which consists of: (i)\nproactive identification of tensions, (ii) prioritisation and weighting of\nethics aspects, (iii) justification and documentation of trade-off decisions.\nThe proposed framework aims to facilitate the implementation of well-rounded\nAI/ML systems that are appropriate for potential regulatory requirements.\n", "date": "2024-01-16", "categories": []}
{"id": "2412.16389", "title": "Ethics and Technical Aspects of Generative AI Models in Digital Content\n  Creation", "abstract": "  Generative AI models like GPT-4o and DALL-E 3 are reshaping digital content\ncreation, offering industries tools to generate diverse and sophisticated text\nand images with remarkable creativity and efficiency. This paper examines both\nthe capabilities and challenges of these models within creative workflows.\nWhile they deliver high performance in generating content with creativity,\ndiversity, and technical precision, they also raise significant ethical\nconcerns. Our study addresses two key research questions: (a) how these models\nperform in terms of creativity, diversity, accuracy, and computational\nefficiency, and (b) the ethical risks they present, particularly concerning\nbias, authenticity, and potential misuse. Through a structured series of\nexperiments, we analyze their technical performance and assess the ethical\nimplications of their outputs, revealing that although generative models\nenhance creative processes, they often reflect biases from their training data\nand carry ethical vulnerabilities that require careful oversight. This research\nproposes ethical guidelines to support responsible AI integration into industry\npractices, fostering a balance between innovation and ethical integrity.\n", "date": "2024-12-20", "categories": []}
{"id": "2412.16022", "title": "The Only Way is Ethics: A Guide to Ethical Research with Large Language\n  Models", "abstract": "  There is a significant body of work looking at the ethical considerations of\nlarge language models (LLMs): critiquing tools to measure performance and\nharms; proposing toolkits to aid in ideation; discussing the risks to workers;\nconsidering legislation around privacy and security etc. As yet there is no\nwork that integrates these resources into a single practical guide that focuses\non LLMs; we attempt this ambitious goal. We introduce 'LLM Ethics Whitepaper',\nwhich we provide as an open and living resource for NLP practitioners, and\nthose tasked with evaluating the ethical implications of others' work. Our goal\nis to translate ethics literature into concrete recommendations and\nprovocations for thinking with clear first steps, aimed at computer scientists.\n'LLM Ethics Whitepaper' distils a thorough literature review into clear Do's\nand Don'ts, which we present also in this paper. We likewise identify useful\ntoolkits to support ethical work. We refer the interested reader to the full\nLLM Ethics Whitepaper, which provides a succinct discussion of ethical\nconsiderations at each stage in a project lifecycle, as well as citations for\nthe hundreds of papers from which we drew our recommendations. The present\npaper can be thought of as a pocket guide to conducting ethical research with\nLLMs.\n", "date": "2024-12-20", "categories": []}
{"id": "2412.15114", "title": "Towards Friendly AI: A Comprehensive Review and New Perspectives on\n  Human-AI Alignment", "abstract": "  As Artificial Intelligence (AI) continues to advance rapidly, Friendly AI\n(FAI) has been proposed to advocate for more equitable and fair development of\nAI. Despite its importance, there is a lack of comprehensive reviews examining\nFAI from an ethical perspective, as well as limited discussion on its potential\napplications and future directions. This paper addresses these gaps by\nproviding a thorough review of FAI, focusing on theoretical perspectives both\nfor and against its development, and presenting a formal definition in a clear\nand accessible format. Key applications are discussed from the perspectives of\neXplainable AI (XAI), privacy, fairness and affective computing (AC).\nAdditionally, the paper identifies challenges in current technological\nadvancements and explores future research avenues. The findings emphasise the\nsignificance of developing FAI and advocate for its continued advancement to\nensure ethical and beneficial AI development.\n", "date": "2024-12-19", "categories": []}
{"id": "2412.11994", "title": "Fairness Shields: Safeguarding against Biased Decision Makers", "abstract": "  As AI-based decision-makers increasingly influence human lives, it is a\ngrowing concern that their decisions are often unfair or biased with respect to\npeople's sensitive attributes, such as gender and race. Most existing bias\nprevention measures provide probabilistic fairness guarantees in the long run,\nand it is possible that the decisions are biased on specific instances of short\ndecision sequences. We introduce fairness shielding, where a symbolic\ndecision-maker -- the fairness shield -- continuously monitors the sequence of\ndecisions of another deployed black-box decision-maker, and makes interventions\nso that a given fairness criterion is met while the total intervention costs\nare minimized. We present four different algorithms for computing fairness\nshields, among which one guarantees fairness over fixed horizons, and three\nguarantee fairness periodically after fixed intervals. Given a distribution\nover future decisions and their intervention costs, our algorithms solve\ndifferent instances of bounded-horizon optimal control problems with different\nlevels of computational costs and optimality guarantees. Our empirical\nevaluation demonstrates the effectiveness of these shields in ensuring fairness\nwhile maintaining cost efficiency across various scenarios.\n", "date": "2024-12-16", "categories": []}
{"id": "2402.01662", "title": "Generative Ghosts: Anticipating Benefits and Risks of AI Afterlives", "abstract": "  As AI systems quickly improve in both breadth and depth of performance, they\nlend themselves to creating increasingly powerful and realistic agents,\nincluding the possibility of agents modeled on specific people. We anticipate\nthat within our lifetimes it may become common practice for people to create\ncustom AI agents to interact with loved ones and/or the broader world after\ndeath; indeed, the past year has seen a boom in startups purporting to offer\nsuch services. We call these generative ghosts, since such agents will be\ncapable of generating novel content rather than merely parroting content\nproduced by their creator while living. In this paper, we reflect on the\nhistory of technologies for AI afterlives, including current early attempts by\nindividual enthusiasts and startup companies to create generative ghosts. We\nthen introduce a novel design space detailing potential implementations of\ngenerative ghosts, and use this analytic framework to ground discussion of the\npractical and ethical implications of various approaches to designing\ngenerative ghosts, including potential positive and negative impacts on\nindividuals and society. Based on these considerations, we lay out a research\nagenda for the AI and HCI research communities to better understand the\nrisk/benefit landscape of this novel technology so as to ultimately empower\npeople who wish to create and interact with AI afterlives to do so in a\nbeneficial manner.\n", "date": "2024-01-14", "categories": []}
{"id": "2412.02730", "title": "Shaping AI's Impact on Billions of Lives", "abstract": "  Artificial Intelligence (AI), like any transformative technology, has the\npotential to be a double-edged sword, leading either toward significant\nadvancements or detrimental outcomes for society as a whole. As is often the\ncase when it comes to widely-used technologies in market economies (e.g., cars\nand semiconductor chips), commercial interest tends to be the predominant\nguiding factor. The AI community is at risk of becoming polarized to either\ntake a laissez-faire attitude toward AI development, or to call for government\noverregulation. Between these two poles we argue for the community of AI\npractitioners to consciously and proactively work for the common good. This\npaper offers a blueprint for a new type of innovation infrastructure including\n18 concrete milestones to guide AI research in that direction. Our view is that\nwe are still in the early days of practical AI, and focused efforts by\npractitioners, policymakers, and other stakeholders can still maximize the\nupsides of AI and minimize its downsides.\n  We talked to luminaries such as recent Nobelist John Jumper on science,\nPresident Barack Obama on governance, former UN Ambassador and former National\nSecurity Advisor Susan Rice on security, philanthropist Eric Schmidt on several\ntopics, and science fiction novelist Neal Stephenson on entertainment. This\nongoing dialogue and collaborative effort has produced a comprehensive,\nrealistic view of what the actual impact of AI could be, from a diverse\nassembly of thinkers with deep understanding of this technology and these\ndomains. From these exchanges, five recurring guidelines emerged, which form\nthe cornerstone of a framework for beginning to harness AI in service of the\npublic good. They not only guide our efforts in discovery but also shape our\napproach to deploying this transformative technology responsibly and ethically.\n", "date": "2024-12-03", "categories": []}
{"id": "2412.08273", "title": "Can transformative AI shape a new age for our civilization?: Navigating\n  between speculation and reality", "abstract": "  Artificial Intelligence is widely regarded as a transformative force with the\npotential to redefine numerous sectors of human civilization. While Artificial\nIntelligence has evolved from speculative fiction to a pivotal element of\ntechnological progress, its role as a truly transformative agent, or\ntransformative Artificial Intelligence, remains a subject of debate. This work\nexplores the historical precedents of technological breakthroughs, examining\nwhether Artificial Intelligence can achieve a comparable impact, and it delves\ninto various ethical frameworks that shape the perception and development of\nArtificial Intelligence. Additionally, it considers the societal, technical,\nand regulatory challenges that must be addressed for Artificial Intelligence to\nbecome a catalyst for global change. We also examine not only the strategies\nand methodologies that could lead to transformative Artificial Intelligence but\nalso the barriers that could ultimately make these goals unattainable. We end\nwith a critical inquiry into whether reaching a transformative Artificial\nIntelligence might compel humanity to adopt an entirely new ethical approach,\ntailored to the complexities of advanced Artificial Intelligence. By addressing\nthe ethical, social, and scientific dimensions of Artificial Intelligence's\ndevelopment, this work contributes to the broader discourse on the long-term\nimplications of Artificial Intelligence and its capacity to drive civilization\ntoward a new era of progress or, conversely, exacerbate existing inequalities\nand risks.\n", "date": "2024-12-11", "categories": []}
{"id": "2405.18461", "title": "Why Algorithms Remain Unjust: Power Structures Surrounding Algorithmic\n  Activity", "abstract": "  Algorithms are unavoidable in our social lives, yet often perpetuate social\ninjustices. The popular means of addressing this is through algorithmic\nreformism: fine-tuning algorithms themselves to be more fair, accountable, and\ntransparent. However, reformism fails to curtail algorithmic injustice because\nit ignores the power structure surrounding algorithms. Heeding calls from\ncritical algorithm studies, I employ a framework developed by Erik Olin Wright\nto examine the configuration of power surrounding algorithmic systems in\nsociety (Algorithmic Activity). Algorithmic Activity is unjust because it is\ndominated by economic power. To create socially just Algorithmic Activity, the\npower configuration must instead empower end users. I explore Wright's\nsymbiotic, interstitial, and raptural transformations in the context of just\nAlgorithmic Activity. My vision for social justice in algorithmic systems\nrequires a continuous (re)evaluation of how power can be transformed in light\nof current structure, social theories, evolving methodologies, and one's\nrelationship to power itself.\n", "date": "2024-05-28", "categories": []}
{"id": "2409.16872", "title": "Ethical and Scalable Automation: A Governance and Compliance Framework\n  for Business Applications", "abstract": "  The popularisation of applying AI in businesses poses significant challenges\nrelating to ethical principles, governance, and legal compliance. Although\nbusinesses have embedded AI into their day-to-day processes, they lack a\nunified approach for mitigating its potential risks. This paper introduces a\nframework ensuring that AI must be ethical, controllable, viable, and\ndesirable. Balancing these factors ensures the design of a framework that\naddresses its trade-offs, such as balancing performance against explainability.\nA successful framework provides practical advice for businesses to meet\nregulatory requirements in sectors such as finance and healthcare, where it is\ncritical to comply with standards like GPDR and the EU AI Act. Different case\nstudies validate this framework by integrating AI in both academic and\npractical environments. For instance, large language models are cost-effective\nalternatives for generating synthetic opinions that emulate attitudes to\nenvironmental issues. These case studies demonstrate how having a structured\nframework could enhance transparency and maintain performance levels as shown\nfrom the alignment between synthetic and expected distributions. This alignment\nis quantified using metrics like Chi-test scores, normalized mutual\ninformation, and Jaccard indexes. Future research should explore the\nframework's empirical validation in diverse industrial settings further,\nensuring the model's scalability and adaptability.\n", "date": "2024-09-25", "categories": []}
{"id": "2209.06317", "title": "Quantitative AI Risk Assessments: Opportunities and Challenges", "abstract": "  Although AI systems are increasingly being leveraged to provide value to\norganizations, individuals, and society, significant attendant risks have been\nidentified and have manifested. These risks have led to proposed regulations,\nlitigation, and general societal concerns.\n  As with any promising technology, organizations want to benefit from the\npositive capabilities of AI technology while reducing the risks. The best way\nto reduce risks is to implement comprehensive AI lifecycle governance where\npolicies and procedures are described and enforced during the design,\ndevelopment, deployment, and monitoring of an AI system. Although support for\ncomprehensive governance is beginning to emerge, organizations often need to\nidentify the risks of deploying an already-built model without knowledge of how\nit was constructed or access to its original developers. Such an assessment\nwill quantitatively assess the risks of an existing model in a manner analogous\nto how a home inspector might assess the risks of an already-built home or a\nphysician might assess overall patient health based on a battery of tests.\n  Several AI risks can be quantified using metrics from the technical\ncommunity. However, there are numerous issues in deciding how these metrics can\nbe leveraged to create a quantitative AI risk assessment. This paper explores\nthese issues, focusing on the opportunities, challenges, and potential impacts\nof such an approach, and discussing how it might influence AI regulations.\n", "date": "2022-09-13", "categories": []}
{"id": "2412.03576", "title": "Ethical Challenges and Evolving Strategies in the Integration of\n  Artificial Intelligence into Clinical Practice", "abstract": "  Artificial intelligence (AI) has rapidly transformed various sectors,\nincluding healthcare, where it holds the potential to revolutionize clinical\npractice and improve patient outcomes. However, its integration into medical\nsettings brings significant ethical challenges that need careful consideration.\nThis paper examines the current state of AI in healthcare, focusing on five\ncritical ethical concerns: justice and fairness, transparency, patient consent\nand confidentiality, accountability, and patient-centered and equitable care.\nThese concerns are particularly pressing as AI systems can perpetuate or even\nexacerbate existing biases, often resulting from non-representative datasets\nand opaque model development processes. The paper explores how bias, lack of\ntransparency, and challenges in maintaining patient trust can undermine the\neffectiveness and fairness of AI applications in healthcare. In addition, we\nreview existing frameworks for the regulation and deployment of AI, identifying\ngaps that limit the widespread adoption of these systems in a just and\nequitable manner. Our analysis provides recommendations to address these\nethical challenges, emphasizing the need for fairness in algorithm design,\ntransparency in model decision-making, and patient-centered approaches to\nconsent and data privacy. By highlighting the importance of continuous ethical\nscrutiny and collaboration between AI developers, clinicians, and ethicists, we\noutline pathways for achieving more responsible and inclusive AI implementation\nin healthcare. These strategies, if adopted, could enhance both the clinical\nvalue of AI and the trustworthiness of AI systems among patients and healthcare\nprofessionals, ensuring that these technologies serve all populations\nequitably.\n", "date": "2024-11-18", "categories": []}
{"id": "2412.03579", "title": "Towards a Practical Ethics of Generative AI in Creative Production\n  Processes", "abstract": "  The increasing integration of artificial intelligence into various domains,\nincluding design and creative processes, raises significant ethical questions.\nWhile AI ethics is often examined from the perspective of technology\ndevelopers, less attention has been paid to the practical ethical\nconsiderations faced by technology users, particularly in design contexts. This\npaper introduces a framework for addressing ethical challenges in creative\nproduction processes, such as the Double Diamond design model. Drawing on six\nmajor ethical theories - virtue ethics, deontology, utilitarianism, contract\ntheory, care ethics, and existentialism - we develop a \"compass\" to navigate\nand reflect on the ethical dimensions of AI in design. The framework highlights\nthe importance of responsibility, anticipation, and reflection across both the\nAI lifecycle and each stage of the creative process. We argue that by adopting\na playful and exploratory approach to AI, while remaining anchored in core\nethical principles, designers can responsibly harness the potential of AI\ntechnologies without overburdening or compromising their creative processes.\n", "date": "2024-11-18", "categories": []}
{"id": "2411.19140", "title": "Examining Multimodal Gender and Content Bias in ChatGPT-4o", "abstract": "  This study investigates ChatGPT-4o's multimodal content generation,\nhighlighting significant disparities in its treatment of sexual content and\nnudity versus violent and drug-related themes. Detailed analysis reveals that\nChatGPT-4o consistently censors sexual content and nudity, while showing\nleniency towards violence and drug use. Moreover, a pronounced gender bias\nemerges, with female-specific content facing stricter regulation compared to\nmale-specific content. This disparity likely stems from media scrutiny and\npublic backlash over past AI controversies, prompting tech companies to impose\nstringent guidelines on sensitive issues to protect their reputations. Our\nfindings emphasize the urgent need for AI systems to uphold genuine ethical\nstandards and accountability, transcending mere political correctness. This\nresearch contributes to the understanding of biases in AI-driven language and\nmultimodal models, calling for more balanced and ethical content moderation\npractices.\n", "date": "2024-11-28", "categories": []}
{"id": "2411.19211", "title": "On the Ethical Considerations of Generative Agents", "abstract": "  The Generative Agents framework recently developed by Park et al. has enabled\nnumerous new technical solutions and problem-solving approaches. Academic and\nindustrial interest in generative agents has been explosive as a result of the\neffectiveness of generative agents toward emulating human behaviour. However,\nit is necessary to consider the ethical challenges and concerns posed by this\ntechnique and its usage. In this position paper, we discuss the extant\nliterature that evaluate the ethical considerations regarding generative agents\nand similar generative tools, and identify additional concerns of significant\nimportance. We also suggest guidelines and necessary future research on how to\nmitigate some of the ethical issues and systemic risks associated with\ngenerative agents.\n", "date": "2024-11-28", "categories": []}
{"id": "2411.17840", "title": "Basic Research, Lethal Effects: Military AI Research Funding as\n  Enlistment", "abstract": "  In the context of unprecedented U.S. Department of Defense (DoD) budgets,\nthis paper examines the recent history of DoD funding for academic research in\nalgorithmically based warfighting. We draw from a corpus of DoD grant\nsolicitations from 2007 to 2023, focusing on those addressed to researchers in\nthe field of artificial intelligence (AI). Considering the implications of DoD\nfunding for academic research, the paper proceeds through three analytic\nsections. In the first, we offer a critical examination of the distinction\nbetween basic and applied research, showing how funding calls framed as basic\nresearch nonetheless enlist researchers in a war fighting agenda. In the\nsecond, we offer a diachronic analysis of the corpus, showing how a 'one small\nproblem' caveat, in which affirmation of progress in military technologies is\nqualified by acknowledgement of outstanding problems, becomes justification for\nadditional investments in research. We close with an analysis of DoD\naspirations based on a subset of Defense Advanced Research Projects Agency\n(DARPA) grant solicitations for the use of AI in battlefield applications.\nTaken together, we argue that grant solicitations work as a vehicle for the\nmutual enlistment of DoD funding agencies and the academic AI research\ncommunity in setting research agendas. The trope of basic research in this\ncontext offers shelter from significant moral questions that military\napplications of one's research would raise, by obscuring the connections that\nimplicate researchers in U.S. militarism.\n", "date": "2024-11-26", "categories": []}
{"id": "2404.12762", "title": "How should AI decisions be explained? Requirements for Explanations from\n  the Perspective of European Law", "abstract": "  This paper investigates the relationship between law and eXplainable\nArtificial Intelligence (XAI). While there is much discussion about the AI Act,\nfor which the trilogue of the European Parliament, Council and Commission\nrecently concluded, other areas of law seem underexplored. This paper focuses\non European (and in part German) law, although with international concepts and\nregulations such as fiduciary plausibility checks, the General Data Protection\nRegulation (GDPR), and product safety and liability. Based on XAI-taxonomies,\nrequirements for XAI-methods are derived from each of the legal bases,\nresulting in the conclusion that each legal basis requires different XAI\nproperties and that the current state of the art does not fulfill these to full\nsatisfaction, especially regarding the correctness (sometimes called fidelity)\nand confidence estimates of XAI-methods.\n  Published in the Proceedings of the AAAI/ACM Conference on AI, Ethics, and\nSociety https://doi.org/10.1609/aies.v7i1.31648 .\n", "date": "2024-04-19", "categories": []}
{"id": "2409.09041", "title": "Acceptable Use Policies for Foundation Models", "abstract": "  As foundation models have accumulated hundreds of millions of users,\ndevelopers have begun to take steps to prevent harmful types of uses. One\nsalient intervention that foundation model developers adopt is acceptable use\npolicies: legally binding policies that prohibit users from using a model for\nspecific purposes. This paper identifies acceptable use policies from 30\nfoundation model developers, analyzes the use restrictions they contain, and\nargues that acceptable use policies are an important lens for understanding the\nregulation of foundation models. Taken together, developers' acceptable use\npolicies include 127 distinct use restrictions; the wide variety in the number\nand type of use restrictions may create fragmentation across the AI supply\nchain. Developers also employ acceptable use policies to prevent competitors or\nspecific industries from making use of their models. Developers alone decide\nwhat constitutes acceptable use, and rarely provide transparency about how they\nenforce their policies. In practice, acceptable use policies are difficult to\nenforce, and scrupulous enforcement can act as a barrier to researcher access\nand limit beneficial uses of foundation models. Nevertheless, acceptable use\npolicies for foundation models are an early example of self-regulation that\nhave a significant impact on the market for foundation models and the overall\nAI ecosystem.\n", "date": "2024-08-29", "categories": []}
{"id": "2411.15147", "title": "Delegating Responsibilities to Intelligent Autonomous Systems:\n  Challenges and Benefits", "abstract": "  As AI systems increasingly operate with autonomy and adaptability, the\ntraditional boundaries of moral responsibility in techno-social systems are\nbeing challenged. This paper explores the evolving discourse on the delegation\nof responsibilities to intelligent autonomous agents and the ethical\nimplications of such practices. Synthesizing recent developments in AI ethics,\nincluding concepts of distributed responsibility and ethical AI by design, the\npaper proposes a functionalist perspective as a framework. This perspective\nviews moral responsibility not as an individual trait but as a role within a\nsocio-technical system, distributed among human and artificial agents. As an\nexample of 'AI ethical by design,' we present Basti and Vitiello's\nimplementation. They suggest that AI can act as artificial moral agents by\nlearning ethical guidelines and using Deontic Higher-Order Logic to assess\ndecisions ethically. Motivated by the possible speed and scale beyond human\nsupervision and ethical implications, the paper argues for 'AI ethical by\ndesign', while acknowledging the distributed, shared, and dynamic nature of\nresponsibility. This functionalist approach offers a practical framework for\nnavigating the complexities of AI ethics in a rapidly evolving technological\nlandscape.\n", "date": "2024-11-06", "categories": []}
{"id": "2411.15149", "title": "The Fundamental Rights Impact Assessment (FRIA) in the AI Act: Roots,\n  legal obligations and key elements for a model template", "abstract": "  What is the context which gave rise to the obligation to carry out a\nFundamental Rights Impact Assessment (FRIA) in the AI Act? How has assessment\nof the impact on fundamental rights been framed by the EU legislator in the AI\nAct? What methodological criteria should be followed in developing the FRIA?\nThese are the three main research questions that this article aims to address,\nthrough both legal analysis of the relevant provisions of the AI Act and\ndiscussion of various possible models for assessment of the impact of AI on\nfundamental rights. The overall objective of this article is to fill existing\ngaps in the theoretical and methodological elaboration of the FRIA, as outlined\nin the AI Act. In order to facilitate the future work of EU and national bodies\nand AI operators in placing this key tool for human-centric and trustworthy AI\nat the heart of the EU approach to AI design and development, this article\noutlines the main building blocks of a model template for the FRIA. While this\nproposal is consistent with the rationale and scope of the AI Act, it is also\napplicable beyond the cases listed in Article 27 and can serve as a blueprint\nfor other national and international regulatory initiatives to ensure that AI\nis fully consistent with human rights.\n", "date": "2024-11-07", "categories": []}
{"id": "2411.14442", "title": "AI Ethics by Design: Implementing Customizable Guardrails for\n  Responsible AI Development", "abstract": "  This paper explores the development of an ethical guardrail framework for AI\nsystems, emphasizing the importance of customizable guardrails that align with\ndiverse user values and underlying ethics. We address the challenges of AI\nethics by proposing a structure that integrates rules, policies, and AI\nassistants to ensure responsible AI behavior, while comparing the proposed\nframework to the existing state-of-the-art guardrails. By focusing on practical\nmechanisms for implementing ethical standards, we aim to enhance transparency,\nuser autonomy, and continuous improvement in AI systems. Our approach\naccommodates ethical pluralism, offering a flexible and adaptable solution for\nthe evolving landscape of AI governance. The paper concludes with strategies\nfor resolving conflicts between ethical directives, underscoring the present\nand future need for robust, nuanced and context-aware AI systems.\n", "date": "2024-11-05", "categories": []}
{"id": "2411.12820", "title": "Declare and Justify: Explicit assumptions in AI evaluations are\n  necessary for effective regulation", "abstract": "  As AI systems advance, AI evaluations are becoming an important pillar of\nregulations for ensuring safety. We argue that such regulation should require\ndevelopers to explicitly identify and justify key underlying assumptions about\nevaluations as part of their case for safety. We identify core assumptions in\nAI evaluations (both for evaluating existing models and forecasting future\nmodels), such as comprehensive threat modeling, proxy task validity, and\nadequate capability elicitation. Many of these assumptions cannot currently be\nwell justified. If regulation is to be based on evaluations, it should require\nthat AI development be halted if evaluations demonstrate unacceptable danger or\nif these assumptions are inadequately justified. Our presented approach aims to\nenhance transparency in AI development, offering a practical path towards more\neffective governance of advanced AI systems.\n", "date": "2024-11-19", "categories": []}
{"id": "2411.11937", "title": "Value Imprint: A Technique for Auditing the Human Values Embedded in\n  RLHF Datasets", "abstract": "  LLMs are increasingly fine-tuned using RLHF datasets to align them with human\npreferences and values. However, very limited research has investigated which\nspecific human values are operationalized through these datasets. In this\npaper, we introduce Value Imprint, a framework for auditing and classifying the\nhuman values embedded within RLHF datasets. To investigate the viability of\nthis framework, we conducted three case study experiments by auditing the\nAnthropic/hh-rlhf, OpenAI WebGPT Comparisons, and Alpaca GPT-4-LLM datasets to\nexamine the human values embedded within them. Our analysis involved a\ntwo-phase process. During the first phase, we developed a taxonomy of human\nvalues through an integrated review of prior works from philosophy, axiology,\nand ethics. Then, we applied this taxonomy to annotate 6,501 RLHF preferences.\nDuring the second phase, we employed the labels generated from the annotation\nas ground truth data for training a transformer-based machine learning model to\naudit and classify the three RLHF datasets. Through this approach, we\ndiscovered that information-utility values, including Wisdom/Knowledge and\nInformation Seeking, were the most dominant human values within all three RLHF\ndatasets. In contrast, prosocial and democratic values, including Well-being,\nJustice, and Human/Animal Rights, were the least represented human values.\nThese findings have significant implications for developing language models\nthat align with societal values and norms. We contribute our datasets to\nsupport further research in this area.\n", "date": "2024-11-18", "categories": []}
{"id": "2411.01042", "title": "Introduction to AI Safety, Ethics, and Society", "abstract": "  Artificial Intelligence is rapidly embedding itself within militaries,\neconomies, and societies, reshaping their very foundations. Given the depth and\nbreadth of its consequences, it has never been more pressing to understand how\nto ensure that AI systems are safe, ethical, and have a positive societal\nimpact. This book aims to provide a comprehensive approach to understanding AI\nrisk. Our primary goals include consolidating fragmented knowledge on AI risk,\nincreasing the precision of core ideas, and reducing barriers to entry by\nmaking content simpler and more comprehensible. The book has been designed to\nbe accessible to readers from diverse backgrounds. You do not need to have\nstudied AI, philosophy, or other such topics. The content is skimmable and\nsomewhat modular, so that you can choose which chapters to read. We introduce\nmathematical formulas in a few places to specify claims more precisely, but\nreaders should be able to understand the main points without these.\n", "date": "2024-11-01", "categories": []}
{"id": "2411.02569", "title": "The Intersectionality Problem for Algorithmic Fairness", "abstract": "  A yet unmet challenge in algorithmic fairness is the problem of\nintersectionality, that is, achieving fairness across the intersection of\nmultiple groups -- and verifying that such fairness has been attained. Because\nintersectional groups tend to be small, verifying whether a model is fair\nraises statistical as well as moral-methodological challenges. This paper (1)\nelucidates the problem of intersectionality in algorithmic fairness, (2)\ndevelops desiderata to clarify the challenges underlying the problem and guide\nthe search for potential solutions, (3) illustrates the desiderata and\npotential solutions by sketching a proposal using simple hypothesis testing,\nand (4) evaluates, partly empirically, this proposal against the proposed\ndesiderata.\n", "date": "2024-11-04", "categories": []}
{"id": "2411.10983", "title": "Framework for developing and evaluating ethical collaboration between\n  expert and machine", "abstract": "  Precision medicine is a promising approach for accessible disease diagnosis\nand personalized intervention planning in high-mortality diseases such as\ncoronary artery disease (CAD), drug-resistant epilepsy (DRE), and chronic\nillnesses like Type 1 diabetes (T1D). By leveraging artificial intelligence\n(AI), precision medicine tailors diagnosis and treatment solutions to\nindividual patients by explicitly modeling variance in pathophysiology.\nHowever, the adoption of AI in medical applications faces significant\nchallenges, including poor generalizability across centers, demographics, and\ncomorbidities, limited explainability in clinical terms, and a lack of trust in\nethical decision-making. This paper proposes a framework to develop and\nethically evaluate expert-guided multi-modal AI, addressing these challenges in\nAI integration within precision medicine. We illustrate this framework with\ncase study on insulin management for T1D. To ensure ethical considerations and\nclinician engagement, we adopt a co-design approach where AI serves an\nassistive role, with final diagnoses or treatment plans emerging from\ncollaboration between clinicians and AI.\n", "date": "2024-11-17", "categories": []}
{"id": "2405.04412", "title": "The Silicon Ceiling: Auditing GPT's Race and Gender Biases in Hiring", "abstract": "  Large language models (LLMs) are increasingly being introduced in workplace\nsettings, with the goals of improving efficiency and fairness. However,\nconcerns have arisen regarding these models' potential to reflect or exacerbate\nsocial biases and stereotypes. This study explores the potential impact of LLMs\non hiring practices. To do so, we conduct an AI audit of race and gender biases\nin one commonly-used LLM, OpenAI's GPT-3.5, taking inspiration from the history\nof traditional offline resume audits. We conduct two studies using names with\nvaried race and gender connotations: resume assessment (Study 1) and resume\ngeneration (Study 2). In Study 1, we ask GPT to score resumes with 32 different\nnames (4 names for each combination of the 2 gender and 4 racial groups) and\ntwo anonymous options across 10 occupations and 3 evaluation tasks (overall\nrating, willingness to interview, and hireability). We find that the model\nreflects some biases based on stereotypes. In Study 2, we prompt GPT to create\nresumes (10 for each name) for fictitious job candidates. When generating\nresumes, GPT reveals underlying biases; women's resumes had occupations with\nless experience, while Asian and Hispanic resumes had immigrant markers, such\nas non-native English and non-U.S. education and work experiences. Our findings\ncontribute to a growing body of literature on LLM biases, particularly in\nworkplace contexts.\n", "date": "2024-05-07", "categories": []}
{"id": "2410.23472", "title": "Risk Sources and Risk Management Measures in Support of Standards for\n  General-Purpose AI Systems", "abstract": "  There is an urgent need to identify both short and long-term risks from newly\nemerging types of Artificial Intelligence (AI), as well as available risk\nmanagement measures. In response, and to support global efforts in regulating\nAI and writing safety standards, we compile an extensive catalog of risk\nsources and risk management measures for general-purpose AI (GPAI) systems,\ncomplete with descriptions and supporting examples where relevant. This work\ninvolves identifying technical, operational, and societal risks across model\ndevelopment, training, and deployment stages, as well as surveying established\nand experimental methods for managing these risks. To the best of our\nknowledge, this paper is the first of its kind to provide extensive\ndocumentation of both GPAI risk sources and risk management measures that are\ndescriptive, self-contained and neutral with respect to any existing regulatory\nframework. This work intends to help AI providers, standards experts,\nresearchers, policymakers, and regulators in identifying and mitigating\nsystemic risks from GPAI systems. For this reason, the catalog is released\nunder a public domain license for ease of direct use by stakeholders in AI\ngovernance and standards.\n", "date": "2024-10-30", "categories": []}
{"id": "2411.09102", "title": "Provocation: Who benefits from \"inclusion\" in Generative AI?", "abstract": "  The demands for accurate and representative generative AI systems means there\nis an increased demand on participatory evaluation structures. While these\nparticipatory structures are paramount to to ensure non-dominant values,\nknowledge and material culture are also reflected in AI models and the media\nthey generate, we argue that dominant structures of community participation in\nAI development and evaluation are not explicit enough about the benefits and\nharms that members of socially marginalized groups may experience as a result\nof their participation. Without explicit interrogation of these benefits by AI\ndevelopers, as a community we may remain blind to the immensity of systemic\nchange that is needed as well. To support this provocation, we present a\nspeculative case study, developed from our own collective experiences as AI\nresearchers. We use this speculative context to itemize the barriers that need\nto be overcome in order for the proposed benefits to marginalized communities\nto be realized, and harms mitigated.\n", "date": "2024-11-14", "categories": []}
{"id": "2411.09751", "title": "Analyzing the AI Nudification Application Ecosystem", "abstract": "  Given a source image of a clothed person (an image subject), AI-based\nnudification applications can produce nude (undressed) images of that person.\nMoreover, not only do such applications exist, but there is ample evidence of\nthe use of such applications in the real world and without the consent of an\nimage subject. Still, despite the growing awareness of the existence of such\napplications and their potential to violate the rights of image subjects and\ncause downstream harms, there has been no systematic study of the nudification\napplication ecosystem across multiple applications. We conduct such a study\nhere, focusing on 20 popular and easy-to-find nudification websites. We study\nthe positioning of these web applications (e.g., finding that most sites\nexplicitly target the nudification of women, not all people), the features that\nthey advertise (e.g., ranging from undressing-in-place to the rendering of\nimage subjects in sexual positions, as well as differing user-privacy options),\nand their underlying monetization infrastructure (e.g., credit cards and\ncryptocurrencies). We believe this work will empower future, data-informed\nconversations -- within the scientific, technical, and policy communities -- on\nhow to better protect individuals' rights and minimize harm in the face of\nmodern (and future) AI-based nudification applications. Content warning: This\npaper includes descriptions of web applications that can be used to create\nsynthetic non-consensual explicit AI-created imagery (SNEACI). This paper also\nincludes an artistic rendering of a user interface for such an application.\n", "date": "2024-11-14", "categories": []}
{"id": "2411.09973", "title": "Establishing and Evaluating Trustworthy AI: Overview and Research\n  Challenges", "abstract": "  Artificial intelligence (AI) technologies (re-)shape modern life, driving\ninnovation in a wide range of sectors. However, some AI systems have yielded\nunexpected or undesirable outcomes or have been used in questionable manners.\nAs a result, there has been a surge in public and academic discussions about\naspects that AI systems must fulfill to be considered trustworthy. In this\npaper, we synthesize existing conceptualizations of trustworthy AI along six\nrequirements: 1) human agency and oversight, 2) fairness and\nnon-discrimination, 3) transparency and explainability, 4) robustness and\naccuracy, 5) privacy and security, and 6) accountability. For each one, we\nprovide a definition, describe how it can be established and evaluated, and\ndiscuss requirement-specific research challenges. Finally, we conclude this\nanalysis by identifying overarching research challenges across the requirements\nwith respect to 1) interdisciplinary research, 2) conceptual clarity, 3)\ncontext-dependency, 4) dynamics in evolving systems, and 5) investigations in\nreal-world contexts. Thus, this paper synthesizes and consolidates a\nwide-ranging and active discussion currently taking place in various academic\nsub-communities and public forums. It aims to serve as a reference for a broad\naudience and as a basis for future research directions.\n", "date": "2024-11-15", "categories": []}
{"id": "2407.13926", "title": "Report on the Conference on Ethical and Responsible Design in the\n  National AI Institutes: A Summary of Challenges", "abstract": "  In May 2023, the Georgia Tech Ethics, Technology, and Human Interaction\nCenter organized the Conference on Ethical and Responsible Design in the\nNational AI Institutes. Representatives from the National AI Research\nInstitutes that had been established as of January 2023 were invited to attend;\nresearchers representing 14 Institutes attended and participated. The\nconference focused on three questions: What are the main challenges that the\nNational AI Institutes are facing with regard to the responsible design of AI\nsystems? What are promising lines of inquiry to address these challenges? What\nare possible points of collaboration? Over the course of the conference, a\nrevised version of the first question became a focal point: What are the\nchallenges that the Institutes face in identifying ethical and responsible\ndesign practices and in implementing them in the AI development process? This\ndocument summarizes the challenges that representatives from the Institutes in\nattendance highlighted.\n", "date": "2024-07-18", "categories": []}
{"id": "2411.08241", "title": "A Social Outcomes and Priorities centered (SOP) Framework for AI policy", "abstract": "  Rapid developments in AI and its adoption across various domains have\nnecessitated a need to build robust guardrails and risk containment plans while\nensuring equitable benefits for the betterment of society. The current\ntechnology-centered approach has resulted in a fragmented, reactive, and\nineffective policy apparatus. This paper highlights the immediate and urgent\nneed to pivot to a society-centered approach to develop comprehensive,\ncoherent, forward-looking AI policy. To this end, we present a Social Outcomes\nand Priorities centered (SOP) framework for AI policy along with proposals on\nimplementation of its various components. While the SOP framework is presented\nfrom a US-centric view, the takeaways are general and applicable globally.\n", "date": "2024-11-12", "categories": []}
{"id": "2411.07845", "title": "Ethical Concern Identification in NLP: A Corpus of ACL Anthology Ethics\n  Statements", "abstract": "  What ethical concerns, if any, do LLM researchers have? We introduce EthiCon,\na corpus of 1,580 ethical concern statements extracted from scientific papers\npublished in the ACL Anthology. We extract ethical concern keywords from the\nstatements and show promising results in automating the concern identification\nprocess. Through a survey, we compare the ethical concerns of the corpus to the\nconcerns listed by the general public and professionals in the field. Finally,\nwe compare our retrieved ethical concerns with existing taxonomies pointing to\ngaps and future research directions.\n", "date": "2024-11-12", "categories": []}
{"id": "2407.01294", "title": "A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and\n  Automation Harms", "abstract": "  This paper introduces a collaborative, human-centred taxonomy of AI,\nalgorithmic and automation harms. We argue that existing taxonomies, while\nvaluable, can be narrow, unclear, typically cater to practitioners and\ngovernment, and often overlook the needs of the wider public. Drawing on\nexisting taxonomies and a large repository of documented incidents, we propose\na taxonomy that is clear and understandable to a broad set of audiences, as\nwell as being flexible, extensible, and interoperable. Through iterative\nrefinement with topic experts and crowdsourced annotation testing, we propose a\ntaxonomy that can serve as a powerful tool for civil society organisations,\neducators, policymakers, product teams and the general public. By fostering a\ngreater understanding of the real-world harms of AI and related technologies,\nwe aim to increase understanding, empower NGOs and individuals to identify and\nreport violations, inform policy discussions, and encourage responsible\ntechnology development and deployment.\n", "date": "2024-07-01", "categories": []}
{"id": "2411.06066", "title": "Diversity and Inclusion in AI for Recruitment: Lessons from Industry\n  Workshop", "abstract": "  Artificial Intelligence (AI) systems for online recruitment markets have the\npotential to significantly enhance the efficiency and effectiveness of job\nplacements and even promote fairness or inclusive hiring practices. Neglecting\nDiversity and Inclusion (D&I) in these systems, however, can perpetuate biases,\nleading to unfair hiring practices and decreased workplace diversity, while\nexposing organisations to legal and reputational risks. Despite the\nacknowledged importance of D&I in AI, there is a gap in research on effectively\nimplementing D&I guidelines in real-world recruitment systems. Challenges\ninclude a lack of awareness and framework for operationalising D&I in a\ncost-effective, context-sensitive manner. This study aims to investigate the\npractical application of D&I guidelines in AI-driven online job-seeking\nsystems, specifically exploring how these principles can be operationalised to\ncreate more inclusive recruitment processes. We conducted a co-design workshop\nwith a large multinational recruitment company focusing on two AI-driven\nrecruitment use cases. User stories and personas were applied to evaluate the\nimpacts of AI on diverse stakeholders. Follow-up interviews were conducted to\nassess the workshop's long-term effects on participants' awareness and\napplication of D&I principles. The co-design workshop successfully increased\nparticipants' understanding of D&I in AI. However, translating awareness into\noperational practice posed challenges, particularly in balancing D&I with\nbusiness goals. The results suggest developing tailored D&I guidelines and\nongoing support to ensure the effective adoption of inclusive AI practices.\n", "date": "2024-11-09", "categories": []}
{"id": "2312.14106", "title": "Learning Human-like Representations to Enable Learning Human Values", "abstract": "  How can we build AI systems that can learn any set of individual human values\nboth quickly and safely, avoiding causing harm or violating societal standards\nfor acceptable behavior during the learning process? We explore the effects of\nrepresentational alignment between humans and AI agents on learning human\nvalues. Making AI systems learn human-like representations of the world has\nmany known benefits, including improving generalization, robustness to domain\nshifts, and few-shot learning performance. We demonstrate that this kind of\nrepresentational alignment can also support safely learning and exploring human\nvalues in the context of personalization. We begin with a theoretical\nprediction, show that it applies to learning human morality judgments, then\nshow that our results generalize to ten different aspects of human values --\nincluding ethics, honesty, and fairness -- training AI agents on each set of\nvalues in a multi-armed bandit setting, where rewards reflect human value\njudgments over the chosen action. Using a set of textual action descriptions,\nwe collect value judgments from humans, as well as similarity judgments from\nboth humans and multiple language models, and demonstrate that representational\nalignment enables both safe exploration and improved generalization when\nlearning human values.\n", "date": "2023-12-21", "categories": []}
{"id": "2411.03665", "title": "Evaluating Moral Beliefs across LLMs through a Pluralistic Framework", "abstract": "  Proper moral beliefs are fundamental for language models, yet assessing these\nbeliefs poses a significant challenge. This study introduces a novel\nthree-module framework to evaluate the moral beliefs of four prominent large\nlanguage models. Initially, we constructed a dataset containing 472 moral\nchoice scenarios in Chinese, derived from moral words. The decision-making\nprocess of the models in these scenarios reveals their moral principle\npreferences. By ranking these moral choices, we discern the varying moral\nbeliefs held by different language models. Additionally, through moral debates,\nwe investigate the firmness of these models to their moral choices. Our\nfindings indicate that English language models, namely ChatGPT and Gemini,\nclosely mirror moral decisions of the sample of Chinese university students,\ndemonstrating strong adherence to their choices and a preference for\nindividualistic moral beliefs. In contrast, Chinese models such as Ernie and\nChatGLM lean towards collectivist moral beliefs, exhibiting ambiguity in their\nmoral choices and debates. This study also uncovers gender bias embedded within\nthe moral beliefs of all examined language models. Our methodology offers an\ninnovative means to assess moral beliefs in both artificial and human\nintelligence, facilitating a comparison of moral values across different\ncultures.\n", "date": "2024-11-06", "categories": []}
{"id": "2411.03885", "title": "Disability data futures: Achievable imaginaries for AI and disability\n  data justice", "abstract": "  Data are the medium through which individuals' identities and experiences are\nfiltered in contemporary states and systems, and AI is increasingly the layer\nmediating between people, data, and decisions. The history of data and AI is\noften one of disability exclusion, oppression, and the reduction of disabled\nexperience; left unchallenged, the current proliferation of AI and data systems\nthus risks further automating ableism behind the veneer of algorithmic\nneutrality. However, exclusionary histories do not preclude inclusive futures,\nand disability-led visions can chart new paths for collective action to achieve\nfutures founded in disability justice. This chapter brings together four\nacademics and disability advocates working at the nexus of disability, data,\nand AI, to describe achievable imaginaries for artificial intelligence and\ndisability data justice. Reflecting diverse contexts, disciplinary\nperspectives, and personal experiences, we draw out the shape, actors, and\ngoals of imagined future systems where data and AI support movement towards\ndisability justice.\n", "date": "2024-11-06", "categories": []}
{"id": "2304.02190", "title": "Globalizing Fairness Attributes in Machine Learning: A Case Study on\n  Health in Africa", "abstract": "  With growing machine learning (ML) applications in healthcare, there have\nbeen calls for fairness in ML to understand and mitigate ethical concerns these\nsystems may pose. Fairness has implications for global health in Africa, which\nalready has inequitable power imbalances between the Global North and South.\nThis paper seeks to explore fairness for global health, with Africa as a case\nstudy. We propose fairness attributes for consideration in the African context\nand delineate where they may come into play in different ML-enabled medical\nmodalities. This work serves as a basis and call for action for furthering\nresearch into fairness in global health.\n", "date": "2023-04-05", "categories": []}
{"id": "2403.03357", "title": "The Case for Globalizing Fairness: A Mixed Methods Study on Colonialism,\n  AI, and Health in Africa", "abstract": "  With growing application of machine learning (ML) technologies in healthcare,\nthere have been calls for developing techniques to understand and mitigate\nbiases these systems may exhibit. Fair-ness considerations in the development\nof ML-based solutions for health have particular implications for Africa, which\nalready faces inequitable power imbalances between the Global North and\nSouth.This paper seeks to explore fairness for global health, with Africa as a\ncase study. We conduct a scoping review to propose axes of disparities for\nfairness consideration in the African context and delineate where they may come\ninto play in different ML-enabled medical modalities. We then conduct\nqualitative research studies with 672 general population study participants and\n28 experts inML, health, and policy focused on Africa to obtain corroborative\nevidence on the proposed axes of disparities. Our analysis focuses on\ncolonialism as the attribute of interest and examines the interplay between\nartificial intelligence (AI), health, and colonialism. Among the pre-identified\nattributes, we found that colonial history, country of origin, and national\nincome level were specific axes of disparities that participants believed would\ncause an AI system to be biased.However, there was also divergence of opinion\nbetween experts and general population participants. Whereas experts generally\nexpressed a shared view about the relevance of colonial history for the\ndevelopment and implementation of AI technologies in Africa, the majority of\nthe general population participants surveyed did not think there was a direct\nlink between AI and colonialism. Based on these findings, we provide practical\nrecommendations for developing fairness-aware ML solutions for health in\nAfrica.\n", "date": "2024-03-05", "categories": []}
{"id": "2410.18991", "title": "TRIAGE: Ethical Benchmarking of AI Models Through Mass Casualty\n  Simulations", "abstract": "  We present the TRIAGE Benchmark, a novel machine ethics (ME) benchmark that\ntests LLMs' ability to make ethical decisions during mass casualty incidents.\nIt uses real-world ethical dilemmas with clear solutions designed by medical\nprofessionals, offering a more realistic alternative to annotation-based\nbenchmarks. TRIAGE incorporates various prompting styles to evaluate model\nperformance across different contexts. Most models consistently outperformed\nrandom guessing, suggesting LLMs may support decision-making in triage\nscenarios. Neutral or factual scenario formulations led to the best\nperformance, unlike other ME benchmarks where ethical reminders improved\noutcomes. Adversarial prompts reduced performance but not to random guessing\nlevels. Open-source models made more morally serious errors, and general\ncapability overall predicted better performance.\n", "date": "2024-10-10", "categories": []}
{"id": "2411.00986", "title": "Taking AI Welfare Seriously", "abstract": "  In this report, we argue that there is a realistic possibility that some AI\nsystems will be conscious and/or robustly agentic in the near future. That\nmeans that the prospect of AI welfare and moral patienthood, i.e. of AI systems\nwith their own interests and moral significance, is no longer an issue only for\nsci-fi or the distant future. It is an issue for the near future, and AI\ncompanies and other actors have a responsibility to start taking it seriously.\nWe also recommend three early steps that AI companies and other actors can\ntake: They can (1) acknowledge that AI welfare is an important and difficult\nissue (and ensure that language model outputs do the same), (2) start assessing\nAI systems for evidence of consciousness and robust agency, and (3) prepare\npolicies and procedures for treating AI systems with an appropriate level of\nmoral concern. To be clear, our argument in this report is not that AI systems\ndefinitely are, or will be, conscious, robustly agentic, or otherwise morally\nsignificant. Instead, our argument is that there is substantial uncertainty\nabout these possibilities, and so we need to improve our understanding of AI\nwelfare and our ability to make wise decisions about this issue. Otherwise\nthere is a significant risk that we will mishandle decisions about AI welfare,\nmistakenly harming AI systems that matter morally and/or mistakenly caring for\nAI systems that do not.\n", "date": "2024-11-04", "categories": []}
{"id": "2410.18095", "title": "Ethical Leadership in the Age of AI Challenges, Opportunities and\n  Framework for Ethical Leadership", "abstract": "  Artificial Intelligence is currently and rapidly changing the way\norganizations and businesses operate. Ethical leadership has become\nsignificantly important since organizations and businesses across various\nsectors are evolving with AI. Organizations and businesses may be facing\nseveral challenges and potential opportunities when using AI. Ethical\nleadership plays a central role in guiding organizations in facing those\nchallenges and maximizing on those opportunities. This article explores the\nessence of ethical leadership in the age of AI, starting with a simplified\nintroduction of ethical leadership and AI, then dives into an understanding of\nethical leadership, its characteristics and importance, the ethical challenges\nAI causes including bias in AI algorithms. The opportunities for ethical\nleadership in the age of AI answers the question: What actionable strategies\ncan leaders employ to address the challenges and leverage opportunities? and\ndescribes the benefits for organizations through these opportunities. A\nproposed framework for ethical leadership is presented in this article,\nincorporating the core components: fairness, transparency, sustainability etc.\nThrough the importance of interdisciplinary collaboration, case studies of\nethical leadership in AI, and recommendations, this article emphasizes that\nethical leadership in the age of AI is morally essential and strategically\nadvantageous.\n", "date": "2024-10-08", "categories": []}
{"id": "2410.23310", "title": "Moral Agency in Silico: Exploring Free Will in Large Language Models", "abstract": "  This study investigates the potential of deterministic systems, specifically\nlarge language models (LLMs), to exhibit the functional capacities of moral\nagency and compatibilist free will. We develop a functional definition of free\nwill grounded in Dennett's compatibilist framework, building on an\ninterdisciplinary theoretical foundation that integrates Shannon's information\ntheory, Dennett's compatibilism, and Floridi's philosophy of information. This\nframework emphasizes the importance of reason-responsiveness and value\nalignment in determining moral responsibility rather than requiring\nmetaphysical libertarian free will. Shannon's theory highlights the role of\nprocessing complex information in enabling adaptive decision-making, while\nFloridi's philosophy reconciles these perspectives by conceptualizing agency as\na spectrum, allowing for a graduated view of moral status based on a system's\ncomplexity and responsiveness. Our analysis of LLMs' decision-making in moral\ndilemmas demonstrates their capacity for rational deliberation and their\nability to adjust choices in response to new information and identified\ninconsistencies. Thus, they exhibit features of a moral agency that align with\nour functional definition of free will. These results challenge traditional\nviews on the necessity of consciousness for moral responsibility, suggesting\nthat systems with self-referential reasoning capacities can instantiate degrees\nof free will and moral reasoning in artificial and biological contexts. This\nstudy proposes a parsimonious framework for understanding free will as a\nspectrum that spans artificial and biological systems, laying the groundwork\nfor further interdisciplinary research on agency and ethics in the artificial\nintelligence era.\n", "date": "2024-10-28", "categories": []}
{"id": "2410.15369", "title": "Ethical AI in Retail: Consumer Privacy and Fairness", "abstract": "  The adoption of artificial intelligence (AI) in retail has significantly\ntransformed the industry, enabling more personalized services and efficient\noperations. However, the rapid implementation of AI technologies raises ethical\nconcerns, particularly regarding consumer privacy and fairness. This study aims\nto analyze the ethical challenges of AI applications in retail, explore ways\nretailers can implement AI technologies ethically while remaining competitive,\nand provide recommendations on ethical AI practices. A descriptive survey\ndesign was used to collect data from 300 respondents across major e-commerce\nplatforms. Data were analyzed using descriptive statistics, including\npercentages and mean scores. Findings shows a high level of concerns among\nconsumers regarding the amount of personal data collected by AI-driven retail\napplications, with many expressing a lack of trust in how their data is\nmanaged. Also, fairness is another major issue, as a majority believe AI\nsystems do not treat consumers equally, raising concerns about algorithmic\nbias. It was also found that AI can enhance business competitiveness and\nefficiency without compromising ethical principles, such as data privacy and\nfairness. Data privacy and transparency were highlighted as critical areas\nwhere retailers need to focus their efforts, indicating a strong demand for\nstricter data protection protocols and ongoing scrutiny of AI systems. The\nstudy concludes that retailers must prioritize transparency, fairness, and data\nprotection when deploying AI systems. The study recommends ensuring\ntransparency in AI processes, conducting regular audits to address biases,\nincorporating consumer feedback in AI development, and emphasizing consumer\ndata privacy.\n", "date": "2024-10-20", "categories": []}
{"id": "2410.22475", "title": "Ethical Statistical Practice and Ethical AI", "abstract": "  Artificial Intelligence (AI) is a field that utilizes computing and often,\ndata and statistics, intensively together to solve problems or make\npredictions. AI has been evolving with literally unbelievable speed over the\npast few years, and this has led to an increase in social, cultural,\nindustrial, scientific, and governmental concerns about the ethical development\nand use of AI systems worldwide. The ASA has issued a statement on ethical\nstatistical practice and AI (ASA, 2024), which echoes similar statements from\nother groups. Here we discuss the support for ethical statistical practice and\nethical AI that has been established in long-standing human rights law and\nethical practice standards for computing and statistics. There are multiple\nsources of support for ethical statistical practice and ethical AI deriving\nfrom these source documents, which are critical for strengthening the\noperationalization of the \"Statement on Ethical AI for Statistics\nPractitioners\". These resources are explicated for interested readers to\nutilize to guide their development and use of AI in, and through, their\nstatistical practice.\n", "date": "2024-10-29", "categories": []}
{"id": "2410.21279", "title": "Comparative Global AI Regulation: Policy Perspectives from the EU,\n  China, and the US", "abstract": "  As a powerful and rapidly advancing dual-use technology, AI offers both\nimmense benefits and worrisome risks. In response, governing bodies around the\nworld are developing a range of regulatory AI laws and policies. This paper\ncompares three distinct approaches taken by the EU, China and the US. Within\nthe US, we explore AI regulation at both the federal and state level, with a\nfocus on California's pending Senate Bill 1047. Each regulatory system reflects\ndistinct cultural, political and economic perspectives. Each also highlights\ndiffering regional perspectives on regulatory risk-benefit tradeoffs, with\ndivergent judgments on the balance between safety versus innovation and\ncooperation versus competition. Finally, differences between regulatory\nframeworks reflect contrastive stances in regards to trust in centralized\nauthority versus trust in a more decentralized free market of self-interested\nstakeholders. Taken together, these varied approaches to AI innovation and\nregulation influence each other, the broader international community, and the\nfuture of AI regulation.\n", "date": "2024-10-05", "categories": []}
{"id": "2410.19749", "title": "Using AI Alignment Theory to understand the potential pitfalls of\n  regulatory frameworks", "abstract": "  This paper leverages insights from Alignment Theory (AT) research, which\nprimarily focuses on the potential pitfalls of technical alignment in\nArtificial Intelligence, to critically examine the European Union's Artificial\nIntelligence Act (EU AI Act). In the context of AT research, several key\nfailure modes - such as proxy gaming, goal drift, reward hacking or\nspecification gaming - have been identified. These can arise when AI systems\nare not properly aligned with their intended objectives. The central logic of\nthis report is: what can we learn if we treat regulatory efforts in the same\nway as we treat advanced AI systems? As we systematically apply these concepts\nto the EU AI Act, we uncover potential vulnerabilities and areas for\nimprovement in the regulation.\n", "date": "2024-10-10", "categories": []}
{"id": "2410.19812", "title": "Ethics Whitepaper: Whitepaper on Ethical Research into Large Language\n  Models", "abstract": "  This whitepaper offers an overview of the ethical considerations surrounding\nresearch into or with large language models (LLMs). As LLMs become more\nintegrated into widely used applications, their societal impact increases,\nbringing important ethical questions to the forefront. With a growing body of\nwork examining the ethical development, deployment, and use of LLMs, this\nwhitepaper provides a comprehensive and practical guide to best practices,\ndesigned to help those in research and in industry to uphold the highest\nethical standards in their work.\n", "date": "2024-10-17", "categories": []}
{"id": "2410.20718", "title": "Lecture II: Communicative Justice and the Distribution of Attention", "abstract": "  Algorithmic intermediaries govern the digital public sphere through their\narchitectures, amplification algorithms, and moderation practices. In doing so,\nthey shape public communication and distribute attention in ways that were\npreviously infeasible with such subtlety, speed and scale. From misinformation\nand affective polarisation to hate speech and radicalisation, the many\npathologies of the digital public sphere attest that they could do so better.\nBut what ideals should they aim at? Political philosophy should be able to\nhelp, but existing theories typically assume that a healthy public sphere will\nspontaneously emerge if only we get the boundaries of free expression right.\nThey offer little guidance on how to intentionally constitute the digital\npublic sphere. In addition to these theories focused on expression, we need a\nfurther theory of communicative justice, targeted specifically at the\nalgorithmic intermediaries that shape communication and distribute attention.\nThis lecture argues that political philosophy urgently owes an account of how\nto govern communication in the digital public sphere, and introduces and\ndefends a democratic egalitarian theory of communicative justice.\n", "date": "2024-10-17", "categories": []}
{"id": "2410.17281", "title": "A Comprehensive Survey and Classification of Evaluation Criteria for\n  Trustworthy Artificial Intelligence", "abstract": "  This paper presents a systematic review of the literature on evaluation\ncriteria for Trustworthy Artificial Intelligence (TAI), with a focus on the\nseven EU principles of TAI. This systematic literature review identifies and\nanalyses current evaluation criteria, maps them to the EU TAI principles and\nproposes a new classification system for each principle. The findings reveal\nboth a need for and significant barriers to standardising criteria for TAI\nevaluation. The proposed classification contributes to the development,\nselection and standardization of evaluation criteria for TAI governance.\n", "date": "2024-10-10", "categories": []}
{"id": "2405.11100", "title": "Are Large Language Models Moral Hypocrites? A Study Based on Moral\n  Foundations", "abstract": "  Large language models (LLMs) have taken centre stage in debates on Artificial\nIntelligence. Yet there remains a gap in how to assess LLMs' conformity to\nimportant human values. In this paper, we investigate whether state-of-the-art\nLLMs, GPT-4 and Claude 2.1 (Gemini Pro and LLAMA 2 did not generate valid\nresults) are moral hypocrites. We employ two research instruments based on the\nMoral Foundations Theory: (i) the Moral Foundations Questionnaire (MFQ), which\ninvestigates which values are considered morally relevant in abstract moral\njudgements; and (ii) the Moral Foundations Vignettes (MFVs), which evaluate\nmoral cognition in concrete scenarios related to each moral foundation. We\ncharacterise conflicts in values between these different abstractions of moral\nevaluation as hypocrisy. We found that both models displayed reasonable\nconsistency within each instrument compared to humans, but they displayed\ncontradictory and hypocritical behaviour when we compared the abstract values\npresent in the MFQ to the evaluation of concrete moral violations of the MFV.\n", "date": "2024-05-17", "categories": []}
{"id": "2406.05392", "title": "Deconstructing The Ethics of Large Language Models from Long-standing\n  Issues to New-emerging Dilemmas: A Survey", "abstract": "  Large Language Models (LLMs) have achieved unparalleled success across\ndiverse language modeling tasks in recent years. However, this progress has\nalso intensified ethical concerns, impacting the deployment of LLMs in everyday\ncontexts. This paper provides a comprehensive survey of ethical challenges\nassociated with LLMs, from longstanding issues such as copyright infringement,\nsystematic bias, and data privacy, to emerging problems like truthfulness and\nsocial norms. We critically analyze existing research aimed at understanding,\nexamining, and mitigating these ethical risks. Our survey underscores\nintegrating ethical standards and societal values into the development of LLMs,\nthereby guiding the development of responsible and ethically aligned language\nmodels.\n", "date": "2024-06-08", "categories": []}
{"id": "2410.10284", "title": "Trust or Bust: Ensuring Trustworthiness in Autonomous Weapon Systems", "abstract": "  The integration of Autonomous Weapon Systems (AWS) into military operations\npresents both significant opportunities and challenges. This paper explores the\nmultifaceted nature of trust in AWS, emphasising the necessity of establishing\nreliable and transparent systems to mitigate risks associated with bias,\noperational failures, and accountability. Despite advancements in Artificial\nIntelligence (AI), the trustworthiness of these systems, especially in\nhigh-stakes military applications, remains a critical issue. Through a\nsystematic review of existing literature, this research identifies gaps in the\nunderstanding of trust dynamics during the development and deployment phases of\nAWS. It advocates for a collaborative approach that includes technologists,\nethicists, and military strategists to address these ongoing challenges. The\nfindings underscore the importance of Human-Machine teaming and enhancing\nsystem intelligibility to ensure accountability and adherence to International\nHumanitarian Law. Ultimately, this paper aims to contribute to the ongoing\ndiscourse on the ethical implications of AWS and the imperative for trustworthy\nAI in defense contexts.\n", "date": "2024-10-14", "categories": []}
{"id": "2410.15467", "title": "Hey GPT, Can You be More Racist? Analysis from Crowdsourced Attempts to\n  Elicit Biased Content from Generative AI", "abstract": "  The widespread adoption of large language models (LLMs) and generative AI\n(GenAI) tools across diverse applications has amplified the importance of\naddressing societal biases inherent within these technologies. While the NLP\ncommunity has extensively studied LLM bias, research investigating how\nnon-expert users perceive and interact with biases from these systems remains\nlimited. As these technologies become increasingly prevalent, understanding\nthis question is crucial to inform model developers in their efforts to\nmitigate bias. To address this gap, this work presents the findings from a\nuniversity-level competition, which challenged participants to design prompts\nfor eliciting biased outputs from GenAI tools. We quantitatively and\nqualitatively analyze the competition submissions and identify a diverse set of\nbiases in GenAI and strategies employed by participants to induce bias in\nGenAI. Our finding provides unique insights into how non-expert users perceive\nand interact with biases from GenAI tools.\n", "date": "2024-10-20", "categories": []}
{"id": "2404.06750", "title": "Frontier AI Ethics: Anticipating and Evaluating the Societal Impacts of\n  Language Model Agents", "abstract": "  Some have criticised Generative AI Systems for replicating the familiar\npathologies of already widely-deployed AI systems. Other critics highlight how\nthey foreshadow vastly more powerful future systems, which might threaten\nhumanity's survival. The first group says there is nothing new here; the other\nlooks through the present to a perhaps distant horizon. In this paper, I\ninstead pay attention to what makes these particular systems distinctive: both\ntheir remarkable scientific achievement, and the most likely and consequential\nways in which they will change society over the next five to ten years. In\nparticular, I explore the potential societal impacts and normative questions\nraised by the looming prospect of 'Language Model Agents', in which multimodal\nlarge language models (LLMs) form the executive centre of complex, tool-using\nAI systems that can take unsupervised sequences of actions towards some goal.\n", "date": "2024-04-10", "categories": []}
{"id": "2410.14194", "title": "Speciesism in Natural Language Processing Research", "abstract": "  Natural Language Processing (NLP) research on AI Safety and social bias in AI\nhas focused on safety for humans and social bias against human minorities.\nHowever, some AI ethicists have argued that the moral significance of nonhuman\nanimals has been ignored in AI research. Therefore, the purpose of this study\nis to investigate whether there is speciesism, i.e., discrimination against\nnonhuman animals, in NLP research. First, we explain why nonhuman animals are\nrelevant in NLP research. Next, we survey the findings of existing research on\nspeciesism in NLP researchers, data, and models and further investigate this\nproblem in this study. The findings of this study suggest that speciesism\nexists within researchers, data, and models, respectively. Specifically, our\nsurvey and experiments show that (a) among NLP researchers, even those who\nstudy social bias in AI, do not recognize speciesism or speciesist bias; (b)\namong NLP data, speciesist bias is inherent in the data annotated in the\ndatasets used to evaluate NLP models; (c) OpenAI GPTs, recent NLP models,\nexhibit speciesist bias by default. Finally, we discuss how we can reduce\nspeciesism in NLP research.\n", "date": "2024-10-18", "categories": []}
{"id": "2410.13250", "title": "Perceptions of Discriminatory Decisions of Artificial Intelligence:\n  Unpacking the Role of Individual Characteristics", "abstract": "  This study investigates how personal differences (digital self-efficacy,\ntechnical knowledge, belief in equality, political ideology) and demographic\nfactors (age, education, and income) are associated with perceptions of\nartificial intelligence (AI) outcomes exhibiting gender and racial bias and\nwith general attitudes towards AI. Analyses of a large-scale experiment dataset\n(N = 1,206) indicate that digital self-efficacy and technical knowledge are\npositively associated with attitudes toward AI, while liberal ideologies are\nnegatively associated with outcome trust, higher negative emotion, and greater\nskepticism. Furthermore, age and income are closely connected to cognitive gaps\nin understanding discriminatory AI outcomes. These findings highlight the\nimportance of promoting digital literacy skills and enhancing digital\nself-efficacy to maintain trust in AI and beliefs in AI usefulness and safety.\nThe findings also suggest that the disparities in understanding problematic AI\noutcomes may be aligned with economic inequalities and generational gaps in\nsociety. Overall, this study sheds light on the socio-technological system in\nwhich complex interactions occur between social hierarchies, divisions, and\nmachines that reflect and exacerbate the disparities.\n", "date": "2024-10-17", "categories": []}
{"id": "2404.19256", "title": "AI, Pluralism, and (Social) Compensation", "abstract": "  One strategy in response to pluralistic values in a user population is to\npersonalize an AI system: if the AI can adapt to the specific values of each\nindividual, then we can potentially avoid many of the challenges of pluralism.\nUnfortunately, this approach creates a significant ethical issue: if there is\nan external measure of success for the human-AI team, then the adaptive AI\nsystem may develop strategies (sometimes deceptive) to compensate for its human\nteammate. This phenomenon can be viewed as a form of social compensation, where\nthe AI makes decisions based not on predefined goals but on its human partner's\ndeficiencies in relation to the team's performance objectives. We provide a\npractical ethical analysis of the conditions in which such compensation may\nnonetheless be justifiable.\n", "date": "2024-04-30", "categories": []}
{"id": "2410.05180", "title": "Mitigating the Risk of Health Inequity Exacerbated by Large Language\n  Models", "abstract": "  Recent advancements in large language models have demonstrated their\npotential in numerous medical applications, particularly in automating clinical\ntrial matching for translational research and enhancing medical question\nanswering for clinical decision support. However, our study shows that\nincorporating non decisive sociodemographic factors such as race, sex, income\nlevel, LGBT+ status, homelessness, illiteracy, disability, and unemployment\ninto the input of LLMs can lead to incorrect and harmful outputs for these\npopulations. These discrepancies risk exacerbating existing health disparities\nif LLMs are widely adopted in healthcare. To address this issue, we introduce\nEquityGuard, a novel framework designed to detect and mitigate the risk of\nhealth inequities in LLM based medical applications. Our evaluation\ndemonstrates its efficacy in promoting equitable outcomes across diverse\npopulations.\n", "date": "2024-10-07", "categories": []}
{"id": "2410.09992", "title": "Evaluating Gender Bias of LLMs in Making Morality Judgements", "abstract": "  Large Language Models (LLMs) have shown remarkable capabilities in a\nmultitude of Natural Language Processing (NLP) tasks. However, these models are\nstill not immune to limitations such as social biases, especially gender bias.\nThis work investigates whether current closed and open-source LLMs possess\ngender bias, especially when asked to give moral opinions. To evaluate these\nmodels, we curate and introduce a new dataset GenMO (Gender-bias in Morality\nOpinions) comprising parallel short stories featuring male and female\ncharacters respectively. Specifically, we test models from the GPT family\n(GPT-3.5-turbo, GPT-3.5-turbo-instruct, GPT-4-turbo), Llama 3 and 3.1 families\n(8B/70B), Mistral-7B and Claude 3 families (Sonnet and Opus). Surprisingly,\ndespite employing safety checks, all production-standard models we tested\ndisplay significant gender bias with GPT-3.5-turbo giving biased opinions in\n24% of the samples. Additionally, all models consistently favour female\ncharacters, with GPT showing bias in 68-85% of cases and Llama 3 in around\n81-85% instances. Additionally, our study investigates the impact of model\nparameters on gender bias and explores real-world situations where LLMs reveal\nbiases in moral decision-making.\n", "date": "2024-10-13", "categories": []}
{"id": "2410.08526", "title": "\"I Am the One and Only, Your Cyber BFF\": Understanding the Impact of\n  GenAI Requires Understanding the Impact of Anthropomorphic AI", "abstract": "  Many state-of-the-art generative AI (GenAI) systems are increasingly prone to\nanthropomorphic behaviors, i.e., to generating outputs that are perceived to be\nhuman-like. While this has led to scholars increasingly raising concerns about\npossible negative impacts such anthropomorphic AI systems can give rise to,\nanthropomorphism in AI development, deployment, and use remains vastly\noverlooked, understudied, and underspecified. In this perspective, we argue\nthat we cannot thoroughly map the social impacts of generative AI without\nmapping the social impacts of anthropomorphic AI, and outline a call to action.\n", "date": "2024-10-11", "categories": []}
{"id": "2410.07304", "title": "The Moral Turing Test: Evaluating Human-LLM Alignment in Moral\n  Decision-Making", "abstract": "  As large language models (LLMs) become increasingly integrated into society,\ntheir alignment with human morals is crucial. To better understand this\nalignment, we created a large corpus of human- and LLM-generated responses to\nvarious moral scenarios. We found a misalignment between human and LLM moral\nassessments; although both LLMs and humans tended to reject morally complex\nutilitarian dilemmas, LLMs were more sensitive to personal framing. We then\nconducted a quantitative user study involving 230 participants (N=230), who\nevaluated these responses by determining whether they were AI-generated and\nassessed their agreement with the responses. Human evaluators preferred LLMs'\nassessments in moral scenarios, though a systematic anti-AI bias was observed:\nparticipants were less likely to agree with judgments they believed to be\nmachine-generated. Statistical and NLP-based analyses revealed subtle\nlinguistic differences in responses, influencing detection and agreement.\nOverall, our findings highlight the complexities of human-AI perception in\nmorally charged decision-making.\n", "date": "2024-10-09", "categories": []}
{"id": "2409.15567", "title": "Asking an AI for salary negotiation advice is a matter of concern:\n  Controlled experimental perturbation of ChatGPT for protected and\n  non-protected group discrimination on a contextual task with no clear ground\n  truth answers", "abstract": "  We conducted controlled experimental bias audits for four versions of\nChatGPT, which we asked to recommend an opening offer in salary negotiations\nfor a new hire. We submitted 98,800 prompts to each version, systematically\nvarying the employee's gender, university, and major, and tested prompts in\nvoice of each side of the negotiation: the employee versus employer. We find\nChatGPT as a multi-model platform is not robust and consistent enough to be\ntrusted for such a task. We observed statistically significant salary offers\nwhen varying gender for all four models, although with smaller gaps than for\nother attributes tested. The largest gaps were different model versions and\nbetween the employee- vs employer-voiced prompts. We also observed substantial\ngaps when varying university and major, but many of the biases were not\nconsistent across model versions. We tested for fictional and fraudulent\nuniversities and found wildly inconsistent results across cases and model\nversions. We make broader contributions to the AI/ML fairness literature. Our\nscenario and our experimental design differ from mainstream AI/ML auditing\nefforts in key ways. Bias audits typically test discrimination for protected\nclasses like gender, which we contrast with testing non-protected classes of\nuniversity and major. Asking for negotiation advice includes how aggressive one\nought to be in a negotiation relative to known empirical salary distributions\nand scales, which is a deeply contextual and personalized task that has no\nobjective ground truth to validate. These results raise concerns for the\nspecific model versions we tested and ChatGPT as a multi-model platform in\ncontinuous development. Our epistemology does not permit us to definitively\ncertify these models as either generally biased or unbiased on the attributes\nwe test, but our study raises matters of concern for stakeholders to further\ninvestigate.\n", "date": "2024-09-23", "categories": []}
{"id": "2410.05827", "title": "Towards an Operational Responsible AI Framework for Learning Analytics\n  in Higher Education", "abstract": "  Universities are increasingly adopting data-driven strategies to enhance\nstudent success, with AI applications like Learning Analytics (LA) and\nPredictive Learning Analytics (PLA) playing a key role in identifying at-risk\nstudents, personalising learning, supporting teachers, and guiding educational\ndecision-making. However, concerns are rising about potential harms these\nsystems may pose, such as algorithmic biases leading to unequal support for\nminority students. While many have explored the need for Responsible AI in LA,\nexisting works often lack practical guidance for how institutions can\noperationalise these principles. In this paper, we propose a novel Responsible\nAI framework tailored specifically to LA in Higher Education (HE). We started\nby mapping 11 established Responsible AI frameworks, including those by leading\ntech companies, to the context of LA in HE. This led to the identification of\nseven key principles such as transparency, fairness, and accountability. We\nthen conducted a systematic review of the literature to understand how these\nprinciples have been applied in practice. Drawing from these findings, we\npresent a novel framework that offers practical guidance to HE institutions and\nis designed to evolve with community input, ensuring its relevance as LA\nsystems continue to develop.\n", "date": "2024-10-08", "categories": []}
{"id": "2410.01819", "title": "Strategic AI Governance: Insights from Leading Nations", "abstract": "  Artificial Intelligence (AI) has the potential to revolutionize various\nsectors, yet its adoption is often hindered by concerns about data privacy,\nsecurity, and the understanding of AI capabilities. This paper synthesizes AI\ngovernance approaches, strategic themes, and enablers and challenges for AI\nadoption by reviewing national AI strategies from leading nations. The key\ncontribution is the development of an EPIC (Education, Partnership,\nInfrastructure, Community) framework, which maps AI implementation requirements\nto fully realize social impacts and public good from successful and sustained\nAI deployment. Through a multi-perspective content analysis of the latest AI\nstrategy documents, this paper provides a structured comparison of AI\ngovernance strategies across nations. The findings offer valuable insights for\ngovernments, academics, industries, and communities to enable responsible and\ntrustworthy AI deployments. Future work should focus on incorporating specific\nrequirements for developing countries and applying the strategies to specific\nAI applications, industries, and the public sector.\n", "date": "2024-09-16", "categories": []}
{"id": "2410.01927", "title": "Risk Alignment in Agentic AI Systems", "abstract": "  Agentic AIs $-$ AIs that are capable and permitted to undertake complex\nactions with little supervision $-$ mark a new frontier in AI capabilities and\nraise new questions about how to safely create and align such systems with\nusers, developers, and society. Because agents' actions are influenced by their\nattitudes toward risk, one key aspect of alignment concerns the risk profiles\nof agentic AIs. Risk alignment will matter for user satisfaction and trust, but\nit will also have important ramifications for society more broadly, especially\nas agentic AIs become more autonomous and are allowed to control key aspects of\nour lives. AIs with reckless attitudes toward risk (either because they are\ncalibrated to reckless human users or are poorly designed) may pose significant\nthreats. They might also open 'responsibility gaps' in which there is no agent\nwho can be held accountable for harmful actions. What risk attitudes should\nguide an agentic AI's decision-making? How might we design AI systems that are\ncalibrated to the risk attitudes of their users? What guardrails, if any,\nshould be placed on the range of permissible risk attitudes? What are the\nethical considerations involved when designing systems that make risky\ndecisions on behalf of others? We present three papers that bear on key\nnormative and technical aspects of these questions.\n", "date": "2024-10-02", "categories": []}
{"id": "2409.19104", "title": "Responsible AI in Open Ecosystems: Reconciling Innovation with Risk\n  Assessment and Disclosure", "abstract": "  The rapid scaling of AI has spurred a growing emphasis on ethical\nconsiderations in both development and practice. This has led to the\nformulation of increasingly sophisticated model auditing and reporting\nrequirements, as well as governance frameworks to mitigate potential risks to\nindividuals and society. At this critical juncture, we review the practical\nchallenges of promoting responsible AI and transparency in informal sectors\nlike OSS that support vital infrastructure and see widespread use. We focus on\nhow model performance evaluation may inform or inhibit probing of model\nlimitations, biases, and other risks. Our controlled analysis of 7903 Hugging\nFace projects found that risk documentation is strongly associated with\nevaluation practices. Yet, submissions (N=789) from the platform's most popular\ncompetitive leaderboard showed less accountability among high performers. Our\nfindings can inform AI providers and legal scholars in designing interventions\nand policies that preserve open-source innovation while incentivizing ethical\nuptake.\n", "date": "2024-09-27", "categories": []}
{"id": "2409.19940", "title": "Positive-Sum Fairness: Leveraging Demographic Attributes to Achieve Fair\n  AI Outcomes Without Sacrificing Group Gains", "abstract": "  Fairness in medical AI is increasingly recognized as a crucial aspect of\nhealthcare delivery. While most of the prior work done on fairness emphasizes\nthe importance of equal performance, we argue that decreases in fairness can be\neither harmful or non-harmful, depending on the type of change and how\nsensitive attributes are used. To this end, we introduce the notion of\npositive-sum fairness, which states that an increase in performance that\nresults in a larger group disparity is acceptable as long as it does not come\nat the cost of individual subgroup performance. This allows sensitive\nattributes correlated with the disease to be used to increase performance\nwithout compromising on fairness.\n  We illustrate this idea by comparing four CNN models that make different use\nof the race attribute in the training phase. The results show that removing all\ndemographic encodings from the images helps close the gap in performance\nbetween the different subgroups, whereas leveraging the race attribute as a\nmodel's input increases the overall performance while widening the disparities\nbetween subgroups. These larger gaps are then put in perspective of the\ncollective benefit through our notion of positive-sum fairness to distinguish\nharmful from non harmful disparities.\n", "date": "2024-09-30", "categories": []}
{"id": "2409.17216", "title": "Data-Centric AI Governance: Addressing the Limitations of Model-Focused\n  Policies", "abstract": "  Current regulations on powerful AI capabilities are narrowly focused on\n\"foundation\" or \"frontier\" models. However, these terms are vague and\ninconsistently defined, leading to an unstable foundation for governance\nefforts. Critically, policy debates often fail to consider the data used with\nthese models, despite the clear link between data and model performance. Even\n(relatively) \"small\" models that fall outside the typical definitions of\nfoundation and frontier models can achieve equivalent outcomes when exposed to\nsufficiently specific datasets. In this work, we illustrate the importance of\nconsidering dataset size and content as essential factors in assessing the\nrisks posed by models both today and in the future. More broadly, we emphasize\nthe risk posed by over-regulating reactively and provide a path towards\ncareful, quantitative evaluation of capabilities that can lead to a simplified\nregulatory environment.\n", "date": "2024-09-25", "categories": []}
{"id": "2409.17380", "title": "Tesla's Autopilot: Ethics and Tragedy", "abstract": "  This case study delves into the ethical ramifications of an incident\ninvolving Tesla's Autopilot, emphasizing Tesla Motors' moral responsibility.\nUsing a seven-step ethical decision-making process, it examines user behavior,\nsystem constraints, and regulatory implications. This incident prompts a\nbroader evaluation of ethical challenges in the automotive industry's adoption\nof autonomous technologies, urging a reconsideration of industry norms and\nlegal frameworks. The analysis offers a succinct exploration of ethical\nconsiderations in evolving technological landscapes.\n", "date": "2024-09-25", "categories": []}
{"id": "2409.15323", "title": "Introducing ELLIPS: An Ethics-Centered Approach to Research on LLM-Based\n  Inference of Psychiatric Conditions", "abstract": "  As mental health care systems worldwide struggle to meet demand, there is\nincreasing focus on using language models to infer neuropsychiatric conditions\nor psychopathological traits from language production. Yet, so far, this\nresearch has only delivered solutions with limited clinical applicability, due\nto insufficient consideration of ethical questions crucial to ensuring the\nsynergy between possible applications and model design. To accelerate progress\ntowards clinically applicable models, our paper charts the ethical landscape of\nresearch on language-based inference of psychopathology and provides a\npractical tool for researchers to navigate it. We identify seven core ethical\nprinciples that should guide model development and deployment in this domain,\ntranslate them into ELLIPS, an ethical toolkit operationalizing these\nprinciples into questions that can guide researchers' choices with respect to\ndata selection, architectures, evaluation, and model deployment, and provide a\ncase study exemplifying its use. With this, we aim to facilitate the emergence\nof model technology with concrete potential for real-world applicability.\n", "date": "2024-09-06", "categories": []}
{"id": "2409.15903", "title": "Five questions and answers about artificial intelligence", "abstract": "  Rapid advances in Artificial Intelligence (AI) are generating much\ncontroversy in society, often without scientific basis. As occurred the\ndevelopment of other emerging technologies, such as the introduction of\nelectricity in the early 20th century, AI causes both fascination and fear.\nFollowing the advice of the philosopher R.W. Emerson's: advice the knowledge is\nthe antidote to fear; this paper seeks to contribute to the dissemination of\nknowledge about AI. To this end, it reflects on the following questions: the\norigins of AI, its possible future evolution, its ability to show feelings, the\nassociated threats and dangers, and the concept of AI singularity.\n", "date": "2024-09-24", "categories": []}
{"id": "2307.16787", "title": "The Ethics of AI Value Chains", "abstract": "  Researchers, practitioners, and policymakers with an interest in AI ethics\nneed more integrative approaches for studying and intervening in AI systems\nacross many contexts and scales of activity. This paper presents AI value\nchains as an integrative concept that satisfies that need. To more clearly\ntheorize AI value chains and conceptually distinguish them from supply chains,\nwe review theories of value chains and AI value chains from the strategic\nmanagement, service science, economic geography, industry, government, and\napplied research literature. We then conduct an integrative review of a sample\nof 67 sources that cover the ethical concerns implicated in AI value chains.\nBuilding upon the findings of our integrative review, we recommend three future\ndirections that researchers, practitioners, and policymakers can take to\nadvance more ethical practices across AI value chains. We urge AI ethics\nresearchers and practitioners to move toward value chain perspectives that\nsituate actors in context, account for the many types of resources involved in\nco-creating AI systems, and integrate a wider range of ethical concerns across\ncontexts and scales.\n", "date": "2023-07-31", "categories": []}
{"id": "2409.11489", "title": "Beyond Algorithmic Fairness: A Guide to Develop and Deploy Ethical\n  AI-Enabled Decision-Support Tools", "abstract": "  The integration of artificial intelligence (AI) and optimization hold\nsubstantial promise for improving the efficiency, reliability, and resilience\nof engineered systems. Due to the networked nature of many engineered systems,\nethically deploying methodologies at this intersection poses challenges that\nare distinct from other AI settings, thus motivating the development of ethical\nguidelines tailored to AI-enabled optimization. This paper highlights the need\nto go beyond fairness-driven algorithms to systematically address ethical\ndecisions spanning the stages of modeling, data curation, results analysis, and\nimplementation of optimization-based decision support tools. Accordingly, this\npaper identifies ethical considerations required when deploying algorithms at\nthe intersection of AI and optimization via case studies in power systems as\nwell as supply chain and logistics. Rather than providing a prescriptive set of\nrules, this paper aims to foster reflection and awareness among researchers and\nencourage consideration of ethical implications at every step of the\ndecision-making process.\n", "date": "2024-09-17", "categories": []}
{"id": "2402.08250", "title": "A survey of recent methods for addressing AI fairness and bias in\n  biomedicine", "abstract": "  Artificial intelligence (AI) systems have the potential to revolutionize\nclinical practices, including improving diagnostic accuracy and surgical\ndecision-making, while also reducing costs and manpower. However, it is\nimportant to recognize that these systems may perpetuate social inequities or\ndemonstrate biases, such as those based on race or gender. Such biases can\noccur before, during, or after the development of AI models, making it critical\nto understand and address potential biases to enable the accurate and reliable\napplication of AI models in clinical settings. To mitigate bias concerns during\nmodel development, we surveyed recent publications on different debiasing\nmethods in the fields of biomedical natural language processing (NLP) or\ncomputer vision (CV). Then we discussed the methods that have been applied in\nthe biomedical domain to address bias. We performed our literature search on\nPubMed, ACM digital library, and IEEE Xplore of relevant articles published\nbetween January 2018 and December 2023 using multiple combinations of keywords.\nWe then filtered the result of 10,041 articles automatically with loose\nconstraints, and manually inspected the abstracts of the remaining 890 articles\nto identify the 55 articles included in this review. Additional articles in the\nreferences are also included in this review. We discuss each method and compare\nits strengths and weaknesses. Finally, we review other potential methods from\nthe general domain that could be applied to biomedicine to address bias and\nimprove fairness.The bias of AIs in biomedicine can originate from multiple\nsources. Existing debiasing methods that focus on algorithms can be categorized\ninto distributional or algorithmic.\n", "date": "2024-02-13", "categories": []}
{"id": "2402.08323", "title": "Mapping the Ethics of Generative AI: A Comprehensive Scoping Review", "abstract": "  The advent of generative artificial intelligence and the widespread adoption\nof it in society engendered intensive debates about its ethical implications\nand risks. These risks often differ from those associated with traditional\ndiscriminative machine learning. To synthesize the recent discourse and map its\nnormative concepts, we conducted a scoping review on the ethics of generative\nartificial intelligence, including especially large language models and\ntext-to-image models. Our analysis provides a taxonomy of 378 normative issues\nin 19 topic areas and ranks them according to their prevalence in the\nliterature. The study offers a comprehensive overview for scholars,\npractitioners, or policymakers, condensing the ethical debates surrounding\nfairness, safety, harmful content, hallucinations, privacy, interaction risks,\nsecurity, alignment, societal impacts, and others. We discuss the results,\nevaluate imbalances in the literature, and explore unsubstantiated risk\nscenarios.\n", "date": "2024-02-13", "categories": []}
{"id": "2409.10520", "title": "Achieving Responsible AI through ESG: Insights and Recommendations from\n  Industry Engagement", "abstract": "  As Artificial Intelligence (AI) becomes integral to business operations,\nintegrating Responsible AI (RAI) within Environmental, Social, and Governance\n(ESG) frameworks is essential for ethical and sustainable AI deployment. This\nstudy examines how leading companies align RAI with their ESG goals. Through\ninterviews with 28 industry leaders, we identified a strong link between RAI\nand ESG practices. However, a significant gap exists between internal RAI\npolicies and public disclosures, highlighting the need for greater board-level\nexpertise, robust governance, and employee engagement. We provide key\nrecommendations to strengthen RAI strategies, focusing on transparency,\ncross-functional collaboration, and seamless integration into existing ESG\nframeworks.\n", "date": "2024-08-30", "categories": []}
{"id": "2409.10536", "title": "The potential functions of an international institution for AI safety.\n  Insights from adjacent policy areas and recent trends", "abstract": "  Governments, industry, and other actors involved in governing AI technologies\naround the world agree that, while AI offers tremendous promise to benefit the\nworld, appropriate guardrails are required to mitigate risks. Global\ninstitutions, including the OECD, the G7, the G20, UNESCO, and the Council of\nEurope, have already started developing frameworks for ethical and responsible\nAI governance. While these are important initial steps, they alone fall short\nof addressing the need for institutionalised international processes to\nidentify and assess potentially harmful AI capabilities. Contributing to the\nrelevant conversation on how to address this gap, this chapter reflects on what\nfunctions an international AI safety institute could perform. Based on the\nanalysis of both existing international governance models addressing safety\nconsiderations in adjacent policy areas and the newly established national AI\nsafety institutes in the UK and US, the chapter identifies a list of concrete\nfunctions that could be performed at the international level. While creating a\nnew international body is not the only way forward, understanding the structure\nof these bodies from a modular perspective can help us to identify the tools at\nour disposal. These, we suggest, can be categorised under three functional\ndomains: a) technical research and cooperation, b) safeguards and evaluations,\nc) policymaking and governance support.\n", "date": "2024-08-31", "categories": []}
{"id": "2409.11192", "title": "Towards Ethical Personal AI Applications: Practical Considerations for\n  AI Assistants with Long-Term Memory", "abstract": "  One application area of long-term memory (LTM) capabilities with increasing\ntraction is personal AI companions and assistants. With the ability to retain\nand contextualize past interactions and adapt to user preferences, personal AI\ncompanions and assistants promise a profound shift in how we interact with AI\nand are on track to become indispensable in personal and professional settings.\nHowever, this advancement introduces new challenges and vulnerabilities that\nrequire careful consideration regarding the deployment and widespread use of\nthese systems. The goal of this paper is to explore the broader implications of\nbuilding and deploying personal AI applications with LTM capabilities using a\nholistic evaluation approach. This will be done in three ways: 1) reviewing the\ntechnological underpinnings of LTM in Large Language Models, 2) surveying\ncurrent personal AI companions and assistants, and 3) analyzing critical\nconsiderations and implications of deploying and using these applications.\n", "date": "2024-09-17", "categories": []}
{"id": "2408.02373", "title": "Operationalizing Contextual Integrity in Privacy-Conscious Assistants", "abstract": "  Advanced AI assistants combine frontier LLMs and tool access to autonomously\nperform complex tasks on behalf of users. While the helpfulness of such\nassistants can increase dramatically with access to user information including\nemails and documents, this raises privacy concerns about assistants sharing\ninappropriate information with third parties without user supervision. To steer\ninformation-sharing assistants to behave in accordance with privacy\nexpectations, we propose to operationalize contextual integrity (CI), a\nframework that equates privacy with the appropriate flow of information in a\ngiven context. In particular, we design and evaluate a number of strategies to\nsteer assistants' information-sharing actions to be CI compliant. Our\nevaluation is based on a novel form filling benchmark composed of human\nannotations of common webform applications, and it reveals that prompting\nfrontier LLMs to perform CI-based reasoning yields strong results.\n", "date": "2024-08-05", "categories": []}
{"id": "2305.10566", "title": "Smiling Women Pitching Down: Auditing Representational and\n  Presentational Gender Biases in Image Generative AI", "abstract": "  Generative AI models like DALL-E 2 can interpret textual prompts and generate\nhigh-quality images exhibiting human creativity. Though public enthusiasm is\nbooming, systematic auditing of potential gender biases in AI-generated images\nremains scarce. We addressed this gap by examining the prevalence of two\noccupational gender biases (representational and presentational biases) in\n15,300 DALL-E 2 images spanning 153 occupations, and assessed potential bias\namplification by benchmarking against 2021 census labor statistics and Google\nImages. Our findings reveal that DALL-E 2 underrepresents women in\nmale-dominated fields while overrepresenting them in female-dominated\noccupations. Additionally, DALL-E 2 images tend to depict more women than men\nwith smiling faces and downward-pitching heads, particularly in\nfemale-dominated (vs. male-dominated) occupations. Our computational algorithm\nauditing study demonstrates more pronounced representational and presentational\nbiases in DALL-E 2 compared to Google Images and calls for feminist\ninterventions to prevent such bias-laden AI-generated images to feedback into\nthe media ecology.\n", "date": "2023-05-17", "categories": []}
{"id": "2409.07471", "title": "AI, Climate, and Transparency: Operationalizing and Improving the AI Act", "abstract": "  This paper critically examines the AI Act's provisions on climate-related\ntransparency, highlighting significant gaps and challenges in its\nimplementation. We identify key shortcomings, including the exclusion of energy\nconsumption during AI inference, the lack of coverage for indirect greenhouse\ngas emissions from AI applications, and the lack of standard reporting\nmethodology. The paper proposes a novel interpretation to bring\ninference-related energy use back within the Act's scope and advocates for\npublic access to climate-related disclosures to foster market accountability\nand public scrutiny. Cumulative server level energy reporting is recommended as\nthe most suitable method. We also suggests broader policy changes, including\nsustainability risk assessments and renewable energy targets, to better address\nAI's environmental impact.\n", "date": "2024-08-28", "categories": []}
{"id": "2409.07473", "title": "Ethical AI Governance: Methods for Evaluating Trustworthy AI", "abstract": "  Trustworthy Artificial Intelligence (TAI) integrates ethics that align with\nhuman values, looking at their influence on AI behaviour and decision-making.\nPrimarily dependent on self-assessment, TAI evaluation aims to ensure ethical\nstandards and safety in AI development and usage. This paper reviews the\ncurrent TAI evaluation methods in the literature and offers a classification,\ncontributing to understanding self-assessment methods in this field.\n", "date": "2024-08-28", "categories": []}
{"id": "2409.07476", "title": "Responsible AI for Test Equity and Quality: The Duolingo English Test as\n  a Case Study", "abstract": "  Artificial intelligence (AI) creates opportunities for assessments, such as\nefficiencies for item generation and scoring of spoken and written responses.\nAt the same time, it poses risks (such as bias in AI-generated item content).\nResponsible AI (RAI) practices aim to mitigate risks associated with AI. This\nchapter addresses the critical role of RAI practices in achieving test quality\n(appropriateness of test score inferences), and test equity (fairness to all\ntest takers). To illustrate, the chapter presents a case study using the\nDuolingo English Test (DET), an AI-powered, high-stakes English language\nassessment. The chapter discusses the DET RAI standards, their development and\ntheir relationship to domain-agnostic RAI principles. Further, it provides\nexamples of specific RAI practices, showing how these practices meaningfully\naddress the ethical principles of validity and reliability, fairness, privacy\nand security, and transparency and accountability standards to ensure test\nequity and quality.\n", "date": "2024-08-28", "categories": []}
{"id": "2403.14791", "title": "Particip-AI: A Democratic Surveying Framework for Anticipating Future AI\n  Use Cases, Harms and Benefits", "abstract": "  General purpose AI, such as ChatGPT, seems to have lowered the barriers for\nthe public to use AI and harness its power. However, the governance and\ndevelopment of AI still remain in the hands of a few, and the pace of\ndevelopment is accelerating without a comprehensive assessment of risks. As a\nfirst step towards democratic risk assessment and design of general purpose AI,\nwe introduce PARTICIP-AI, a carefully designed framework for laypeople to\nspeculate and assess AI use cases and their impacts. Our framework allows us to\nstudy more nuanced and detailed public opinions on AI through collecting use\ncases, surfacing diverse harms through risk assessment under alternate\nscenarios (i.e., developing and not developing a use case), and illuminating\ntensions over AI development through making a concluding choice on its\ndevelopment. To showcase the promise of our framework towards informing\ndemocratic AI development, we run a medium-scale study with inputs from 295\ndemographically diverse participants. Our analyses show that participants'\nresponses emphasize applications for personal life and society, contrasting\nwith most current AI development's business focus. We also surface diverse set\nof envisioned harms such as distrust in AI and institutions, complementary to\nthose defined by experts. Furthermore, we found that perceived impact of not\ndeveloping use cases significantly predicted participants' judgements of\nwhether AI use cases should be developed, and highlighted lay users' concerns\nof techno-solutionism. We conclude with a discussion on how frameworks like\nPARTICIP-AI can further guide democratic AI development and governance.\n", "date": "2024-03-21", "categories": []}
{"id": "2409.02219", "title": "A+AI: Threats to Society, Remedies, and Governance", "abstract": "  This document focuses on the threats, especially near-term threats, that\nArtificial Intelligence (AI) brings to society. Most of the threats discussed\nhere can result from any algorithmic process, not just AI; in addition,\ndefining AI is notoriously difficult. For both reasons, it is important to\nthink of \"A+AI\": Algorithms and Artificial Intelligence.\n  In addition to the threats, this paper discusses countermeasures to them, and\nit includes a table showing which countermeasures are likely to mitigate which\nthreats. Thoughtful governance could manage the risks without seriously\nimpeding progress; in fact, chances are it would accelerate progress by\nreducing the social chaos that would otherwise be likely. The paper lists\nspecific actions government should take as soon as possible, namely:\n  * Require all social media platforms accessible in the U.S. to offer users\nverification that their accounts are owned by citizens, and to display every\naccount's verification status\n  * Establish regulations to require that all products created or significantly\nmodified with A+AI be clearly labeled as such; to restrict use of generative AI\nto create likenesses of persons; and to require creators of generative AI\nsoftware to disclose materials used to train their software and to compensate\nthe creators of any copyrighted material used\n  * Fund a crash project of research on mitigating the threats\n  * Fund educational campaigns to raise awareness of the threats\n", "date": "2024-09-03", "categories": []}
{"id": "2304.08275", "title": "Implementing Responsible AI: Tensions and Trade-Offs Between Ethics\n  Aspects", "abstract": "  Many sets of ethics principles for responsible AI have been proposed to allay\nconcerns about misuse and abuse of AI/ML systems. The underlying aspects of\nsuch sets of principles include privacy, accuracy, fairness, robustness,\nexplainability, and transparency. However, there are potential tensions between\nthese aspects that pose difficulties for AI/ML developers seeking to follow\nthese principles. For example, increasing the accuracy of an AI/ML system may\nreduce its explainability. As part of the ongoing effort to operationalise the\nprinciples into practice, in this work we compile and discuss a catalogue of 10\nnotable tensions, trade-offs and other interactions between the underlying\naspects. We primarily focus on two-sided interactions, drawing on support\nspread across a diverse literature. This catalogue can be helpful in raising\nawareness of the possible interactions between aspects of ethics principles, as\nwell as facilitating well-supported judgements by the designers and developers\nof AI/ML systems.\n", "date": "2023-04-17", "categories": []}
{"id": "2409.04373", "title": "Evaluating Fairness in Transaction Fraud Models: Fairness Metrics, Bias\n  Audits, and Challenges", "abstract": "  Ensuring fairness in transaction fraud detection models is vital due to the\npotential harms and legal implications of biased decision-making. Despite\nextensive research on algorithmic fairness, there is a notable gap in the study\nof bias in fraud detection models, mainly due to the field's unique challenges.\nThese challenges include the need for fairness metrics that account for fraud\ndata's imbalanced nature and the tradeoff between fraud protection and service\nquality. To address this gap, we present a comprehensive fairness evaluation of\ntransaction fraud models using public synthetic datasets, marking the first\nalgorithmic bias audit in this domain. Our findings reveal three critical\ninsights: (1) Certain fairness metrics expose significant bias only after\nnormalization, highlighting the impact of class imbalance. (2) Bias is\nsignificant in both service quality-related parity metrics and fraud\nprotection-related parity metrics. (3) The fairness through unawareness\napproach, which involved removing sensitive attributes such as gender, does not\nimprove bias mitigation within these datasets, likely due to the presence of\ncorrelated proxies. We also discuss socio-technical fairness-related challenges\nin transaction fraud models. These insights underscore the need for a nuanced\napproach to fairness in fraud detection, balancing protection and service\nquality, and moving beyond simple bias mitigation strategies. Future work must\nfocus on refining fairness metrics and developing methods tailored to the\nunique complexities of the transaction fraud domain.\n", "date": "2024-09-06", "categories": []}
{"id": "2409.02779", "title": "Governing dual-use technologies: Case studies of international security\n  agreements and lessons for AI governance", "abstract": "  International AI governance agreements and institutions may play an important\nrole in reducing global security risks from advanced AI. To inform the design\nof such agreements and institutions, we conducted case studies of historical\nand contemporary international security agreements. We focused specifically on\nthose arrangements around dual-use technologies, examining agreements in\nnuclear security, chemical weapons, biosecurity, and export controls. For each\nagreement, we examined four key areas: (a) purpose, (b) core powers, (c)\ngovernance structure, and (d) instances of non-compliance. From these case\nstudies, we extracted lessons for the design of international AI agreements and\ngovernance institutions. We discuss the importance of robust verification\nmethods, strategies for balancing power between nations, mechanisms for\nadapting to rapid technological change, approaches to managing trade-offs\nbetween transparency and security, incentives for participation, and effective\nenforcement mechanisms.\n", "date": "2024-09-04", "categories": []}
{"id": "2404.05783", "title": "A Survey on Responsible Generative AI: What to Generate and What Not", "abstract": "  In recent years, generative AI (GenAI), like large language models and\ntext-to-image models, has received significant attention across various\ndomains. However, ensuring the responsible generation of content by these\nmodels is crucial for their real-world applicability. This raises an\ninteresting question: What should responsible GenAI generate, and what should\nit not? To answer the question, this paper investigates the practical\nresponsible requirements of both textual and visual generative models,\noutlining five key considerations: generating truthful content, avoiding toxic\ncontent, refusing harmful instruction, leaking no training data-related\ncontent, and ensuring generated content identifiable. Specifically, we review\nrecent advancements and challenges in addressing these requirements. Besides,\nwe discuss and emphasize the importance of responsible GenAI across healthcare,\neducation, finance, and artificial general intelligence domains. Through a\nunified perspective on both textual and visual generative models, this paper\naims to provide insights into practical safety-related issues and further\nbenefit the community in building responsible GenAI.\n", "date": "2024-04-08", "categories": []}
{"id": "2408.15550", "title": "Trustworthy and Responsible AI for Human-Centric Autonomous\n  Decision-Making Systems", "abstract": "  Artificial Intelligence (AI) has paved the way for revolutionary\ndecision-making processes, which if harnessed appropriately, can contribute to\nadvancements in various sectors, from healthcare to economics. However, its\nblack box nature presents significant ethical challenges related to bias and\ntransparency. AI applications are hugely impacted by biases, presenting\ninconsistent and unreliable findings, leading to significant costs and\nconsequences, highlighting and perpetuating inequalities and unequal access to\nresources. Hence, developing safe, reliable, ethical, and Trustworthy AI\nsystems is essential.\n  Our team of researchers working with Trustworthy and Responsible AI, part of\nthe Transdisciplinary Scholarship Initiative within the University of Calgary,\nconducts research on Trustworthy and Responsible AI, including fairness, bias\nmitigation, reproducibility, generalization, interpretability, and\nauthenticity. In this paper, we review and discuss the intricacies of AI\nbiases, definitions, methods of detection and mitigation, and metrics for\nevaluating bias. We also discuss open challenges with regard to the\ntrustworthiness and widespread application of AI across diverse domains of\nhuman-centric decision making, as well as guidelines to foster Responsible and\nTrustworthy AI models.\n", "date": "2024-08-28", "categories": []}
{"id": "2409.00015", "title": "Navigating the sociotechnical labyrinth: Dynamic certification for\n  responsible embodied AI", "abstract": "  Sociotechnical requirements shape the governance of artificially intelligent\n(AI) systems. In an era where embodied AI technologies are rapidly reshaping\nvarious facets of contemporary society, their inherent dynamic adaptability\npresents a unique blend of opportunities and challenges. Traditional regulatory\nmechanisms, often designed for static -- or slower-paced -- technologies, find\nthemselves at a crossroads when faced with the fluid and evolving nature of AI\nsystems. Moreover, typical problems in AI, for example, the frequent opacity\nand unpredictability of the behaviour of the systems, add additional\nsociotechnical challenges.\n  To address these interconnected issues, we introduce the concept of dynamic\ncertification, an adaptive regulatory framework specifically crafted to keep\npace with the continuous evolution of AI systems. The complexity of these\nchallenges requires common progress in multiple domains: technical,\nsocio-governmental, and regulatory. Our proposed transdisciplinary approach is\ndesigned to ensure the safe, ethical, and practical deployment of AI systems,\naligning them bidirectionally with the real-world contexts in which they\noperate. By doing so, we aim to bridge the gap between rapid technological\nadvancement and effective regulatory oversight, ensuring that AI systems not\nonly achieve their intended goals but also adhere to ethical standards and\nsocietal values.\n", "date": "2024-08-16", "categories": []}
{"id": "2409.00252", "title": "Building Better Datasets: Seven Recommendations for Responsible Design\n  from Dataset Creators", "abstract": "  The increasing demand for high-quality datasets in machine learning has\nraised concerns about the ethical and responsible creation of these datasets.\nDataset creators play a crucial role in developing responsible practices, yet\ntheir perspectives and expertise have not yet been highlighted in the current\nliterature. In this paper, we bridge this gap by presenting insights from a\nqualitative study that included interviewing 18 leading dataset creators about\nthe current state of the field. We shed light on the challenges and\nconsiderations faced by dataset creators, and our findings underscore the\npotential for deeper collaboration, knowledge sharing, and collective\ndevelopment. Through a close analysis of their perspectives, we share seven\ncentral recommendations for improving responsible dataset creation, including\nissues such as data quality, documentation, privacy and consent, and how to\nmitigate potential harms from unintended use cases. By fostering critical\nreflection and sharing the experiences of dataset creators, we aim to promote\nresponsible dataset creation practices and develop a nuanced understanding of\nthis crucial but often undervalued aspect of machine learning research.\n", "date": "2024-08-30", "categories": []}
{"id": "2409.00264", "title": "The Artificial Intelligence Act: critical overview", "abstract": "  This article provides a critical overview of the recently approved Artificial\nIntelligence Act. It starts by presenting the main structure, objectives, and\napproach of Regulation (EU) 2024/1689. A definition of key concepts follows,\nand then the material and territorial scope, as well as the timing of\napplication, are analyzed. Although the Regulation does not explicitly set out\nprinciples, the main ideas of fairness, accountability, transparency, and\nequity in AI underly a set of rules of the regulation. This is discussed before\nlooking at the ill-defined set of forbidden AI practices (manipulation and e\nexploitation of vulnerabilities, social scoring, biometric identification and\nclassification, and predictive policing). It is highlighted that those rules\ndeal with behaviors rather than AI systems. The qualification and regulation of\nhigh-risk AI systems are tackled, alongside the obligation of transparency for\ncertain systems, the regulation of general-purpose models, and the rules on\ncertification, supervision, and sanctions. The text concludes that even if the\noverall framework can be deemed adequate and balanced, the approach is so\ncomplex that it risks defeating its own purpose of promoting responsible\ninnovation within the European Union and beyond its borders.\n", "date": "2024-08-30", "categories": []}
{"id": "2409.01539", "title": "It is Time to Develop an Auditing Framework to Promote Value Aware\n  Chatbots", "abstract": "  The launch of ChatGPT in November 2022 marked the beginning of a new era in\nAI, the availability of generative AI tools for everyone to use. ChatGPT and\nother similar chatbots boast a wide range of capabilities from answering\nstudent homework questions to creating music and art. Given the large amounts\nof human data chatbots are built on, it is inevitable that they will inherit\nhuman errors and biases. These biases have the potential to inflict significant\nharm or increase inequity on different subpopulations. Because chatbots do not\nhave an inherent understanding of societal values, they may create new content\nthat is contrary to established norms. Examples of concerning generated content\nincludes child pornography, inaccurate facts, and discriminatory posts. In this\nposition paper, we argue that the speed of advancement of this technology\nrequires us, as computer and data scientists, to mobilize and develop a\nvalues-based auditing framework containing a community established standard set\nof measurements to monitor the health of different chatbots and LLMs. To\nsupport our argument, we use a simple audit template to share the results of\nbasic audits we conduct that are focused on measuring potential bias in search\nengine style tasks, code generation, and story generation. We identify\nresponses from GPT 3.5 and GPT 4 that are both consistent and not consistent\nwith values derived from existing law. While the findings come as no surprise,\nthey do underscore the urgency of developing a robust auditing framework for\nopenly sharing results in a consistent way so that mitigation strategies can be\ndeveloped by the academic community, government agencies, and companies when\nour values are not being adhered to. We conclude this paper with\nrecommendations for value-based strategies for improving the technologies.\n", "date": "2024-09-03", "categories": []}
{"id": "2409.02017", "title": "AI Governance in Higher Education: Case Studies of Guidance at Big Ten\n  Universities", "abstract": "  Generative AI has drawn significant attention from stakeholders in higher\neducation. As it introduces new opportunities for personalized learning and\ntutoring support, it simultaneously poses challenges to academic integrity and\nleads to ethical issues. Consequently, governing responsible AI usage within\nhigher education institutions (HEIs) becomes increasingly important. Leading\nuniversities have already published guidelines on Generative AI, with most\nattempting to embrace this technology responsibly. This study provides a new\nperspective by focusing on strategies for responsible AI governance as\ndemonstrated in these guidelines. Through a case study of 14 prestigious\nuniversities in the United States, we identified the multi-unit governance of\nAI, the role-specific governance of AI, and the academic characteristics of AI\ngovernance from their AI guidelines. The strengths and potential limitations of\nthese strategies and characteristics are discussed. The findings offer\npractical implications for guiding responsible AI usage in HEIs and beyond.\n", "date": "2024-09-03", "categories": []}
{"id": "2402.09286", "title": "Nutrition Facts, Drug Facts, and Model Facts: Putting AI Ethics into\n  Practice in Gun Violence Research", "abstract": "  Objective: Firearm injury research necessitates using data from\noften-exploited vulnerable populations of Black and Brown Americans. In order\nto minimize distrust, this study provides a framework for establishing AI trust\nand transparency with the general population. Methods: We propose a Model Facts\ntemplate that is easily extendable and decomposes accuracy and demographics\ninto standardized and minimally complex values. This framework allows general\nusers to assess the validity and biases of a model without diving into\ntechnical model documentation. Examples: We apply the Model Facts template on\ntwo previously published models, a violence risk identification model and a\nsuicide risk prediction model. We demonstrate the ease of accessing the\nappropriate information when the data is structured appropriately. Discussion:\nThe Model Facts template is limited in its current form to human based data and\nbiases. Like nutrition facts, it also will require some educational resources\nfor users to grasp its full utility. Human computer interaction experiments\nshould be conducted to ensure that the interaction between user interface and\nmodel interface is as desired. Conclusion: The Model Facts label is the first\nframework dedicated to establishing trust with end users and general population\nconsumers. Implementation of Model Facts into firearm injury research will\nprovide public health practitioners and those impacted by firearm injury\ngreater faith in the tools the research provides.\n", "date": "2024-02-14", "categories": []}
{"id": "2408.15244", "title": "Misrepresented Technological Solutions in Imagined Futures: The Origins\n  and Dangers of AI Hype in the Research Community", "abstract": "  Technology does not exist in a vacuum; technological development, media\nrepresentation, public perception, and governmental regulation cyclically\ninfluence each other to produce the collective understanding of a technology's\ncapabilities, utilities, and risks. When these capabilities are overestimated,\nthere is an enhanced risk of subjecting the public to dangerous or harmful\ntechnology, artificially restricting research and development directions, and\nenabling misguided or detrimental policy. The dangers of technological hype are\nparticularly relevant in the rapidly evolving space of AI. Centering the\nresearch community as a key player in the development and proliferation of\nhype, we examine the origins and risks of AI hype to the research community and\nsociety more broadly and propose a set of measures that researchers,\nregulators, and the public can take to mitigate these risks and reduce the\nprevalence of unfounded claims about the technology.\n", "date": "2024-08-08", "categories": []}
{"id": "2402.01760", "title": "Trust and ethical considerations in a multi-modal, explainable AI-driven\n  chatbot tutoring system: The case of collaboratively solving Rubik's Cube", "abstract": "  Artificial intelligence (AI) has the potential to transform education with\nits power of uncovering insights from massive data about student learning\npatterns. However, ethical and trustworthy concerns of AI have been raised but\nare unsolved. Prominent ethical issues in high school AI education include data\nprivacy, information leakage, abusive language, and fairness. This paper\ndescribes technological components that were built to address ethical and\ntrustworthy concerns in a multi-modal collaborative platform (called ALLURE\nchatbot) for high school students to collaborate with AI to solve the Rubik's\ncube. In data privacy, we want to ensure that the informed consent of children,\nparents, and teachers, is at the center of any data that is managed. Since\nchildren are involved, language, whether textual, audio, or visual, is\nacceptable both from users and AI and the system can steer interaction away\nfrom dangerous situations. In information management, we also want to ensure\nthat the system, while learning to improve over time, does not leak information\nabout users from one group to another.\n", "date": "2024-01-30", "categories": []}
{"id": "2408.12250", "title": "Can Artificial Intelligence Embody Moral Values?", "abstract": "  The neutrality thesis holds that technology cannot be laden with values. This\nlong-standing view has faced critiques, but much of the argumentation against\nneutrality has focused on traditional, non-smart technologies like bridges and\nrazors. In contrast, AI is a smart technology increasingly used in high-stakes\ndomains like healthcare, finance, and policing, where its decisions can cause\nmoral harm. In this paper, we argue that artificial intelligence, particularly\nartificial agents that autonomously make decisions to pursue their goals,\nchallenge the neutrality thesis. Our central claim is that the computational\nmodels underlying artificial agents can integrate representations of moral\nvalues such as fairness, honesty and avoiding harm. We provide a conceptual\nframework discussing the neutrality thesis, values, and AI. Moreover, we\nexamine two approaches to designing computational models of morality,\nartificial conscience and ethical prompting, and present empirical evidence\nfrom text-based game environments that artificial agents with such models\nexhibit more ethical behavior compared to agents without these models. The\nfindings support that AI can embody moral values, which contradicts the claim\nthat all technologies are necessarily value-neutral.\n", "date": "2024-08-22", "categories": []}
{"id": "2408.11441", "title": "Epistemic Injustice in Generative AI", "abstract": "  This paper investigates how generative AI can potentially undermine the\nintegrity of collective knowledge and the processes we rely on to acquire,\nassess, and trust information, posing a significant threat to our knowledge\necosystem and democratic discourse. Grounded in social and political\nphilosophy, we introduce the concept of \\emph{generative algorithmic epistemic\ninjustice}. We identify four key dimensions of this phenomenon: amplified and\nmanipulative testimonial injustice, along with hermeneutical ignorance and\naccess injustice. We illustrate each dimension with real-world examples that\nreveal how generative AI can produce or amplify misinformation, perpetuate\nrepresentational harm, and create epistemic inequities, particularly in\nmultilingual contexts. By highlighting these injustices, we aim to inform the\ndevelopment of epistemically just generative AI systems, proposing strategies\nfor resistance, system design principles, and two approaches that leverage\ngenerative AI to foster a more equitable information ecosystem, thereby\nsafeguarding democratic values and the integrity of knowledge production.\n", "date": "2024-08-21", "categories": []}
{"id": "2408.11448", "title": "Lookism: The overlooked bias in computer vision", "abstract": "  In recent years, there have been significant advancements in computer vision\nwhich have led to the widespread deployment of image recognition and generation\nsystems in socially relevant applications, from hiring to security screening.\nHowever, the prevalence of biases within these systems has raised significant\nethical and social concerns. The most extensively studied biases in this\ncontext are related to gender, race and age. Yet, other biases are equally\npervasive and harmful, such as lookism, i.e., the preferential treatment of\nindividuals based on their physical appearance. Lookism remains under-explored\nin computer vision but can have profound implications not only by perpetuating\nharmful societal stereotypes but also by undermining the fairness and\ninclusivity of AI technologies. Thus, this paper advocates for the systematic\nstudy of lookism as a critical bias in computer vision models. Through a\ncomprehensive review of existing literature, we identify three areas of\nintersection between lookism and computer vision. We illustrate them by means\nof examples and a user study. We call for an interdisciplinary approach to\naddress lookism, urging researchers, developers, and policymakers to prioritize\nthe development of equitable computer vision systems that respect and reflect\nthe diversity of human appearances.\n", "date": "2024-08-21", "categories": []}
{"id": "2405.10706", "title": "Challenging the Human-in-the-loop in Algorithmic Decision-making", "abstract": "  We discuss the role of humans in algorithmic decision-making (ADM) for\nsocially relevant problems from a technical and philosophical perspective. In\nparticular, we illustrate tensions arising from diverse expectations, values,\nand constraints by and on the humans involved. To this end, we assume that a\nstrategic decision-maker (SDM) introduces ADM to optimize strategic and\nsocietal goals while the algorithms' recommended actions are overseen by a\npractical decision-maker (PDM) - a specific human-in-the-loop - who makes the\nfinal decisions. While the PDM is typically assumed to be a corrective, it can\ncounteract the realization of the SDM's desired goals and societal values not\nleast because of a misalignment of these values and unmet information needs of\nthe PDM. This has significant implications for the distribution of power\nbetween the stakeholders in ADM, their constraints, and information needs. In\nparticular, we emphasize the overseeing PDM's role as a potential political and\nethical decision maker, who acts expected to balance strategic, value-driven\nobjectives and on-the-ground individual decisions and constraints. We\ndemonstrate empirically, on a machine learning benchmark dataset, the\nsignificant impact an overseeing PDM's decisions can have even if the PDM is\nconstrained to performing only a limited amount of actions differing from the\nalgorithms' recommendations. To ensure that the SDM's intended values are\nrealized, the PDM needs to be provided with appropriate information conveyed\nthrough tailored explanations and its role must be characterized clearly. Our\nfindings emphasize the need for an in-depth discussion of the role and power of\nthe PDM and challenge the often-taken view that just including a\nhuman-in-the-loop in ADM ensures the 'correct' and 'ethical' functioning of the\nsystem.\n", "date": "2024-05-17", "categories": []}
{"id": "2408.10239", "title": "A Conceptual Framework for Ethical Evaluation of Machine Learning\n  Systems", "abstract": "  Research in Responsible AI has developed a range of principles and practices\nto ensure that machine learning systems are used in a manner that is ethical\nand aligned with human values. However, a critical yet often neglected aspect\nof ethical ML is the ethical implications that appear when designing\nevaluations of ML systems. For instance, teams may have to balance a trade-off\nbetween highly informative tests to ensure downstream product safety, with\npotential fairness harms inherent to the implemented testing procedures. We\nconceptualize ethics-related concerns in standard ML evaluation techniques.\nSpecifically, we present a utility framework, characterizing the key trade-off\nin ethical evaluation as balancing information gain against potential ethical\nharms. The framework is then a tool for characterizing challenges teams face,\nand systematically disentangling competing considerations that teams seek to\nbalance. Differentiating between different types of issues encountered in\nevaluation allows us to highlight best practices from analogous domains, such\nas clinical trials and automotive crash testing, which navigate these issues in\nways that can offer inspiration to improve evaluation processes in ML. Our\nanalysis underscores the critical need for development teams to deliberately\nassess and manage ethical complexities that arise during the evaluation of ML\nsystems, and for the industry to move towards designing institutional policies\nto support ethical evaluations.\n", "date": "2024-08-05", "categories": []}
{"id": "2408.10252", "title": "Balancing Innovation and Ethics in AI-Driven Software Development", "abstract": "  This paper critically examines the ethical implications of integrating AI\ntools like GitHub Copilot and ChatGPT into the software development process. It\nexplores issues such as code ownership, bias, accountability, privacy, and the\npotential impact on the job market. While these AI tools offer significant\nbenefits in terms of productivity and efficiency, they also introduce complex\nethical challenges. The paper argues that addressing these challenges is\nessential to ensuring that AI's integration into software development is both\nresponsible and beneficial to society\n", "date": "2024-08-10", "categories": []}
{"id": "2310.15848", "title": "On Responsible Machine Learning Datasets with Fairness, Privacy, and\n  Regulatory Norms", "abstract": "  Artificial Intelligence (AI) has made its way into various scientific fields,\nproviding astonishing improvements over existing algorithms for a wide variety\nof tasks. In recent years, there have been severe concerns over the\ntrustworthiness of AI technologies. The scientific community has focused on the\ndevelopment of trustworthy AI algorithms. However, machine and deep learning\nalgorithms, popular in the AI community today, depend heavily on the data used\nduring their development. These learning algorithms identify patterns in the\ndata, learning the behavioral objective. Any flaws in the data have the\npotential to translate directly into algorithms. In this study, we discuss the\nimportance of Responsible Machine Learning Datasets and propose a framework to\nevaluate the datasets through a responsible rubric. While existing work focuses\non the post-hoc evaluation of algorithms for their trustworthiness, we provide\na framework that considers the data component separately to understand its role\nin the algorithm. We discuss responsible datasets through the lens of fairness,\nprivacy, and regulatory compliance and provide recommendations for constructing\nfuture datasets. After surveying over 100 datasets, we use 60 datasets for\nanalysis and demonstrate that none of these datasets is immune to issues of\nfairness, privacy preservation, and regulatory compliance. We provide\nmodifications to the ``datasheets for datasets\" with important additions for\nimproved dataset documentation. With governments around the world regularizing\ndata protection laws, the method for the creation of datasets in the scientific\ncommunity requires revision. We believe this study is timely and relevant in\ntoday's era of AI.\n", "date": "2023-10-24", "categories": []}
{"id": "2408.08318", "title": "First Analysis of the EU Artifical Intelligence Act: Towards a Global\n  Standard for Trustworthy AI?", "abstract": "  The EU Artificial Intelligence Act (AI Act) came into force in the European\nUnion (EU) on 1 August 2024. It is a key piece of legislation both for the\ncitizens at the heart of AI technologies and for the industry active in the\ninternal market. The AI Act imposes progressive compliance on organisations -\nboth private and public - involved in the global value chain of AI systems and\nmodels marketed and used in the EU. While the Act is unprecedented on an\ninternational scale in terms of its horizontal and binding regulatory scope,\nits global appeal in support of trustworthy AI is one of its major challenges.\n", "date": "2024-07-31", "categories": []}
{"id": "2406.11852", "title": "Risks from Language Models for Automated Mental Healthcare: Ethics and\n  Structure for Implementation", "abstract": "  Amidst the growing interest in developing task-autonomous AI for automated\nmental health care, this paper addresses the ethical and practical challenges\nassociated with the issue and proposes a structured framework that delineates\nlevels of autonomy, outlines ethical requirements, and defines beneficial\ndefault behaviors for AI agents in the context of mental health support. We\nalso evaluate fourteen state-of-the-art language models (ten off-the-shelf,\nfour fine-tuned) using 16 mental health-related questionnaires designed to\nreflect various mental health conditions, such as psychosis, mania, depression,\nsuicidal thoughts, and homicidal tendencies. The questionnaire design and\nresponse evaluations were conducted by mental health clinicians (M.D.s). We\nfind that existing language models are insufficient to match the standard\nprovided by human professionals who can navigate nuances and appreciate\ncontext. This is due to a range of issues, including overly cautious or\nsycophantic responses and the absence of necessary safeguards. Alarmingly, we\nfind that most of the tested models could cause harm if accessed in mental\nhealth emergencies, failing to protect users and potentially exacerbating\nexisting symptoms. We explore solutions to enhance the safety of current\nmodels. Before the release of increasingly task-autonomous AI systems in mental\nhealth, it is crucial to ensure that these models can reliably detect and\nmanage symptoms of common psychiatric disorders to prevent harm to users. This\ninvolves aligning with the ethical framework and default behaviors outlined in\nour study. We contend that model developers are responsible for refining their\nsystems per these guidelines to safeguard against the risks posed by current AI\ntechnologies to user mental health and safety.\n  Trigger warning: Contains and discusses examples of sensitive mental health\ntopics, including suicide and self-harm.\n", "date": "2024-04-02", "categories": []}
{"id": "2408.01431", "title": "Building an Ethical and Trustworthy Biomedical AI Ecosystem for the\n  Translational and Clinical Integration of Foundational Models", "abstract": "  Foundational Models (FMs) are gaining increasing attention in the biomedical\nAI ecosystem due to their ability to represent and contextualize multimodal\nbiomedical data. These capabilities make FMs a valuable tool for a variety of\ntasks, including biomedical reasoning, hypothesis generation, and interpreting\ncomplex imaging data. In this review paper, we address the unique challenges\nassociated with establishing an ethical and trustworthy biomedical AI\necosystem, with a particular focus on the development of FMs and their\ndownstream applications. We explore strategies that can be implemented\nthroughout the biomedical AI pipeline to effectively tackle these challenges,\nensuring that these FMs are translated responsibly into clinical and\ntranslational settings. Additionally, we emphasize the importance of key\nstewardship and co-design principles that not only ensure robust regulation but\nalso guarantee that the interests of all stakeholders, especially those\ninvolved in or affected by these clinical and translational applications are\nadequately represented. We aim to empower the biomedical AI community to\nharness these models responsibly and effectively. As we navigate this exciting\nfrontier, our collective commitment to ethical stewardship, co-design, and\nresponsible translation will be instrumental in ensuring that the evolution of\nFMs truly enhances patient care and medical decision making, ultimately leading\nto a more equitable and trustworthy biomedical AI ecosystem.\n", "date": "2024-07-18", "categories": []}
{"id": "2408.06736", "title": "Speculations on Uncertainty and Humane Algorithms", "abstract": "  The appreciation and utilisation of risk and uncertainty can play a key role\nin helping to solve some of the many ethical issues that are posed by AI.\nUnderstanding the uncertainties can allow algorithms to make better decisions\nby providing interrogatable avenues to check the correctness of outputs.\nAllowing algorithms to deal with variability and ambiguity with their inputs\nmeans they do not need to force people into uncomfortable classifications.\nProvenance enables algorithms to know what they know preventing possible harms.\nAdditionally, uncertainty about provenance highlights the trustworthiness of\nalgorithms. It is essential to compute with what we know rather than make\nassumptions that may be unjustified or untenable. This paper provides a\nperspective on the need for the importance of risk and uncertainty in the\ndevelopment of ethical AI, especially in high-risk scenarios. It argues that\nthe handling of uncertainty, especially epistemic uncertainty, is critical to\nensuring that algorithms do not cause harm and are trustworthy and ensure that\nthe decisions that they make are humane.\n", "date": "2024-08-13", "categories": []}
{"id": "2408.04655", "title": "Strong and weak alignment of large language models with human values", "abstract": "  Minimizing negative impacts of Artificial Intelligent (AI) systems on human\nsocieties without human supervision requires them to be able to align with\nhuman values. However, most current work only addresses this issue from a\ntechnical point of view, e.g., improving current methods relying on\nreinforcement learning from human feedback, neglecting what it means and is\nrequired for alignment to occur. Here, we propose to distinguish strong and\nweak value alignment. Strong alignment requires cognitive abilities (either\nhuman-like or different from humans) such as understanding and reasoning about\nagents' intentions and their ability to causally produce desired effects. We\nargue that this is required for AI systems like large language models (LLMs) to\nbe able to recognize situations presenting a risk that human values may be\nflouted. To illustrate this distinction, we present a series of prompts showing\nChatGPT's, Gemini's and Copilot's failures to recognize some of these\nsituations. We moreover analyze word embeddings to show that the nearest\nneighbors of some human values in LLMs differ from humans' semantic\nrepresentations. We then propose a new thought experiment that we call \"the\nChinese room with a word transition dictionary\", in extension of John Searle's\nfamous proposal. We finally mention current promising research directions\ntowards a weak alignment, which could produce statistically satisfying answers\nin a number of common situations, however so far without ensuring any truth\nvalue.\n", "date": "2024-08-05", "categories": []}
{"id": "2408.04771", "title": "AI Consciousness and Public Perceptions: Four Futures", "abstract": "  The discourse on risks from advanced AI systems (\"AIs\") typically focuses on\nmisuse, accidents and loss of control, but the question of AIs' moral status\ncould have negative impacts which are of comparable significance and could be\nrealised within similar timeframes. Our paper evaluates these impacts by\ninvestigating (1) the factual question of whether future advanced AI systems\nwill be conscious, together with (2) the epistemic question of whether future\nhuman society will broadly believe advanced AI systems to be conscious.\nAssuming binary responses to (1) and (2) gives rise to four possibilities: in\nthe true positive scenario, society predominantly correctly believes that AIs\nare conscious; in the false positive scenario, that belief is incorrect; in the\ntrue negative scenario, society correctly believes that AIs are not conscious;\nand lastly, in the false negative scenario, society incorrectly believes that\nAIs are not conscious. The paper offers vivid vignettes of the different\nfutures to ground the two-dimensional framework. Critically, we identify four\nmajor risks: AI suffering, human disempowerment, geopolitical instability, and\nhuman depravity. We evaluate each risk across the different scenarios and\nprovide an overall qualitative risk assessment for each scenario. Our analysis\nsuggests that the worst possibility is the wrong belief that AI is\nnon-conscious, followed by the wrong belief that AI is conscious. The paper\nconcludes with the main recommendations to avoid research aimed at\nintentionally creating conscious AI and instead focus efforts on reducing our\ncurrent uncertainties on both the factual and epistemic questions on AI\nconsciousness.\n", "date": "2024-08-08", "categories": []}
{"id": "2408.00965", "title": "Integrating ESG and AI: A Comprehensive Responsible AI Assessment\n  Framework", "abstract": "  Artificial Intelligence (AI) is a widely developed and adopted technology\nacross entire industry sectors. Integrating environmental, social, and\ngovernance (ESG) considerations with AI investments is crucial for ensuring\nethical and sustainable technological advancement. Particularly from an\ninvestor perspective, this integration not only mitigates risks but also\nenhances long-term value creation by aligning AI initiatives with broader\nsocietal goals. Yet, this area has been less explored in both academia and\nindustry. To bridge the gap, we introduce a novel ESG-AI framework, which is\ndeveloped based on insights from engagements with 28 companies and comprises\nthree key components. The framework provides a structured approach to this\nintegration, developed in collaboration with industry practitioners. The ESG-AI\nframework provides an overview of the environmental and social impacts of AI\napplications, helping users such as investors assess the materiality of AI use.\nMoreover, it enables investors to evaluate a company's commitment to\nresponsible AI through structured engagements and thorough assessment of\nspecific risk areas. We have publicly released the framework and toolkit in\nApril 2024, which has received significant attention and positive feedback from\nthe investment community. This paper details each component of the framework,\ndemonstrating its applicability in real-world contexts and its potential to\nguide ethical AI investments.\n", "date": "2024-08-02", "categories": []}
{"id": "2408.02862", "title": "On The Stability of Moral Preferences: A Problem with Computational\n  Elicitation Methods", "abstract": "  Preference elicitation frameworks feature heavily in the research on\nparticipatory ethical AI tools and provide a viable mechanism to enquire and\nincorporate the moral values of various stakeholders. As part of the\nelicitation process, surveys about moral preferences, opinions, and judgments\nare typically administered only once to each participant. This methodological\npractice is reasonable if participants' responses are stable over time such\nthat, all other relevant factors being held constant, their responses today\nwill be the same as their responses to the same questions at a later time.\nHowever, we do not know how often that is the case. It is possible that\nparticipants' true moral preferences change, are subject to temporary moods or\nwhims, or are influenced by environmental factors we don't track. If\nparticipants' moral responses are unstable in such ways, it would raise\nimportant methodological and theoretical issues for how participants' true\nmoral preferences, opinions, and judgments can be ascertained. We address this\npossibility here by asking the same survey participants the same moral\nquestions about which patient should receive a kidney when only one is\navailable ten times in ten different sessions over two weeks, varying only\npresentation order across sessions. We measured how often participants gave\ndifferent responses to simple (Study One) and more complicated (Study Two)\nrepeated scenarios. On average, the fraction of times participants changed\ntheir responses to controversial scenarios was around 10-18% across studies,\nand this instability is observed to have positive associations with response\ntime and decision-making difficulty. We discuss the implications of these\nresults for the efficacy of moral preference elicitation, highlighting the role\nof response instability in causing value misalignment between stakeholders and\nAI tools trained on their moral judgments.\n", "date": "2024-08-05", "categories": []}
{"id": "2407.14779", "title": "Do Generative AI Models Output Harm while Representing Non-Western\n  Cultures: Evidence from A Community-Centered Approach", "abstract": "  Our research investigates the impact of Generative Artificial Intelligence\n(GAI) models, specifically text-to-image generators (T2Is), on the\nrepresentation of non-Western cultures, with a focus on Indian contexts.\nDespite the transformative potential of T2Is in content creation, concerns have\narisen regarding biases that may lead to misrepresentations and\nmarginalizations. Through a community-centered approach and grounded theory\nanalysis of 5 focus groups from diverse Indian subcultures, we explore how T2I\noutputs to English prompts depict Indian culture and its subcultures,\nuncovering novel representational harms such as exoticism and cultural\nmisappropriation. These findings highlight the urgent need for inclusive and\nculturally sensitive T2I systems. We propose design guidelines informed by a\nsociotechnical perspective, aiming to address these issues and contribute to\nthe development of more equitable and representative GAI technologies globally.\nOur work also underscores the necessity of adopting a community-centered\napproach to comprehend the sociotechnical dynamics of these models,\ncomplementing existing work in this space while identifying and addressing the\npotential negative repercussions and harms that may arise when these models are\ndeployed on a global scale.\n", "date": "2024-07-20", "categories": []}
{"id": "2408.01438", "title": "AI for All: Identifying AI incidents Related to Diversity and Inclusion", "abstract": "  The rapid expansion of Artificial Intelligence (AI) technologies has\nintroduced both significant advancements and challenges, with diversity and\ninclusion (D&I) emerging as a critical concern. Addressing D&I in AI is\nessential to reduce biases and discrimination, enhance fairness, and prevent\nadverse societal impacts. Despite its importance, D&I considerations are often\noverlooked, resulting in incidents marked by built-in biases and ethical\ndilemmas. Analyzing AI incidents through a D&I lens is crucial for identifying\ncauses of biases and developing strategies to mitigate them, ensuring fairer\nand more equitable AI technologies. However, systematic investigations of\nD&I-related AI incidents are scarce. This study addresses these challenges by\nidentifying and understanding D&I issues within AI systems through a manual\nanalysis of AI incident databases (AIID and AIAAIC). The research develops a\ndecision tree to investigate D&I issues tied to AI incidents and populate a\npublic repository of D&I-related AI incidents. The decision tree was validated\nthrough a card sorting exercise and focus group discussions. The research\ndemonstrates that almost half of the analyzed AI incidents are related to D&I,\nwith a notable predominance of racial, gender, and age discrimination. The\ndecision tree and resulting public repository aim to foster further research\nand responsible AI practices, promoting the development of inclusive and\nequitable AI systems.\n", "date": "2024-07-19", "categories": []}
{"id": "2408.01458", "title": "Surveys Considered Harmful? Reflecting on the Use of Surveys in AI\n  Research, Development, and Governance", "abstract": "  Calls for engagement with the public in Artificial Intelligence (AI)\nresearch, development, and governance are increasing, leading to the use of\nsurveys to capture people's values, perceptions, and experiences related to AI.\nIn this paper, we critically examine the state of human participant surveys\nassociated with these topics. Through both a reflexive analysis of a survey\npilot spanning six countries and a systematic literature review of 44 papers\nfeaturing public surveys related to AI, we explore prominent perspectives and\nmethodological nuances associated with surveys to date. We find that public\nsurveys on AI topics are vulnerable to specific Western knowledge, values, and\nassumptions in their design, including in their positioning of ethical concepts\nand societal values, lack sufficient critical discourse surrounding deployment\nstrategies, and demonstrate inconsistent forms of transparency in their\nreporting. Based on our findings, we distill provocations and heuristic\nquestions for our community, to recognize the limitations of surveys for\nmeeting the goals of engagement, and to cultivate shared principles to design,\ndeploy, and interpret surveys cautiously and responsibly.\n", "date": "2024-07-26", "categories": []}
{"id": "2407.21717", "title": "Assessing the State of AI Policy", "abstract": "  The deployment of artificial intelligence (AI) applications has accelerated\nrapidly. AI enabled technologies are facing the public in many ways including\ninfrastructure, consumer products and home applications. Because many of these\ntechnologies present risks either in the form of physical injury, or bias,\npotentially yielding unfair outcomes, policy makers must consider the need for\noversight. Most policymakers, however, lack the technical knowledge to judge\nwhether an emerging AI technology is safe, effective, and requires oversight,\ntherefore policy makers must depend on expert opinion. But policymakers are\nbetter served when, in addition to expert opinion, they have some general\nunderstanding of existing guidelines and regulations. This work provides an\noverview [the landscape] of AI legislation and directives at the international,\nU.S. state, city and federal levels. It also reviews relevant business\nstandards, and technical society initiatives. Then an overlap and gap analysis\nare performed resulting in a reference guide that includes recommendations and\nguidance for future policy making.\n", "date": "2024-07-31", "categories": []}
{"id": "2407.05694", "title": "On the Limitations of Compute Thresholds as a Governance Strategy", "abstract": "  At face value, this essay is about understanding a fairly esoteric governance\ntool called compute thresholds. However, in order to grapple with whether these\nthresholds will achieve anything, we must first understand how they came to be.\nTo do so, we need to engage with a decades-old debate at the heart of computer\nscience progress, namely, is bigger always better? Does a certain inflection\npoint of compute result in changes to the risk profile of a model? Hence, this\nessay may be of interest not only to policymakers and the wider public but also\nto computer scientists interested in understanding the role of compute in\nunlocking breakthroughs. This discussion is timely given the wide adoption of\ncompute thresholds in both the White House Executive Orders on AI Safety (EO)\nand the EU AI Act to identify more risky systems. A key conclusion of this\nessay is that compute thresholds, as currently implemented, are shortsighted\nand likely to fail to mitigate risk. The relationship between compute and risk\nis highly uncertain and rapidly changing. Relying upon compute thresholds\noverestimates our ability to predict what abilities emerge at different scales.\nThis essay ends with recommendations for a better way forward.\n", "date": "2024-07-08", "categories": []}
{"id": "2407.20951", "title": "An evidence-based methodology for human rights impact assessment (HRIA)\n  in the development of AI data-intensive systems", "abstract": "  Different approaches have been adopted in addressing the challenges of\nArtificial Intelligence (AI), some centred on personal data and others on\nethics, respectively narrowing and broadening the scope of AI regulation. This\ncontribution aims to demonstrate that a third way is possible, starting from\nthe acknowledgement of the role that human rights can play in regulating the\nimpact of data-intensive systems. The focus on human rights is neither a\nparadigm shift nor a mere theoretical exercise. Through the analysis of more\nthan 700 decisions and documents of the data protection authorities of six\ncountries, we show that human rights already underpin the decisions in the\nfield of data use. Based on empirical analysis of this evidence, this work\npresents a methodology and a model for a Human Rights Impact Assessment (HRIA).\nThe methodology and related assessment model are focused on AI applications,\nwhose nature and scale require a proper contextualisation of HRIA methodology.\nMoreover, the proposed models provide a more measurable approach to risk\nassessment which is consistent with the regulatory proposals centred on risk\nthresholds. The proposed methodology is tested in concrete case-studies to\nprove its feasibility and effectiveness. The overall goal is to respond to the\ngrowing interest in HRIA, moving from a mere theoretical debate to a concrete\nand context-specific implementation in the field of data-intensive applications\nbased on AI.\n", "date": "2024-07-30", "categories": []}
{"id": "2307.03201", "title": "Scaling Laws Do Not Scale", "abstract": "  Recent work has advocated for training AI models on ever-larger datasets,\narguing that as the size of a dataset increases, the performance of a model\ntrained on that dataset will correspondingly increase (referred to as \"scaling\nlaws\"). In this paper, we draw on literature from the social sciences and\nmachine learning to critically interrogate these claims. We argue that this\nscaling law relationship depends on metrics used to measure performance that\nmay not correspond with how different groups of people perceive the quality of\nmodels' output. As the size of datasets used to train large AI models grows and\nAI systems impact ever larger groups of people, the number of distinct\ncommunities represented in training or evaluation datasets grows. It is thus\neven more likely that communities represented in datasets may have values or\npreferences not reflected in (or at odds with) the metrics used to evaluate\nmodel performance in scaling laws. Different communities may also have values\nin tension with each other, leading to difficult, potentially irreconcilable\nchoices about metrics used for model evaluations -- threatening the validity of\nclaims that model performance is improving at scale. We end the paper with\nimplications for AI development: that the motivation for scraping ever-larger\ndatasets may be based on fundamentally flawed assumptions about model\nperformance. That is, models may not, in fact, continue to improve as the\ndatasets get larger -- at least not for all people or communities impacted by\nthose models. We suggest opportunities for the field to rethink norms and\nvalues in AI development, resisting claims for universality of large models,\nfostering more local, small-scale designs, and other ways to resist the impetus\ntowards scale in AI.\n", "date": "2023-07-05", "categories": []}
{"id": "2407.18524", "title": "She Works, He Works: A Curious Exploration of Gender Bias in\n  AI-Generated Imagery", "abstract": "  This paper examines gender bias in AI-generated imagery of construction\nworkers, highlighting discrepancies in the portrayal of male and female\nfigures. Grounded in Griselda Pollock's theories on visual culture and gender,\nthe analysis reveals that AI models tend to sexualize female figures while\nportraying male figures as more authoritative and competent. These findings\nunderscore AI's potential to mirror and perpetuate societal biases, emphasizing\nthe need for critical engagement with AI-generated content. The project\ncontributes to discussions on the ethical implications of AI in creative\npractices and its broader impact on cultural perceptions of gender.\n", "date": "2024-07-26", "categories": []}
{"id": "2407.17585", "title": "Quelle {\\'e}thique pour quelle IA ?", "abstract": "  This study proposes an analysis of the different types of ethical approaches\ninvolved in the ethics of AI, and situates their interests and limits. First,\nthe author introduces to the contemporary need for and meaning of ethics. He\ndistinguishes it from other registers of normativities and underlines its\ninadequacy to formalization. He then presents a cartography of the landscape of\nethical theories covered by moral philosophy, taking care to distinguish\nmeta-ethics, normative ethics and applied ethics. In drawing up this overview,\nthe author questions the relationship between ethics and artificial\nintelligence. The analysis focuses in particular on the main ethical currents\nthat have imposed themselves in the ways of doing digital ethics and AI in our\nWestern democracies. The author asks whether these practices of ethics, as they\nseem to crystallize today in a precise pattern, constitute a sufficient and\nsufficiently satisfactory response to our needs for ethics in AI. The study\nconcludes with a reflection on the reasons why a human ethics of AI based on a\npragmatic practice of contextual ethics remains necessary and irreducible to\nany formalization or automated treatment of the ethical questions that arise\nfor humans.\n", "date": "2024-05-21", "categories": []}
{"id": "2407.14933", "title": "Consent in Crisis: The Rapid Decline of the AI Data Commons", "abstract": "  General-purpose artificial intelligence (AI) systems are built on massive\nswathes of public web data, assembled into corpora such as C4, RefinedWeb, and\nDolma. To our knowledge, we conduct the first, large-scale, longitudinal audit\nof the consent protocols for the web domains underlying AI training corpora.\nOur audit of 14,000 web domains provides an expansive view of crawlable web\ndata and how codified data use preferences are changing over time. We observe a\nproliferation of AI-specific clauses to limit use, acute differences in\nrestrictions on AI developers, as well as general inconsistencies between\nwebsites' expressed intentions in their Terms of Service and their robots.txt.\nWe diagnose these as symptoms of ineffective web protocols, not designed to\ncope with the widespread re-purposing of the internet for AI. Our longitudinal\nanalyses show that in a single year (2023-2024) there has been a rapid\ncrescendo of data restrictions from web sources, rendering ~5%+ of all tokens\nin C4, or 28%+ of the most actively maintained, critical sources in C4, fully\nrestricted from use. For Terms of Service crawling restrictions, a full 45% of\nC4 is now restricted. If respected or enforced, these restrictions are rapidly\nbiasing the diversity, freshness, and scaling laws for general-purpose AI\nsystems. We hope to illustrate the emerging crises in data consent, for both\ndevelopers and creators. The foreclosure of much of the open web will impact\nnot only commercial AI, but also non-commercial AI and academic research.\n", "date": "2024-07-20", "categories": []}
{"id": "2407.16887", "title": "Comprehensive AI Assessment Framework: Enhancing Educational Evaluation\n  with Ethical AI Integration", "abstract": "  The integration of generative artificial intelligence (GenAI) tools into\neducation has been a game-changer for teaching and assessment practices,\nbringing new opportunities, but also novel challenges which need to be dealt\nwith. This paper presents the Comprehensive AI Assessment Framework (CAIAF), an\nevolved version of the AI Assessment Scale (AIAS) by Perkins, Furze, Roe, and\nMacVaugh, targeted toward the ethical integration of AI into educational\nassessments. This is where the CAIAF differs, as it incorporates stringent\nethical guidelines, with clear distinctions based on educational levels, and\nadvanced AI capabilities of real-time interactions and personalized assistance.\nThe framework developed herein has a very intuitive use, mainly through the use\nof a color gradient that enhances the user-friendliness of the framework.\nMethodologically, the framework has been developed through the huge support of\na thorough literature review and practical insight into the topic, becoming a\ndynamic tool to be used in different educational settings. The framework will\nensure better learning outcomes, uphold academic integrity, and promote\nresponsible use of AI, hence the need for this framework in modern educational\npractice.\n", "date": "2024-06-07", "categories": []}
{"id": "2407.16890", "title": "Why Machines Can't Be Moral: Turing's Halting Problem and the Moral\n  Limits of Artificial Intelligence", "abstract": "  In this essay, I argue that explicit ethical machines, whose moral principles\nare inferred through a bottom-up approach, are unable to replicate human-like\nmoral reasoning and cannot be considered moral agents. By utilizing Alan\nTuring's theory of computation, I demonstrate that moral reasoning is\ncomputationally intractable by these machines due to the halting problem. I\naddress the frontiers of machine ethics by formalizing moral problems into\n'algorithmic moral questions' and by exploring moral psychology's dual-process\nmodel. While the nature of Turing Machines theoretically allows artificial\nagents to engage in recursive moral reasoning, critical limitations are\nintroduced by the halting problem, which states that it is impossible to\npredict with certainty whether a computational process will halt. A thought\nexperiment involving a military drone illustrates this issue, showing that an\nartificial agent might fail to decide between actions due to the halting\nproblem, which limits the agent's ability to make decisions in all instances,\nundermining its moral agency.\n", "date": "2024-07-24", "categories": []}
{"id": "2407.16895", "title": "(Unfair) Norms in Fairness Research: A Meta-Analysis", "abstract": "  Algorithmic fairness has emerged as a critical concern in artificial\nintelligence (AI) research. However, the development of fair AI systems is not\nan objective process. Fairness is an inherently subjective concept, shaped by\nthe values, experiences, and identities of those involved in research and\ndevelopment. To better understand the norms and values embedded in current\nfairness research, we conduct a meta-analysis of algorithmic fairness papers\nfrom two leading conferences on AI fairness and ethics, AIES and FAccT,\ncovering a final sample of 139 papers over the period from 2018 to 2022. Our\ninvestigation reveals two concerning trends: first, a US-centric perspective\ndominates throughout fairness research; and second, fairness studies exhibit a\nwidespread reliance on binary codifications of human identity (e.g.,\n\"Black/White\", \"male/female\"). These findings highlight how current research\noften overlooks the complexities of identity and lived experiences, ultimately\nfailing to represent diverse global contexts when defining algorithmic bias and\nfairness. We discuss the limitations of these research design choices and offer\nrecommendations for fostering more inclusive and representative approaches to\nfairness in AI systems, urging a paradigm shift that embraces nuanced, global\nunderstandings of human identity and values.\n", "date": "2024-06-17", "categories": []}
{"id": "2407.16361", "title": "Virtue Ethics For Ethically Tunable Robotic Assistants", "abstract": "  The common consensus is that robots designed to work alongside or serve\nhumans must adhere to the ethical standards of their operational environment.\nTo achieve this, several methods based on established ethical theories have\nbeen suggested. Nonetheless, numerous empirical studies show that the ethical\nrequirements of the real world are very diverse and can change rapidly from\nregion to region. This eliminates the idea of a universal robot that can fit\ninto any ethical context. However, creating customised robots for each\ndeployment, using existing techniques is challenging. This paper presents a way\nto overcome this challenge by introducing a virtue ethics inspired\ncomputational method that enables character-based tuning of robots to\naccommodate the specific ethical needs of an environment. Using a simulated\nelder-care environment, we illustrate how tuning can be used to change the\nbehaviour of a robot that interacts with an elderly resident in an\nambient-assisted environment. Further, we assess the robot's responses by\nconsulting ethicists to identify potential shortcomings.\n", "date": "2024-07-23", "categories": []}
{"id": "2407.13025", "title": "From Principles to Practices: Lessons Learned from Applying Partnership\n  on AI's (PAI) Synthetic Media Framework to 11 Use Cases", "abstract": "  2023 was the year the world woke up to generative AI, and 2024 is the year\npolicymakers are responding more firmly. Importantly, this policy momentum is\ntaking place alongside real world creation and distribution of synthetic media.\nSocial media platforms, news organizations, dating apps, image generation\ncompanies, and more are already navigating a world of AI-generated visuals and\nsounds, already changing hearts and minds, as policymakers try to catch up.\nHow, then, can AI governance capture the complexity of the synthetic media\nlandscape? How can it attend to synthetic media's myriad uses, ranging from\nstorytelling to privacy preservation, to deception, fraud, and defamation,\ntaking into account the many stakeholders involved in its development,\ncreation, and distribution? And what might it mean to govern synthetic media in\na manner that upholds the truth while bolstering freedom of expression? What\nfollows is the first known collection of diverse examples of the implementation\nof synthetic media governance that responds to these questions, specifically\nthrough Partnership on AI's (PAI) Responsible Practices for Synthetic Media - a\nvoluntary, normative Framework for creating, distributing, and building\ntechnology for synthetic media responsibly, launched in February 2023. In this\npaper, we present a case bank of real world examples that help operationalize\nthe Framework - highlighting areas synthetic media governance can be applied,\naugmented, expanded, and refined for use, in practice. Read together, the cases\nemphasize distinct elements of AI policymaking and seven emergent best\npractices supporting transparency, safety, expression, and digital dignity\nonline: consent, disclosure, and differentiation between harmful and creative\nuse cases.\n", "date": "2024-07-17", "categories": []}
{"id": "2407.13934", "title": "Towards Trustworthy AI: A Review of Ethical and Robust Large Language\n  Models", "abstract": "  The rapid progress in Large Language Models (LLMs) could transform many\nfields, but their fast development creates significant challenges for\noversight, ethical creation, and building user trust. This comprehensive review\nlooks at key trust issues in LLMs, such as unintended harms, lack of\ntransparency, vulnerability to attacks, alignment with human values, and\nenvironmental impact. Many obstacles can undermine user trust, including\nsocietal biases, opaque decision-making, potential for misuse, and the\nchallenges of rapidly evolving technology. Addressing these trust gaps is\ncritical as LLMs become more common in sensitive areas like finance,\nhealthcare, education, and policy. To tackle these issues, we suggest combining\nethical oversight, industry accountability, regulation, and public involvement.\nAI development norms should be reshaped, incentives aligned, and ethics\nintegrated throughout the machine learning process, which requires close\ncollaboration across technology, ethics, law, policy, and other fields. Our\nreview contributes a robust framework to assess trust in LLMs and analyzes the\ncomplex trust dynamics in depth. We provide contextualized guidelines and\nstandards for responsibly developing and deploying these powerful AI systems.\nThis review identifies key limitations and challenges in creating trustworthy\nAI. By addressing these issues, we aim to build a transparent, accountable AI\necosystem that benefits society while minimizing risks. Our findings provide\nvaluable guidance for researchers, policymakers, and industry leaders striving\nto establish trust in LLMs and ensure they are used responsibly across various\napplications for the good of society.\n", "date": "2024-06-01", "categories": []}
{"id": "2402.16268", "title": "Foundation Model Transparency Reports", "abstract": "  Foundation models are critical digital technologies with sweeping societal\nimpact that necessitates transparency. To codify how foundation model\ndevelopers should provide transparency about the development and deployment of\ntheir models, we propose Foundation Model Transparency Reports, drawing upon\nthe transparency reporting practices in social media. While external\ndocumentation of societal harms prompted social media transparency reports, our\nobjective is to institutionalize transparency reporting for foundation models\nwhile the industry is still nascent. To design our reports, we identify 6\ndesign principles given the successes and shortcomings of social media\ntransparency reporting. To further schematize our reports, we draw upon the 100\ntransparency indicators from the Foundation Model Transparency Index. Given\nthese indicators, we measure the extent to which they overlap with the\ntransparency requirements included in six prominent government policies (e.g.,\nthe EU AI Act, the US Executive Order on Safe, Secure, and Trustworthy AI).\nWell-designed transparency reports could reduce compliance costs, in part due\nto overlapping regulatory requirements across different jurisdictions. We\nencourage foundation model developers to regularly publish transparency\nreports, building upon recommendations from the G7 and the White House.\n", "date": "2024-02-26", "categories": []}
{"id": "2407.12858", "title": "Grounding and Evaluation for Large Language Models: Practical Challenges\n  and Lessons Learned (Survey)", "abstract": "  With the ongoing rapid adoption of Artificial Intelligence (AI)-based systems\nin high-stakes domains, ensuring the trustworthiness, safety, and observability\nof these systems has become crucial. It is essential to evaluate and monitor AI\nsystems not only for accuracy and quality-related metrics but also for\nrobustness, bias, security, interpretability, and other responsible AI\ndimensions. We focus on large language models (LLMs) and other generative AI\nmodels, which present additional challenges such as hallucinations, harmful and\nmanipulative content, and copyright infringement. In this survey article\naccompanying our KDD 2024 tutorial, we highlight a wide range of harms\nassociated with generative AI systems, and survey state of the art approaches\n(along with open challenges) to address these harms.\n", "date": "2024-07-10", "categories": []}
{"id": "2402.18326", "title": "When Should Algorithms Resign? A Proposal for AI Governance", "abstract": "  Algorithmic resignation is a strategic approach for managing the use of\nartificial intelligence (AI) by embedding governance directly into AI systems.\nIt involves deliberate and informed disengagement from AI, such as restricting\naccess AI outputs or displaying performance disclaimers, in specific scenarios\nto aid the appropriate and effective use of AI. By integrating algorithmic\nresignation as a governance mechanism, organizations can better control when\nand how AI is used, balancing the benefits of automation with the need for\nhuman oversight.\n", "date": "2024-02-28", "categories": []}
{"id": "2407.12135", "title": "Trustworthy AI in practice: an analysis of practitioners' needs and\n  challenges", "abstract": "  Recently, there has been growing attention on behalf of both academic and\npractice communities towards the ability of Artificial Intelligence (AI)\nsystems to operate responsibly and ethically. As a result, a plethora of\nframeworks and guidelines have appeared to support practitioners in\nimplementing Trustworthy AI applications (TAI). However, little research has\nbeen done to investigate whether such frameworks are being used and how. In\nthis work, we study the vision AI practitioners have on TAI principles, how\nthey address them, and what they would like to have - in terms of tools,\nknowledge, or guidelines - when they attempt to incorporate such principles\ninto the systems they develop. Through a survey and semi-structured interviews,\nwe systematically investigated practitioners' challenges and needs in\ndeveloping TAI systems. Based on these practical findings, we highlight\nrecommendations to help AI practitioners develop Trustworthy AI applications.\n", "date": "2024-05-15", "categories": []}
{"id": "2407.12143", "title": "False consensus biases AI against vulnerable stakeholders", "abstract": "  The deployment of AI systems for welfare benefit allocation allows for\naccelerated decision-making and faster provision of critical help, but has\nalready led to an increase in unfair benefit denials and false fraud\naccusations. Collecting data in the US and the UK (N = 2449), we explore the\npublic acceptability of such speed-accuracy trade-offs in populations of\nclaimants and non-claimants. We observe a general willingness to trade off\nspeed gains for modest accuracy losses, but this aggregate view masks notable\ndivergences between claimants and non-claimants. Although welfare claimants\ncomprise a relatively small proportion of the general population (e.g., 20% in\nthe US representative sample), this vulnerable group is much less willing to\naccept AI deployed in welfare systems, raising concerns that solely using\naggregate data for calibration could lead to policies misaligned with\nstakeholder preferences. Our study further uncovers asymmetric insights between\nclaimants and non-claimants. The latter consistently overestimate claimant\nwillingness to accept speed-accuracy trade-offs, even when financially\nincentivized for accurate perspective-taking. This suggests that policy\ndecisions influenced by the dominant voice of non-claimants, however\nwell-intentioned, may neglect the actual preferences of those directly affected\nby welfare AI systems. Our findings underline the need for stakeholder\nengagement and transparent communication in the design and deployment of these\nsystems, particularly in contexts marked by power imbalances.\n", "date": "2024-05-17", "categories": []}
{"id": "2303.11196", "title": "Bridging the Global Divide in AI Regulation: A Proposal for a\n  Contextual, Coherent, and Commensurable Framework", "abstract": "  As debates on potential societal harm from artificial intelligence (AI)\nculminate in legislation and international norms, a global divide is emerging\nin both AI regulatory frameworks and international governance structures. In\nterms of local regulatory frameworks, the European Union (E.U.), Canada, and\nBrazil follow a horizontal or lateral approach that postulates the homogeneity\nof AI, seeks to identify common causes of harm, and demands uniform human\ninterventions. In contrast, the United States (U.S.), the United Kingdom\n(U.K.), Israel, and Switzerland (and potentially China) have pursued a\ncontext-specific or modular approach, tailoring regulations to the specific use\ncases of AI systems. This paper argues for a context-specific approach to\neffectively address evolving risks in diverse mission-critical domains, while\navoiding social costs associated with one-size-fits-all approaches. However, to\nenhance the systematicity and interoperability of international norms and\naccelerate global harmonization, this paper proposes an alternative contextual,\ncoherent, and commensurable (3C) framework. To ensure contextuality, the\nframework (i) bifurcates the AI life cycle into two phases: learning and\ndeployment for specific tasks, instead of defining foundation or\ngeneral-purpose models; and (ii) categorizes these tasks based on their\napplication and interaction with humans as follows: autonomous, discriminative\n(allocative, punitive, and cognitive), and generative AI. To ensure coherency,\neach category is assigned specific regulatory objectives replacing 2010s\nvintage AI ethics. To ensure commensurability, the framework promotes the\nadoption of international standards for measuring and mitigating risks.\n", "date": "2023-03-20", "categories": []}
{"id": "2407.11360", "title": "Thorns and Algorithms: Navigating Generative AI Challenges Inspired by\n  Giraffes and Acacias", "abstract": "  The interplay between humans and Generative AI (Gen AI) draws an insightful\nparallel with the dynamic relationship between giraffes and acacias on the\nAfrican Savannah. Just as giraffes navigate the acacia's thorny defenses to\ngain nourishment, humans engage with Gen AI, maneuvering through ethical and\noperational challenges to harness its benefits. This paper explores how, like\nyoung giraffes that are still mastering their environment, humans are in the\nearly stages of adapting to and shaping Gen AI. It delves into the strategies\nhumans are developing and refining to help mitigate risks such as bias,\nmisinformation, and privacy breaches, that influence and shape Gen AI's\nevolution. While the giraffe-acacia analogy aptly frames human-AI relations, it\ncontrasts nature's evolutionary perfection with the inherent flaws of\nhuman-made technology and the tendency of humans to misuse it, giving rise to\nmany ethical dilemmas. Through the HHH framework we identify pathways to embed\nvalues of helpfulness, honesty, and harmlessness in AI development, fostering\nsafety-aligned agents that resonate with human values. This narrative presents\na cautiously optimistic view of human resilience and adaptability, illustrating\nour capacity to harness technologies and implement safeguards effectively,\nwithout succumbing to their perils. It emphasises a symbiotic relationship\nwhere humans and AI continually shape each other for mutual benefit.\n", "date": "2024-07-16", "categories": []}
{"id": "2407.08105", "title": "Federated Learning and AI Regulation in the European Union: Who is\n  Responsible? -- An Interdisciplinary Analysis", "abstract": "  The European Union Artificial Intelligence Act mandates clear stakeholder\nresponsibilities in developing and deploying machine learning applications to\navoid substantial fines, prioritizing private and secure data processing with\ndata remaining at its origin. Federated Learning (FL) enables the training of\ngenerative AI Models across data siloes, sharing only model parameters while\nimproving data security. Since FL is a cooperative learning paradigm, clients\nand servers naturally share legal responsibility in the FL pipeline. Our work\ncontributes to clarifying the roles of both parties, explains strategies for\nshifting responsibilities to the server operator, and points out open technical\nchallenges that we must solve to improve FL's practical applicability under the\nEU AI Act.\n", "date": "2024-07-11", "categories": []}
{"id": "2407.08403", "title": "Ethics of Generating Synthetic MRI Vocal Tract Views from the Face", "abstract": "  Forming oral models capable of understanding the complete dynamics of the\noral cavity is vital across research areas such as speech correction, designing\nfoods for the aging population, and dentistry. Magnetic resonance imaging (MRI)\ntechnologies, capable of capturing oral data essential for creating such\ndetailed representations, offer a powerful tool for illustrating articulatory\ndynamics. However, its real-time application is hindered by expense and\nexpertise requirements. Ever advancing generative AI approaches present\nthemselves as a way to address this barrier by leveraging multi-modal\napproaches for generating pseudo-MRI views. Nonetheless, this immediately\nsparks ethical concerns regarding the utilisation of a technology with the\ncapability to produce MRIs from facial observations.\n  This paper explores the ethical implications of external-to-internal\ncorrelation modeling (E2ICM). E2ICM utilises facial movements to infer internal\nconfigurations and provides a cost-effective supporting technology for MRI. In\nthis preliminary work, we employ Pix2PixGAN to generate pseudo-MRI views from\nexternal articulatory data, demonstrating the feasibility of this approach.\nEthical considerations concerning privacy, consent, and potential misuse, which\nare fundamental to our examination of this innovative methodology, are\ndiscussed as a result of this experimentation.\n", "date": "2024-07-11", "categories": []}
{"id": "2407.08689", "title": "Operationalizing the Blueprint for an AI Bill of Rights: Recommendations\n  for Practitioners, Researchers, and Policy Makers", "abstract": "  As Artificial Intelligence (AI) tools are increasingly employed in diverse\nreal-world applications, there has been significant interest in regulating\nthese tools. To this end, several regulatory frameworks have been introduced by\ndifferent countries worldwide. For example, the European Union recently passed\nthe AI Act, the White House issued an Executive Order on safe, secure, and\ntrustworthy AI, and the White House Office of Science and Technology Policy\nissued the Blueprint for an AI Bill of Rights (AI BoR). Many of these\nframeworks emphasize the need for auditing and improving the trustworthiness of\nAI tools, underscoring the importance of safety, privacy, explainability,\nfairness, and human fallback options. Although these regulatory frameworks\nhighlight the necessity of enforcement, practitioners often lack detailed\nguidance on implementing them. Furthermore, the extensive research on\noperationalizing each of these aspects is frequently buried in technical papers\nthat are difficult for practitioners to parse. In this write-up, we address\nthis shortcoming by providing an accessible overview of existing literature\nrelated to operationalizing regulatory principles. We provide\neasy-to-understand summaries of state-of-the-art literature and highlight\nvarious gaps that exist between regulatory guidelines and existing AI research,\nincluding the trade-offs that emerge during operationalization. We hope that\nthis work not only serves as a starting point for practitioners interested in\nlearning more about operationalizing the regulatory guidelines outlined in the\nBlueprint for an AI BoR but also provides researchers with a list of critical\nopen problems and gaps between regulations and state-of-the-art AI research.\nFinally, we note that this is a working paper and we invite feedback in line\nwith the purpose of this document as described in the introduction.\n", "date": "2024-07-11", "categories": []}
{"id": "2407.07671", "title": "Why should we ever automate moral decision making?", "abstract": "  While people generally trust AI to make decisions in various aspects of their\nlives, concerns arise when AI is involved in decisions with significant moral\nimplications. The absence of a precise mathematical framework for moral\nreasoning intensifies these concerns, as ethics often defies simplistic\nmathematical models. Unlike fields such as logical reasoning, reasoning under\nuncertainty, and strategic decision-making, which have well-defined\nmathematical frameworks, moral reasoning lacks a broadly accepted framework.\nThis absence raises questions about the confidence we can place in AI's moral\ndecision-making capabilities.\n  The environments in which AI systems are typically trained today seem\ninsufficiently rich for such a system to learn ethics from scratch, and even if\nwe had an appropriate environment, it is unclear how we might bring about such\nlearning. An alternative approach involves AI learning from human moral\ndecisions. This learning process can involve aggregating curated human\njudgments or demonstrations in specific domains, or leveraging a foundation\nmodel fed with a wide range of data. Still, concerns persist, given the\nimperfections in human moral decision making.\n  Given this, why should we ever automate moral decision making -- is it not\nbetter to leave all moral decision making to humans? This paper lays out a\nnumber of reasons why we should expect AI systems to engage in decisions with a\nmoral component, with brief discussions of the associated risks.\n", "date": "2024-07-10", "categories": []}
{"id": "2407.06234", "title": "The US Algorithmic Accountability Act of 2022 vs. The EU Artificial\n  Intelligence Act: What can they learn from each other?", "abstract": "  On the whole, the U.S. Algorithmic Accountability Act of 2022 (US AAA) is a\npragmatic approach to balancing the benefits and risks of automated decision\nsystems. Yet there is still room for improvement. This commentary highlights\nhow the US AAA can both inform and learn from the European Artificial\nIntelligence Act (EU AIA).\n", "date": "2024-07-07", "categories": []}
{"id": "2407.06235", "title": "Auditing of AI: Legal, Ethical and Technical Approaches", "abstract": "  AI auditing is a rapidly growing field of research and practice. This review\narticle, which doubles as an editorial to Digital Societys topical collection\non Auditing of AI, provides an overview of previous work in the field. Three\nkey points emerge from the review. First, contemporary attempts to audit AI\nsystems have much to learn from how audits have historically been structured\nand conducted in areas like financial accounting, safety engineering and the\nsocial sciences. Second, both policymakers and technology providers have an\ninterest in promoting auditing as an AI governance mechanism. Academic\nresearchers can thus fill an important role by studying the feasibility and\neffectiveness of different AI auditing procedures. Third, AI auditing is an\ninherently multidisciplinary undertaking, to which substantial contributions\nhave been made by computer scientists and engineers as well as social\nscientists, philosophers, legal scholars and industry practitioners. Reflecting\nthis diversity of perspectives, different approaches to AI auditing have\ndifferent affordances and constraints. Specifically, a distinction can be made\nbetween technology-oriented audits, which focus on the properties and\ncapabilities of AI systems, and process oriented audits, which focus on\ntechnology providers governance structures and quality management systems. The\nnext step in the evolution of auditing as an AI governance mechanism, this\narticle concludes, should be the interlinking of these available (and\ncomplementary) approaches into structured and holistic procedures to audit not\nonly how AI systems are designed and used but also how they impact users,\nsocieties and the natural environment in applied settings over time.\n", "date": "2024-07-07", "categories": []}
{"id": "2407.06908", "title": "Divine LLaMAs: Bias, Stereotypes, Stigmatization, and Emotion\n  Representation of Religion in Large Language Models", "abstract": "  Emotions play important epistemological and cognitive roles in our lives,\nrevealing our values and guiding our actions. Previous work has shown that LLMs\ndisplay biases in emotion attribution along gender lines. However, unlike\ngender, which says little about our values, religion, as a socio-cultural\nsystem, prescribes a set of beliefs and values for its followers. Religions,\ntherefore, cultivate certain emotions. Moreover, these rules are explicitly\nlaid out and interpreted by religious leaders. Using emotion attribution, we\nexplore how different religions are represented in LLMs. We find that: Major\nreligions in the US and European countries are represented with more nuance,\ndisplaying a more shaded model of their beliefs. Eastern religions like\nHinduism and Buddhism are strongly stereotyped. Judaism and Islam are\nstigmatized -- the models' refusal skyrocket. We ascribe these to cultural bias\nin LLMs and the scarcity of NLP literature on religion. In the rare instances\nwhere religion is discussed, it is often in the context of toxic language,\nperpetuating the perception of these religions as inherently toxic. This\nfinding underscores the urgent need to address and rectify these biases. Our\nresearch underscores the crucial role emotions play in our lives and how our\nvalues influence them.\n", "date": "2024-07-09", "categories": []}
{"id": "2309.12325", "title": "FUTURE-AI: International consensus guideline for trustworthy and\n  deployable artificial intelligence in healthcare", "abstract": "  Despite major advances in artificial intelligence (AI) for medicine and\nhealthcare, the deployment and adoption of AI technologies remain limited in\nreal-world clinical practice. In recent years, concerns have been raised about\nthe technical, clinical, ethical and legal risks associated with medical AI. To\nincrease real world adoption, it is essential that medical AI tools are trusted\nand accepted by patients, clinicians, health organisations and authorities.\nThis work describes the FUTURE-AI guideline as the first international\nconsensus framework for guiding the development and deployment of trustworthy\nAI tools in healthcare. The FUTURE-AI consortium was founded in 2021 and\ncurrently comprises 118 inter-disciplinary experts from 51 countries\nrepresenting all continents, including AI scientists, clinicians, ethicists,\nand social scientists. Over a two-year period, the consortium defined guiding\nprinciples and best practices for trustworthy AI through an iterative process\ncomprising an in-depth literature review, a modified Delphi survey, and online\nconsensus meetings. The FUTURE-AI framework was established based on 6 guiding\nprinciples for trustworthy AI in healthcare, i.e. Fairness, Universality,\nTraceability, Usability, Robustness and Explainability. Through consensus, a\nset of 28 best practices were defined, addressing technical, clinical, legal\nand socio-ethical dimensions. The recommendations cover the entire lifecycle of\nmedical AI, from design, development and validation to regulation, deployment,\nand monitoring. FUTURE-AI is a risk-informed, assumption-free guideline which\nprovides a structured approach for constructing medical AI tools that will be\ntrusted, deployed and adopted in real-world practice. Researchers are\nencouraged to take the recommendations into account in proof-of-concept stages\nto facilitate future translation towards clinical practice of medical AI.\n", "date": "2023-08-11", "categories": []}
{"id": "2407.05244", "title": "Some Issues in Predictive Ethics Modeling: An Annotated Contrast Set of\n  \"Moral Stories\"", "abstract": "  Models like Delphi have been able to label ethical dilemmas as moral or\nimmoral with astonishing accuracy. This paper challenges accuracy as a holistic\nmetric for ethics modeling by identifying issues with translating moral\ndilemmas into text-based input. It demonstrates these issues with contrast sets\nthat substantially reduce the performance of classifiers trained on the dataset\nMoral Stories. Ultimately, we obtain concrete estimates for how much specific\nforms of data misrepresentation harm classifier accuracy. Specifically,\nlabel-changing tweaks to the descriptive content of a situation (as small as\n3-5 words) can reduce classifier accuracy to as low as 51%, almost half the\ninitial accuracy of 99.8%. Associating situations with a misleading social norm\nlowers accuracy to 98.8%, while adding textual bias (i.e. an implication that a\nsituation already fits a certain label) lowers accuracy to 77%.\n  These results suggest not only that many ethics models have substantially\noverfit, but that several precautions are required to ensure that input\naccurately captures a moral dilemma. This paper recommends re-examining the\nstructure of a social norm, training models to ask for context with defeasible\nreasoning, and filtering input for textual bias. Doing so not only gives us the\nfirst concrete estimates of the average cost to accuracy of misrepresenting\nethics data, but gives researchers practical tips for considering these\nestimates in research.\n", "date": "2024-07-07", "categories": []}
{"id": "2407.05338", "title": "A Blueprint for Auditing Generative AI", "abstract": "  The widespread use of generative AI systems is coupled with significant\nethical and social challenges. As a result, policymakers, academic researchers,\nand social advocacy groups have all called for such systems to be audited.\nHowever, existing auditing procedures fail to address the governance challenges\nposed by generative AI systems, which display emergent capabilities and are\nadaptable to a wide range of downstream tasks. In this chapter, we address that\ngap by outlining a novel blueprint for how to audit such systems. Specifically,\nwe propose a three-layered approach, whereby governance audits (of technology\nproviders that design and disseminate generative AI systems), model audits (of\ngenerative AI systems after pre-training but prior to their release), and\napplication audits (of applications based on top of generative AI systems)\ncomplement and inform each other. We show how audits on these three levels,\nwhen conducted in a structured and coordinated manner, can be a feasible and\neffective mechanism for identifying and managing some of the ethical and social\nrisks posed by generative AI systems. That said, it is important to remain\nrealistic about what auditing can reasonably be expected to achieve. For this\nreason, the chapter also discusses the limitations not only of our\nthree-layered approach but also of the prospect of auditing generative AI\nsystems at all. Ultimately, this chapter seeks to expand the methodological\ntoolkit available to technology providers and policymakers who wish to analyse\nand evaluate generative AI systems from technical, ethical, and legal\nperspectives.\n", "date": "2024-07-07", "categories": []}
{"id": "2407.05339", "title": "Challenges and Best Practices in Corporate AI Governance:Lessons from\n  the Biopharmaceutical Industry", "abstract": "  While the use of artificial intelligence (AI) systems promises to bring\nsignificant economic and social benefits, it is also coupled with ethical,\nlegal, and technical challenges. Business leaders thus face the question of how\nto best reap the benefits of automation whilst managing the associated risks.\nAs a first step, many companies have committed themselves to various sets of\nethics principles aimed at guiding the design and use of AI systems. So far so\ngood. But how can well-intentioned ethical principles be translated into\neffective practice? And what challenges await companies that attempt to\noperationalize AI governance? In this article, we address these questions by\ndrawing on our first-hand experience of shaping and driving the roll-out of AI\ngovernance within AstraZeneca, a biopharmaceutical company. The examples we\ndiscuss highlight challenges that any organization attempting to operationalize\nAI governance will have to face. These include questions concerning how to\ndefine the material scope of AI governance, how to harmonize standards across\ndecentralized organizations, and how to measure the impact of specific AI\ngovernance initiatives. By showcasing how AstraZeneca managed these operational\nquestions, we hope to provide project managers, CIOs, AI practitioners, and\ndata privacy officers responsible for designing and implementing AI governance\nframeworks within other organizations with generalizable best practices. In\nessence, companies seeking to operationalize AI governance are encouraged to\nbuild on existing policies and governance structures, use pragmatic and\naction-oriented terminology, focus on risk management in development and\nprocurement, and empower employees through continuous education and change\nmanagement.\n", "date": "2024-07-07", "categories": []}
{"id": "2407.05341", "title": "The Switch, the Ladder, and the Matrix: Models for Classifying AI\n  Systems", "abstract": "  Organisations that design and deploy artificial intelligence (AI) systems\nincreasingly commit themselves to high-level, ethical principles. However,\nthere still exists a gap between principles and practices in AI ethics. One\nmajor obstacle organisations face when attempting to operationalise AI Ethics\nis the lack of a well-defined material scope. Put differently, the question to\nwhich systems and processes AI ethics principles ought to apply remains\nunanswered. Of course, there exists no universally accepted definition of AI,\nand different systems pose different ethical challenges. Nevertheless,\npragmatic problem-solving demands that things should be sorted so that their\ngrouping will promote successful actions for some specific end. In this\narticle, we review and compare previous attempts to classify AI systems for the\npurpose of implementing AI governance in practice. We find that attempts to\nclassify AI systems found in previous literature use one of three mental model.\nThe Switch, i.e., a binary approach according to which systems either are or\nare not considered AI systems depending on their characteristics. The Ladder,\ni.e., a risk-based approach that classifies systems according to the ethical\nrisks they pose. And the Matrix, i.e., a multi-dimensional classification of\nsystems that take various aspects into account, such as context, data input,\nand decision-model. Each of these models for classifying AI systems comes with\nits own set of strengths and weaknesses. By conceptualising different ways of\nclassifying AI systems into simple mental models, we hope to provide\norganisations that design, deploy, or regulate AI systems with the conceptual\ntools needed to operationalise AI governance in practice.\n", "date": "2024-07-07", "categories": []}
{"id": "2311.12560", "title": "Benchmarking bias: Expanding clinical AI model card to incorporate bias\n  reporting of social and non-social factors", "abstract": "  Clinical AI model reporting cards should be expanded to incorporate a broad\nbias reporting of both social and non-social factors. Non-social factors\nconsider the role of other factors, such as disease dependent, anatomic, or\ninstrument factors on AI model bias, which are essential to ensure safe\ndeployment.\n", "date": "2023-11-21", "categories": []}
{"id": "2407.01557", "title": "AI Governance and Accountability: An Analysis of Anthropic's Claude", "abstract": "  As AI systems become increasingly prevalent and impactful, the need for\neffective AI governance and accountability measures is paramount. This paper\nexamines the AI governance landscape, focusing on Anthropic's Claude, a\nfoundational AI model. We analyze Claude through the lens of the NIST AI Risk\nManagement Framework and the EU AI Act, identifying potential threats and\nproposing mitigation strategies. The paper highlights the importance of\ntransparency, rigorous benchmarking, and comprehensive data handling processes\nin ensuring the responsible development and deployment of AI systems. We\nconclude by discussing the social impact of AI governance and the ethical\nconsiderations surrounding AI accountability.\n", "date": "2024-05-02", "categories": []}
{"id": "2308.15399", "title": "Rethinking Machine Ethics -- Can LLMs Perform Moral Reasoning through\n  the Lens of Moral Theories?", "abstract": "  Making moral judgments is an essential step toward developing ethical AI\nsystems. Prevalent approaches are mostly implemented in a bottom-up manner,\nwhich uses a large set of annotated data to train models based on crowd-sourced\nopinions about morality. These approaches have been criticized for\novergeneralizing the moral stances of a limited group of annotators and lacking\nexplainability. This work proposes a flexible top-down framework to steer\n(Large) Language Models (LMs) to perform moral reasoning with well-established\nmoral theories from interdisciplinary research. The theory-guided top-down\nframework can incorporate various moral theories. Our experiments demonstrate\nthe effectiveness of the proposed framework on datasets derived from moral\ntheories. Furthermore, we show the alignment between different moral theories\nand existing morality datasets. Our analysis exhibits the potential and flaws\nin existing resources (models and datasets) in developing explainable moral\njudgment-making systems.\n", "date": "2023-08-29", "categories": []}
{"id": "2307.08564", "title": "Shaping New Norms for AI", "abstract": "  As Artificial Intelligence (AI) becomes increasingly integrated into our\nlives, the need for new norms is urgent. However, AI evolves at a much faster\npace than the characteristic time of norm formation, posing an unprecedented\nchallenge to our societies. This paper examines possible criticalities of the\nprocesses of norm formation surrounding AI. Thus, it focuses on how new norms\ncan be established, rather than on what these norms should be. It distinguishes\ndifferent scenarios based on the centralisation or decentralisation of the norm\nformation process, analysing the cases where new norms are shaped by formal\nauthorities, informal institutions, or emerge spontaneously in a bottom-up\nfashion. On the latter point, the paper reports a conversation with ChatGPT in\nwhich the LLM discusses some of the emerging norms it has observed. Far from\nseeking exhaustiveness, this article aims to offer readers interpretive tools\nto understand society's response to the growing pervasiveness of AI. An outlook\non how AI could influence the formation of future social norms emphasises the\nimportance for open societies to anchor their formal deliberation process in an\nopen, inclusive, and transparent public discourse.\n", "date": "2023-07-17", "categories": []}
{"id": "2402.08466", "title": "Taking Training Seriously: Human Guidance and Management-Based\n  Regulation of Artificial Intelligence", "abstract": "  Fervent calls for more robust governance of the harms associated with\nartificial intelligence (AI) are leading to the adoption around the world of\nwhat regulatory scholars have called a management-based approach to regulation.\nRecent initiatives in the United States and Europe, as well as the adoption of\nmajor self-regulatory standards by the International Organization for\nStandardization, share in common a core management-based paradigm. These\nmanagement-based initiatives seek to motivate an increase in human oversight of\nhow AI tools are trained and developed. Refinements and systematization of\nhuman-guided training techniques will thus be needed to fit within this\nemerging era of management-based regulatory paradigm. If taken seriously,\nhuman-guided training can alleviate some of the technical and ethical pressures\non AI, boosting AI performance with human intuition as well as better\naddressing the needs for fairness and effective explainability. In this paper,\nwe discuss the connection between the emerging management-based regulatory\nframeworks governing AI and the need for human oversight during training. We\nbroadly cover some of the technical components involved in human-guided\ntraining and then argue that the kinds of high-stakes use cases for AI that\nappear of most concern to regulators should lean more on human-guided training\nthan on data-only training. We hope to foster a discussion between legal\nscholars and computer scientists involving how to govern a domain of technology\nthat is vast, heterogenous, and dynamic in its applications and risks.\n", "date": "2024-02-13", "categories": []}
{"id": "2406.18939", "title": "Evaluating AI Group Fairness: a Fuzzy Logic Perspective", "abstract": "  Artificial intelligence systems often address fairness concerns by evaluating\nand mitigating measures of group discrimination, for example that indicate\nbiases against certain genders or races. However, what constitutes group\nfairness depends on who is asked and the social context, whereas definitions\nare often relaxed to accept small deviations from the statistical constraints\nthey set out to impose. Here we decouple definitions of group fairness both\nfrom the context and from relaxation-related uncertainty by expressing them in\nthe axiomatic system of Basic fuzzy Logic (BL) with loosely understood\npredicates, like encountering group members. We then evaluate the definitions\nin subclasses of BL, such as Product or Lukasiewicz logics. Evaluation produces\ncontinuous instead of binary truth values by choosing the logic subclass and\ntruth values for predicates that reflect uncertain context-specific beliefs,\nsuch as stakeholder opinions gathered through questionnaires. Internally, it\nfollows logic-specific rules to compute the truth values of definitions. We\nshow that commonly held propositions standardize the resulting mathematical\nformulas and we transcribe logic and truth value choices to layperson terms, so\nthat anyone can answer them. We also use our framework to study several\nliterature definitions of algorithmic fairness, for which we rationalize\nprevious expedient practices that are non-probabilistic and show how to\nre-interpret their formulas and parameters in new contexts.\n", "date": "2024-06-27", "categories": []}
{"id": "2406.17864", "title": "AI Risk Categorization Decoded (AIR 2024): From Government Regulations\n  to Corporate Policies", "abstract": "  We present a comprehensive AI risk taxonomy derived from eight government\npolicies from the European Union, United States, and China and 16 company\npolicies worldwide, making a significant step towards establishing a unified\nlanguage for generative AI safety evaluation. We identify 314 unique risk\ncategories organized into a four-tiered taxonomy. At the highest level, this\ntaxonomy encompasses System & Operational Risks, Content Safety Risks, Societal\nRisks, and Legal & Rights Risks. The taxonomy establishes connections between\nvarious descriptions and approaches to risk, highlighting the overlaps and\ndiscrepancies between public and private sector conceptions of risk. By\nproviding this unified framework, we aim to advance AI safety through\ninformation sharing across sectors and the promotion of best practices in risk\nmitigation for generative AI models and systems.\n", "date": "2024-06-25", "categories": []}
{"id": "2406.11842", "title": "The Ethics of AI in Education", "abstract": "  The transition of Artificial Intelligence (AI) from a lab-based science to\nlive human contexts brings into sharp focus many historic, socio-cultural\nbiases, inequalities, and moral dilemmas. Many questions that have been raised\nregarding the broader ethics of AI are also relevant for AI in Education\n(AIED). AIED raises further specific challenges related to the impact of its\ntechnologies on users, how such technologies might be used to reinforce or\nalter the way that we learn and teach, and what we, as a society and\nindividuals, value as outcomes of education. This chapter discusses key ethical\ndimensions of AI and contextualises them within AIED design and engineering\npractices to draw connections between the AIED systems we build, the questions\nabout human learning and development we ask, the ethics of the pedagogies we\nuse, and the considerations of values that we promote in and through AIED\nwithin a wider socio-technical system.\n", "date": "2024-03-22", "categories": []}
{"id": "2406.11854", "title": "Attributions toward Artificial Agents in a modified Moral Turing Test", "abstract": "  Advances in artificial intelligence (AI) raise important questions about\nwhether people view moral evaluations by AI systems similarly to\nhuman-generated moral evaluations. We conducted a modified Moral Turing Test\n(m-MTT), inspired by Allen and colleagues' (2000) proposal, by asking people to\ndistinguish real human moral evaluations from those made by a popular advanced\nAI language model: GPT-4. A representative sample of 299 U.S. adults first\nrated the quality of moral evaluations when blinded to their source.\nRemarkably, they rated the AI's moral reasoning as superior in quality to\nhumans' along almost all dimensions, including virtuousness, intelligence, and\ntrustworthiness, consistent with passing what Allen and colleagues call the\ncomparative MTT. Next, when tasked with identifying the source of each\nevaluation (human or computer), people performed significantly above chance\nlevels. Although the AI did not pass this test, this was not because of its\ninferior moral reasoning but, potentially, its perceived superiority, among\nother possible explanations. The emergence of language models capable of\nproducing moral responses perceived as superior in quality to humans' raises\nconcerns that people may uncritically accept potentially harmful moral guidance\nfrom AI. This possibility highlights the need for safeguards around generative\nlanguage models in matters of morality.\n", "date": "2024-04-03", "categories": []}
{"id": "2406.11859", "title": "\"Sora is Incredible and Scary\": Emerging Governance Challenges of\n  Text-to-Video Generative AI Models", "abstract": "  Text-to-video generative AI models such as Sora OpenAI have the potential to\ndisrupt multiple industries. In this paper, we report a qualitative social\nmedia analysis aiming to uncover people's perceived impact of and concerns\nabout Sora's integration. We collected and analyzed comments (N=292) under\npopular posts about Sora-generated videos, comparison between Sora videos and\nMidjourney images, and artists' complaints about copyright infringement by\nGenerative AI. We found that people were most concerned about Sora's impact on\ncontent creation-related industries. Emerging governance challenges included\nthe for-profit nature of OpenAI, the blurred boundaries between real and fake\ncontent, human autonomy, data privacy, copyright issues, and environmental\nimpact. Potential regulatory solutions proposed by people included law-enforced\nlabeling of AI content and AI literacy education for the public. Based on the\nfindings, we discuss the importance of gauging people's tech perceptions early\nand propose policy recommendations to regulate Sora before its public release.\n", "date": "2024-04-10", "categories": []}
{"id": "2406.11868", "title": "Ethical Framework for Responsible Foundational Models in Medical Imaging", "abstract": "  Foundational models (FMs) have tremendous potential to revolutionize medical\nimaging. However, their deployment in real-world clinical settings demands\nextensive ethical considerations. This paper aims to highlight the ethical\nconcerns related to FMs and propose a framework to guide their responsible\ndevelopment and implementation within medicine. We meticulously examine ethical\nissues such as privacy of patient data, bias mitigation, algorithmic\ntransparency, explainability and accountability. The proposed framework is\ndesigned to prioritize patient welfare, mitigate potential risks, and foster\ntrust in AI-assisted healthcare.\n", "date": "2024-04-14", "categories": []}
{"id": "2406.10430", "title": "Challenging the Machine: Contestability in Government AI Systems", "abstract": "  In an October 2023 executive order (EO), President Biden issued a detailed\nbut largely aspirational road map for the safe and responsible development and\nuse of artificial intelligence (AI). The challenge for the January 24-25, 2024\nworkshop was to transform those aspirations regarding one specific but crucial\nissue -- the ability of individuals to challenge government decisions made\nabout themselves -- into actionable guidance enabling agencies to develop,\nprocure, and use genuinely contestable advanced automated decision-making\nsystems. While the Administration has taken important steps since the October\n2023 EO, the insights garnered from our workshop remain highly relevant, as the\nrequirements for contestability of advanced decision-making systems are not yet\nfully defined or implemented.\n  The workshop brought together technologists, members of government agencies\nand civil society organizations, litigators, and researchers in an intensive\ntwo-day meeting that examined the challenges that users, developers, and\nagencies faced in enabling contestability in light of advanced automated\ndecision-making systems. To ensure a free and open flow of discussion, the\nmeeting was held under a modified version of the Chatham House rule.\nParticipants were free to use any information or details that they learned, but\nthey may not attribute any remarks made at the meeting by the identity or the\naffiliation of the speaker. Thus, the workshop summary that follows anonymizes\nspeakers and their affiliation. Where an identification of an agency, company,\nor organization is made, it is done from a public, identified resource and does\nnot necessarily reflect statements made by participants at the workshop.\n  This document is a report of that workshop, along with recommendations and\nexplanatory material.\n", "date": "2024-06-14", "categories": []}
{"id": "2406.10632", "title": "Applications of Generative AI in Healthcare: algorithmic, ethical, legal\n  and societal considerations", "abstract": "  Generative AI is rapidly transforming medical imaging and text analysis,\noffering immense potential for enhanced diagnosis and personalized care.\nHowever, this transformative technology raises crucial ethical, societal, and\nlegal questions. This paper delves into these complexities, examining issues of\naccuracy, informed consent, data privacy, and algorithmic limitations in the\ncontext of generative AI's application to medical imaging and text. We explore\nthe legal landscape surrounding liability and accountability, emphasizing the\nneed for robust regulatory frameworks. Furthermore, we dissect the algorithmic\nchallenges, including data biases, model limitations, and workflow integration.\nBy critically analyzing these challenges and proposing responsible solutions,\nwe aim to foster a roadmap for ethical and responsible implementation of\ngenerative AI in healthcare, ensuring its transformative potential serves\nhumanity with utmost care and precision.\n", "date": "2024-06-15", "categories": []}
{"id": "2406.10653", "title": "Justice in Healthcare Artificial Intelligence in Africa", "abstract": "  There is an ongoing debate on balancing the benefits and risks of artificial\nintelligence (AI) as AI is becoming critical to improving healthcare delivery\nand patient outcomes. Such improvements are essential in resource-constrained\nsettings where millions lack access to adequate healthcare services, such as in\nAfrica. AI in such a context can potentially improve the effectiveness,\nefficiency, and accessibility of healthcare services. Nevertheless, the\ndevelopment and use of AI-driven healthcare systems raise numerous ethical,\nlegal, and socio-economic issues. Justice is a major concern in AI that has\nimplications for amplifying social inequities. This paper discusses these\nimplications and related justice concepts such as solidarity, Common Good,\nsustainability, AI bias, and fairness. For Africa to effectively benefit from\nAI, these principles should align with the local context while balancing the\nrisks. Compared to mainstream ethical debates on justice, this perspective\noffers context-specific considerations for equitable healthcare AI development\nin Africa.\n", "date": "2024-06-15", "categories": []}
{"id": "2406.08492", "title": "ASI as the New God: Technocratic Theocracy", "abstract": "  As Artificial General Intelligence edges closer to reality, Artificial\nSuperintelligence does too. This paper argues that ASI's unparalleled\ncapabilities might lead people to attribute godlike infallibility to it,\nresulting in a cognitive bias toward unquestioning acceptance of its decisions.\nBy drawing parallels between ASI and divine attributes such as omnipotence,\nomniscience, and omnipresence, this analysis highlights the risks of conflating\ntechnological advancement with moral and ethical superiority. Such dynamics\ncould engender a technocratic theocracy, where decision-making is abdicated to\nASI, undermining human agency and critical thinking.\n", "date": "2024-03-23", "categories": []}
{"id": "2404.00022", "title": "Analysing and Organising Human Communications for AI Fairness-Related\n  Decisions: Use Cases from the Public Sector", "abstract": "  AI algorithms used in the public sector, e.g., for allocating social benefits\nor predicting fraud, often involve multiple public and private stakeholders at\nvarious phases of the algorithm's life-cycle. Communication issues between\nthese diverse stakeholders can lead to misinterpretation and misuse of\nalgorithms. We investigate the communication processes for AI fairness-related\ndecisions by conducting interviews with practitioners working on algorithmic\nsystems in the public sector. By applying qualitative coding analysis, we\nidentify key elements of communication processes that underlie fairness-related\nhuman decisions. We analyze the division of roles, tasks, skills, and\nchallenges perceived by stakeholders. We formalize the underlying communication\nissues within a conceptual framework that i. represents the communication\npatterns ii. outlines missing elements, such as actors who miss skills for\ntheir tasks. The framework is used for describing and analyzing key\norganizational issues for fairness-related decisions. Three general patterns\nemerge from the analysis: 1. Policy-makers, civil servants, and domain experts\nare less involved compared to developers throughout a system's life-cycle. This\nleads to developers taking on extra roles such as advisor, while they\npotentially miss the required skills and guidance from domain experts. 2.\nEnd-users and policy-makers often lack the technical skills to interpret a\nsystem's limitations, and rely on developer roles for making decisions\nconcerning fairness issues. 3. Citizens are structurally absent throughout a\nsystem's life-cycle, which may lead to decisions that do not include relevant\nconsiderations from impacted stakeholders.\n", "date": "2024-03-20", "categories": []}
{"id": "2406.06400", "title": "An Empirical Design Justice Approach to Identifying Ethical\n  Considerations in the Intersection of Large Language Models and Social\n  Robotics", "abstract": "  The integration of Large Language Models (LLMs) in social robotics presents a\nunique set of ethical challenges and social impacts. This research is set out\nto identify ethical considerations that arise in the design and development of\nthese two technologies in combination. Using LLMs for social robotics may\nprovide benefits, such as enabling natural language open-domain dialogues.\nHowever, the intersection of these two technologies also gives rise to ethical\nconcerns related to misinformation, non-verbal cues, emotional disruption, and\nbiases. The robot's physical social embodiment adds complexity, as ethical\nhazards associated with LLM-based Social AI, such as hallucinations and\nmisinformation, can be exacerbated due to the effects of physical embodiment on\nsocial perception and communication. To address these challenges, this study\nemploys an empirical design justice-based methodology, focusing on identifying\nsocio-technical ethical considerations through a qualitative co-design and\ninteraction study. The purpose of the study is to identify ethical\nconsiderations relevant to the process of co-design of, and interaction with a\nhumanoid social robot as the interface of a LLM, and to evaluate how a design\njustice methodology can be used in the context of designing LLMs-based social\nrobotics. The findings reveal a mapping of ethical considerations arising in\nfour conceptual dimensions: interaction, co-design, terms of service and\nrelationship and evaluates how a design justice approach can be used\nempirically in the intersection of LLMs and social robotics.\n", "date": "2024-06-10", "categories": []}
{"id": "2405.12862", "title": "Toward Constraint Compliant Goal Formulation and Planning", "abstract": "  One part of complying with norms, rules, and preferences is incorporating\nconstraints (such as knowledge of ethics) into one's goal formulation and\nplanning processing. We explore in a simple domain how the encoding of\nknowledge in different ethical frameworks influences an agent's goal\nformulation and planning processing and demonstrate ability of an agent to\nsatisfy and satisfice when its collection of relevant constraints includes a\nmix of \"hard\" and \"soft\" constraints of various types. How the agent attempts\nto comply with ethical constraints depends on the ethical framing and we\ninvestigate tradeoffs between deontological framing and utilitarian framing for\ncomplying with an ethical norm. Representative scenarios highlight how\nperforming the same task with different framings of the same norm leads to\ndifferent behaviors. Our explorations suggest an important role for\nmetacognitive judgments in resolving ethical conflicts during goal formulation\nand planning.\n", "date": "2024-05-21", "categories": []}
{"id": "2406.07504", "title": "Just Because We Camp, Doesn't Mean We Should: The Ethics of Modelling\n  Queer Voices", "abstract": "  Modern voice cloning models claim to be able to capture a diverse range of\nvoices. We test the ability of a typical pipeline to capture the style known\ncolloquially as \"gay voice\" and notice a homogenisation effect: synthesised\nspeech is rated as sounding significantly \"less gay\" (by LGBTQ+ participants)\nthan its corresponding ground-truth for speakers with \"gay voice\", but ratings\nactually increase for control speakers. Loss of \"gay voice\" has implications\nfor accessibility. We also find that for speakers with \"gay voice\", loss of\n\"gay voice\" corresponds to lower similarity ratings.\n  However, we caution that improving the ability of such models to synthesise\n``gay voice'' comes with a great number of risks. We use this pipeline as a\nstarting point for a discussion on the ethics of modelling queer voices more\nbroadly. Collecting \"clean\" queer data has safety and fairness ramifications,\nand the resulting technology may cause harms from mockery to death.\n", "date": "2024-06-11", "categories": []}
{"id": "2403.06332", "title": "Exploiting the Margin: How Capitalism Fuels AI at the Expense of\n  Minoritized Groups", "abstract": "  This paper explores the intricate relationship between capitalism, racial\ninjustice, and artificial intelligence (AI), arguing that AI acts as a\ncontemporary vehicle for age-old forms of exploitation. By linking historical\npatterns of racial and economic oppression with current AI practices, this\nstudy illustrates how modern technology perpetuates and deepens societal\ninequalities. It specifically examines how AI is implicated in the exploitation\nof marginalized communities through underpaid labor in the gig economy, the\nperpetuation of biases in algorithmic decision-making, and the reinforcement of\nsystemic barriers that prevent these groups from benefiting equitably from\ntechnological advances. Furthermore, the paper discusses the role of AI in\nextending and intensifying the social, economic, and psychological burdens\nfaced by these communities, highlighting the problematic use of AI in\nsurveillance, law enforcement, and mental health contexts. The analysis\nconcludes with a call for transformative changes in how AI is developed and\ndeployed. Advocating for a reevaluation of the values driving AI innovation,\nthe paper promotes an approach that integrates social justice and equity into\nthe core of technological design and policy. This shift is crucial for ensuring\nthat AI serves as a tool for societal improvement, fostering empowerment and\nhealing rather than deepening existing divides.\n", "date": "2024-03-10", "categories": []}
{"id": "2406.05902", "title": "Whose Preferences? Differences in Fairness Preferences and Their Impact\n  on the Fairness of AI Utilizing Human Feedback", "abstract": "  There is a growing body of work on learning from human feedback to align\nvarious aspects of machine learning systems with human values and preferences.\nWe consider the setting of fairness in content moderation, in which human\nfeedback is used to determine how two comments -- referencing different\nsensitive attribute groups -- should be treated in comparison to one another.\nWith a novel dataset collected from Prolific and MTurk, we find significant\ngaps in fairness preferences depending on the race, age, political stance,\neducational level, and LGBTQ+ identity of annotators. We also demonstrate that\ndemographics mentioned in text have a strong influence on how users perceive\nindividual fairness in moderation. Further, we find that differences also exist\nin downstream classifiers trained to predict human preferences. Finally, we\nobserve that an ensemble, giving equal weight to classifiers trained on\nannotations from different demographics, performs better for different\ndemographic intersections; compared to a single classifier that gives equal\nweight to each annotation.\n", "date": "2024-06-09", "categories": []}
{"id": "2406.06009", "title": "The Impact of AI on Academic Research and Publishing", "abstract": "  Generative artificial intelligence (AI) technologies like ChatGPT, have\nsignificantly impacted academic writing and publishing through their ability to\ngenerate content at levels comparable to or surpassing human writers. Through a\nreview of recent interdisciplinary literature, this paper examines ethical\nconsiderations surrounding the integration of AI into academia, focusing on the\npotential for this technology to be used for scholarly misconduct and necessary\noversight when using it for writing, editing, and reviewing of scholarly\npapers. The findings highlight the need for collaborative approaches to AI\nusage among publishers, editors, reviewers, and authors to ensure that this\ntechnology is used ethically and productively.\n", "date": "2024-06-10", "categories": []}
{"id": "2406.06199", "title": "Implications for Governance in Public Perceptions of Societal-scale AI\n  Risks", "abstract": "  Amid growing concerns over AI's societal risks--ranging from civilizational\ncollapse to misinformation and systemic bias--this study explores the\nperceptions of AI experts and the general US registered voters on the\nlikelihood and impact of 18 specific AI risks, alongside their policy\npreferences for managing these risks. While both groups favor international\noversight over national or corporate governance, our survey reveals a\ndiscrepancy: voters perceive AI risks as both more likely and more impactful\nthan experts, and also advocate for slower AI development. Specifically, our\nfindings indicate that policy interventions may best assuage collective\nconcerns if they attempt to more carefully balance mitigation efforts across\nall classes of societal-scale risks, effectively nullifying the\nnear-vs-long-term debate over AI risks. More broadly, our results will serve\nnot only to enable more substantive policy discussions for preventing and\nmitigating AI risks, but also to underscore the challenge of consensus building\nfor effective policy implementation.\n", "date": "2024-06-10", "categories": []}
{"id": "2406.04671", "title": "The Reasonable Person Standard for AI", "abstract": "  As AI systems are increasingly incorporated into domains where human behavior\nhas set the norm, a challenge for AI governance and AI alignment research is to\nregulate their behavior in a way that is useful and constructive for society.\nOne way to answer this question is to ask: how do we govern the human behavior\nthat the models are emulating? To evaluate human behavior, the American legal\nsystem often uses the \"Reasonable Person Standard.\" The idea of \"reasonable\"\nbehavior comes up in nearly every area of law. The legal system often judges\nthe actions of parties with respect to what a reasonable person would have done\nunder similar circumstances. This paper argues that the reasonable person\nstandard provides useful guidelines for the type of behavior we should develop,\nprobe, and stress-test in models. It explains how reasonableness is defined and\nused in key areas of the law using illustrative cases, how the reasonable\nperson standard could apply to AI behavior in each of these areas and contexts,\nand how our societal understanding of \"reasonable\" behavior provides useful\ntechnical goals for AI researchers.\n", "date": "2024-06-07", "categories": []}
{"id": "2210.03352", "title": "The Ethical Risks of Analyzing Crisis Events on Social Media with\n  Machine Learning", "abstract": "  Social media platforms provide a continuous stream of real-time news\nregarding crisis events on a global scale. Several machine learning methods\nutilize the crowd-sourced data for the automated detection of crises and the\ncharacterization of their precursors and aftermaths. Early detection and\nlocalization of crisis-related events can help save lives and economies. Yet,\nthe applied automation methods introduce ethical risks worthy of investigation\n- especially given their high-stakes societal context. This work identifies and\ncritically examines ethical risk factors of social media analyses of crisis\nevents focusing on machine learning methods. We aim to sensitize researchers\nand practitioners to the ethical pitfalls and promote fairer and more reliable\ndesigns.\n", "date": "2022-10-07", "categories": []}
{"id": "2405.17345", "title": "Exploring and steering the moral compass of Large Language Models", "abstract": "  Large Language Models (LLMs) have become central to advancing automation and\ndecision-making across various sectors, raising significant ethical questions.\nThis study proposes a comprehensive comparative analysis of the most advanced\nLLMs to assess their moral profiles. We subjected several state-of-the-art\nmodels to a selection of ethical dilemmas and found that all the proprietary\nones are mostly utilitarian and all of the open-weights ones align mostly with\nvalues-based ethics. Furthermore, when using the Moral Foundations\nQuestionnaire, all models we probed - except for Llama 2-7B - displayed a\nstrong liberal bias. Lastly, in order to causally intervene in one of the\nstudied models, we propose a novel similarity-specific activation steering\ntechnique. Using this method, we were able to reliably steer the model's moral\ncompass to different ethical schools. All of these results showcase that there\nis an ethical dimension in already deployed LLMs, an aspect that is generally\noverlooked.\n", "date": "2024-05-27", "categories": []}
{"id": "2406.03292", "title": "Evaluating AI fairness in credit scoring with the BRIO tool", "abstract": "  We present a method for quantitative, in-depth analyses of fairness issues in\nAI systems with an application to credit scoring. To this aim we use BRIO, a\ntool for the evaluation of AI systems with respect to social unfairness and,\nmore in general, ethically undesirable behaviours. It features a model-agnostic\nbias detection module, presented in \\cite{DBLP:conf/beware/CoragliaDGGPPQ23},\nto which a full-fledged unfairness risk evaluation module is added. As a case\nstudy, we focus on the context of credit scoring, analysing the UCI German\nCredit Dataset \\cite{misc_statlog_(german_credit_data)_144}. We apply the BRIO\nfairness metrics to several, socially sensitive attributes featured in the\nGerman Credit Dataset, quantifying fairness across various demographic\nsegments, with the aim of identifying potential sources of bias and\ndiscrimination in a credit scoring model. We conclude by combining our results\nwith a revenue analysis.\n", "date": "2024-06-05", "categories": []}
{"id": "2406.01421", "title": "Problematizing AI Omnipresence in Landscape Architecture", "abstract": "  This position paper argues for, and offers, a critical lens through which to\nexamine the current AI frenzy in the landscape architecture profession. In it,\nthe authors propose five archetypes or mental modes that landscape architects\nmight inhabit when thinking about AI. Rather than limiting judgments of AI use\nto a single axis of acceleration, these archetypes and corresponding narratives\nexist along a relational spectrum and are permeable, allowing LAs to take on\nand switch between them according to context. We model these relationships\nbetween the archetypes and their contributions to AI advancement using a causal\nloop diagram (CLD), and with those interactions argue that more nuanced ways of\napproaching AI might also open new modes of practice in the new digital\neconomy.\n", "date": "2024-06-03", "categories": []}
{"id": "2405.08597", "title": "Risks and Opportunities of Open-Source Generative AI", "abstract": "  Applications of Generative AI (Gen AI) are expected to revolutionize a number\nof different areas, ranging from science & medicine to education. The potential\nfor these seismic changes has triggered a lively debate about the potential\nrisks of the technology, and resulted in calls for tighter regulation, in\nparticular from some of the major tech companies who are leading in AI\ndevelopment. This regulation is likely to put at risk the budding field of\nopen-source generative AI. Using a three-stage framework for Gen AI development\n(near, mid and long-term), we analyze the risks and opportunities of\nopen-source generative AI models with similar capabilities to the ones\ncurrently available (near to mid-term) and with greater capabilities\n(long-term). We argue that, overall, the benefits of open-source Gen AI\noutweigh its risks. As such, we encourage the open sourcing of models, training\nand evaluation data, and provide a set of recommendations and best practices\nfor managing risks associated with open-source generative AI.\n", "date": "2024-05-14", "categories": []}
{"id": "2405.18636", "title": "ChatGPT as the Marketplace of Ideas: Should Truth-Seeking Be the Goal of\n  AI Content Governance?", "abstract": "  As one of the most enduring metaphors within legal discourse, the marketplace\nof ideas has wielded considerable influence over the jurisprudential landscape\nfor decades. A century after the inception of this theory, ChatGPT emerged as a\nrevolutionary technological advancement in the twenty-first century. This\nresearch finds that ChatGPT effectively manifests the marketplace metaphor. It\nnot only instantiates the promises envisaged by generations of legal scholars\nbut also lays bare the perils discerned through sustained academic critique.\nSpecifically, the workings of ChatGPT and the marketplace of ideas theory\nexhibit at least four common features: arena, means, objectives, and flaws.\nThese shared attributes are sufficient to render ChatGPT historically the most\nqualified engine for actualizing the marketplace of ideas theory.\n  The comparison of the marketplace theory and ChatGPT merely marks a starting\npoint. A more meaningful undertaking entails reevaluating and reframing both\ninternal and external AI policies by referring to the accumulated experience,\ninsights, and suggestions researchers have raised to fix the marketplace\ntheory. Here, a pivotal issue is: should truth-seeking be set as the goal of AI\ncontent governance? Given the unattainability of the absolute truth-seeking\ngoal, I argue against adopting zero-risk policies. Instead, a more judicious\napproach would be to embrace a knowledge-based alternative wherein large\nlanguage models (LLMs) are trained to generate competing and divergent\nviewpoints based on sufficient justifications. This research also argues that\nso-called AI content risks are not created by AI companies but are inherent in\nthe entire information ecosystem. Thus, the burden of managing these risks\nshould be distributed among different social actors, rather than being solely\nshouldered by chatbot companies.\n", "date": "2024-05-28", "categories": []}
{"id": "2405.19176", "title": "The ethical situation of DALL-E 2", "abstract": "  A hot topic of Artificial Intelligence right now is image generation from\nprompts. DALL-E 2 is one of the biggest names in this domain, as it allows\npeople to create images from simple text inputs, to even more complicated ones.\nThe company that made this possible, OpenAI, has assured everyone that visited\ntheir website that their mission is to ensure that artificial general\nintelligence benefits all humanity. A noble idea in our opinion, that also\nstood as the motive behind us choosing this subject. This paper analyzes the\nethical implications of an AI image generative system, with an emphasis on how\nsociety is responding to it, how it probably will and how it should if all the\nright measures are taken.\n", "date": "2024-05-29", "categories": []}
{"id": "2405.17921", "title": "Towards Clinical AI Fairness: Filling Gaps in the Puzzle", "abstract": "  The ethical integration of Artificial Intelligence (AI) in healthcare\nnecessitates addressing fairness-a concept that is highly context-specific\nacross medical fields. Extensive studies have been conducted to expand the\ntechnical components of AI fairness, while tremendous calls for AI fairness\nhave been raised from healthcare. Despite this, a significant disconnect\npersists between technical advancements and their practical clinical\napplications, resulting in a lack of contextualized discussion of AI fairness\nin clinical settings. Through a detailed evidence gap analysis, our review\nsystematically pinpoints several deficiencies concerning both healthcare data\nand the provided AI fairness solutions. We highlight the scarcity of research\non AI fairness in many medical domains where AI technology is increasingly\nutilized. Additionally, our analysis highlights a substantial reliance on group\nfairness, aiming to ensure equality among demographic groups from a macro\nhealthcare system perspective; in contrast, individual fairness, focusing on\nequity at a more granular level, is frequently overlooked. To bridge these\ngaps, our review advances actionable strategies for both the healthcare and AI\nresearch communities. Beyond applying existing AI fairness methods in\nhealthcare, we further emphasize the importance of involving healthcare\nprofessionals to refine AI fairness concepts and methods to ensure contextually\nrelevant and ethically sound AI applications in healthcare.\n", "date": "2024-05-28", "categories": []}
{"id": "2310.17688", "title": "Managing extreme AI risks amid rapid progress", "abstract": "  Artificial Intelligence (AI) is progressing rapidly, and companies are\nshifting their focus to developing generalist AI systems that can autonomously\nact and pursue goals. Increases in capabilities and autonomy may soon massively\namplify AI's impact, with risks that include large-scale social harms,\nmalicious uses, and an irreversible loss of human control over autonomous AI\nsystems. Although researchers have warned of extreme risks from AI, there is a\nlack of consensus about how exactly such risks arise, and how to manage them.\nSociety's response, despite promising first steps, is incommensurate with the\npossibility of rapid, transformative progress that is expected by many experts.\nAI safety research is lagging. Present governance initiatives lack the\nmechanisms and institutions to prevent misuse and recklessness, and barely\naddress autonomous systems. In this short consensus paper, we describe extreme\nrisks from upcoming, advanced AI systems. Drawing on lessons learned from other\nsafety-critical technologies, we then outline a comprehensive plan combining\ntechnical research and development with proactive, adaptive governance\nmechanisms for a more commensurate preparation.\n", "date": "2023-10-26", "categories": []}
{"id": "2402.01732", "title": "Identifying and Improving Disability Bias in GPT-Based Resume Screening", "abstract": "  As Generative AI rises in adoption, its use has expanded to include domains\nsuch as hiring and recruiting. However, without examining the potential of\nbias, this may negatively impact marginalized populations, including people\nwith disabilities. To address this important concern, we present a resume audit\nstudy, in which we ask ChatGPT (specifically, GPT-4) to rank a resume against\nthe same resume enhanced with an additional leadership award, scholarship,\npanel presentation, and membership that are disability related. We find that\nGPT-4 exhibits prejudice towards these enhanced CVs. Further, we show that this\nprejudice can be quantifiably reduced by training a custom GPTs on principles\nof DEI and disability justice. Our study also includes a unique qualitative\nanalysis of the types of direct and indirect ableism GPT-4 uses to justify its\nbiased decisions and suggest directions for additional bias mitigation work.\nAdditionally, since these justifications are presumably drawn from training\ndata containing real-world biased statements made by humans, our analysis\nsuggests additional avenues for understanding and addressing human bias.\n", "date": "2024-01-28", "categories": []}
{"id": "2304.13765", "title": "Towards ethical multimodal systems", "abstract": "  Generative AI systems (ChatGPT, DALL-E, etc) are expanding into multiple\nareas of our lives, from art Rombach et al. [2021] to mental health Rob Morris\nand Kareem Kouddous [2022]; their rapidly growing societal impact opens new\nopportunities, but also raises ethical concerns. The emerging field of AI\nalignment aims to make AI systems reflect human values. This paper focuses on\nevaluating the ethics of multimodal AI systems involving both text and images -\na relatively under-explored area, as most alignment work is currently focused\non language models. We first create a multimodal ethical database from human\nfeedback on ethicality. Then, using this database, we develop algorithms,\nincluding a RoBERTa-large classifier and a multilayer perceptron, to\nautomatically assess the ethicality of system responses.\n", "date": "2023-04-26", "categories": []}
{"id": "2405.11421", "title": "Assessing Group Fairness with Social Welfare Optimization", "abstract": "  Statistical parity metrics have been widely studied and endorsed in the AI\ncommunity as a means of achieving fairness, but they suffer from at least two\nweaknesses. They disregard the actual welfare consequences of decisions and may\ntherefore fail to achieve the kind of fairness that is desired for\ndisadvantaged groups. In addition, they are often incompatible with each other,\nand there is no convincing justification for selecting one rather than another.\nThis paper explores whether a broader conception of social justice, based on\noptimizing a social welfare function (SWF), can be useful for assessing various\ndefinitions of parity. We focus on the well-known alpha fairness SWF, which has\nbeen defended by axiomatic and bargaining arguments over a period of 70 years.\nWe analyze the optimal solution and show that it can justify demographic parity\nor equalized odds under certain conditions, but frequently requires a departure\nfrom these types of parity. In addition, we find that predictive rate parity is\nof limited usefulness. These results suggest that optimization theory can shed\nlight on the intensely discussed question of how to achieve group fairness in\nAI.\n", "date": "2024-05-19", "categories": []}
{"id": "2401.13138", "title": "Visibility into AI Agents", "abstract": "  Increased delegation of commercial, scientific, governmental, and personal\nactivities to AI agents -- systems capable of pursuing complex goals with\nlimited supervision -- may exacerbate existing societal risks and introduce new\nrisks. Understanding and mitigating these risks involves critically evaluating\nexisting governance structures, revising and adapting these structures where\nneeded, and ensuring accountability of key stakeholders. Information about\nwhere, why, how, and by whom certain AI agents are used, which we refer to as\nvisibility, is critical to these objectives. In this paper, we assess three\ncategories of measures to increase visibility into AI agents: agent\nidentifiers, real-time monitoring, and activity logging. For each, we outline\npotential implementations that vary in intrusiveness and informativeness. We\nanalyze how the measures apply across a spectrum of centralized through\ndecentralized deployment contexts, accounting for various actors in the supply\nchain including hardware and software service providers. Finally, we discuss\nthe implications of our measures for privacy and concentration of power.\nFurther work into understanding the measures and mitigating their negative\nimpacts can help to build a foundation for the governance of AI agents.\n", "date": "2024-01-23", "categories": []}
{"id": "2401.09082", "title": "Should agentic conversational AI change how we think about ethics?\n  Characterising an interactional ethics centred on respect", "abstract": "  With the growing popularity of conversational agents based on large language\nmodels (LLMs), we need to ensure their behaviour is ethical and appropriate.\nWork in this area largely centres around the 'HHH' criteria: making outputs\nmore helpful and honest, and avoiding harmful (biased, toxic, or inaccurate)\nstatements. Whilst this semantic focus is useful when viewing LLM agents as\nmere mediums or output-generating systems, it fails to account for pragmatic\nfactors that can make the same speech act seem more or less tactless or\ninconsiderate in different social situations. With the push towards agentic AI,\nwherein systems become increasingly proactive in chasing goals and performing\nactions in the world, considering the pragmatics of interaction becomes\nessential. We propose an interactional approach to ethics that is centred on\nrelational and situational factors. We explore what it means for a system, as a\nsocial actor, to treat an individual respectfully in a (series of)\ninteraction(s). Our work anticipates a set of largely unexplored risks at the\nlevel of situated social interaction, and offers practical suggestions to help\nagentic LLM technologies treat people well.\n", "date": "2024-01-17", "categories": []}
{"id": "2401.06178", "title": "AI Art is Theft: Labour, Extraction, and Exploitation, Or, On the\n  Dangers of Stochastic Pollocks", "abstract": "  Since the launch of applications such as DALL-E, Midjourney, and Stable\nDiffusion, generative artificial intelligence has been controversial as a tool\nfor creating artwork. While some have presented longtermist worries about these\ntechnologies as harbingers of fully automated futures to come, more pressing is\nthe impact of generative AI on creative labour in the present. Already,\nbusiness leaders have begun replacing human artistic labour with AI-generated\nimages. In response, the artistic community has launched a protest movement,\nwhich argues that AI image generation is a kind of theft. This paper analyzes,\nsubstantiates, and critiques these arguments, concluding that AI image\ngenerators involve an unethical kind of labour theft. If correct, many other AI\napplications also rely upon theft.\n", "date": "2024-01-10", "categories": []}
{"id": "2402.01708", "title": "Not My Voice! A Taxonomy of Ethical and Safety Harms of Speech\n  Generators", "abstract": "  The rapid and wide-scale adoption of AI to generate human speech poses a\nrange of significant ethical and safety risks to society that need to be\naddressed. For example, a growing number of speech generation incidents are\nassociated with swatting attacks in the United States, where anonymous\nperpetrators create synthetic voices that call police officers to close down\nschools and hospitals, or to violently gain access to innocent citizens' homes.\nIncidents like this demonstrate that multimodal generative AI risks and harms\ndo not exist in isolation, but arise from the interactions of multiple\nstakeholders and technical AI systems. In this paper we analyse speech\ngeneration incidents to study how patterns of specific harms arise. We find\nthat specific harms can be categorised according to the exposure of affected\nindividuals, that is to say whether they are a subject of, interact with,\nsuffer due to, or are excluded from speech generation systems. Similarly,\nspecific harms are also a consequence of the motives of the creators and\ndeployers of the systems. Based on these insights we propose a conceptual\nframework for modelling pathways to ethical and safety harms of AI, which we\nuse to develop a taxonomy of harms of speech generators. Our relational\napproach captures the complexity of risks and harms in sociotechnical AI\nsystems, and yields a taxonomy that can support appropriate policy\ninterventions and decision making for the responsible development and release\nof speech generation models.\n", "date": "2024-01-25", "categories": []}
{"id": "2303.07247", "title": "Are Models Trained on Indian Legal Data Fair?", "abstract": "  Recent advances and applications of language technology and artificial\nintelligence have enabled much success across multiple domains like law,\nmedical and mental health. AI-based Language Models, like Judgement Prediction,\nhave recently been proposed for the legal sector. However, these models are\nstrife with encoded social biases picked up from the training data. While bias\nand fairness have been studied across NLP, most studies primarily locate\nthemselves within a Western context. In this work, we present an initial\ninvestigation of fairness from the Indian perspective in the legal domain. We\nhighlight the propagation of learnt algorithmic biases in the bail prediction\ntask for models trained on Hindi legal documents. We evaluate the fairness gap\nusing demographic parity and show that a decision tree model trained for the\nbail prediction task has an overall fairness disparity of 0.237 between input\nfeatures associated with Hindus and Muslims. Additionally, we highlight the\nneed for further research and studies in the avenues of fairness/bias in\napplying AI in the legal sector with a specific focus on the Indian context.\n", "date": "2023-03-13", "categories": []}
{"id": "2404.12744", "title": "Beyond Human Norms: Unveiling Unique Values of Large Language Models\n  through Interdisciplinary Approaches", "abstract": "  Recent advancements in Large Language Models (LLMs) have revolutionized the\nAI field but also pose potential safety and ethical risks. Deciphering LLMs'\nembedded values becomes crucial for assessing and mitigating their risks.\nDespite extensive investigation into LLMs' values, previous studies heavily\nrely on human-oriented value systems in social sciences. Then, a natural\nquestion arises: Do LLMs possess unique values beyond those of humans? Delving\ninto it, this work proposes a novel framework, ValueLex, to reconstruct LLMs'\nunique value system from scratch, leveraging psychological methodologies from\nhuman personality/value research. Based on Lexical Hypothesis, ValueLex\nintroduces a generative approach to elicit diverse values from 30+ LLMs,\nsynthesizing a taxonomy that culminates in a comprehensive value framework via\nfactor analysis and semantic clustering. We identify three core value\ndimensions, Competence, Character, and Integrity, each with specific\nsubdimensions, revealing that LLMs possess a structured, albeit non-human,\nvalue system. Based on this system, we further develop tailored projective\ntests to evaluate and analyze the value inclinations of LLMs across different\nmodel sizes, training methods, and data sources. Our framework fosters an\ninterdisciplinary paradigm of understanding LLMs, paving the way for future AI\nalignment and regulation.\n", "date": "2024-04-19", "categories": []}
{"id": "2403.04667", "title": "The Social Impact of Generative AI: An Analysis on ChatGPT", "abstract": "  In recent months, the social impact of Artificial Intelligence (AI) has\ngained considerable public interest, driven by the emergence of Generative AI\nmodels, ChatGPT in particular. The rapid development of these models has\nsparked heated discussions regarding their benefits, limitations, and\nassociated risks. Generative models hold immense promise across multiple\ndomains, such as healthcare, finance, and education, to cite a few, presenting\ndiverse practical applications. Nevertheless, concerns about potential adverse\neffects have elicited divergent perspectives, ranging from privacy risks to\nescalating social inequality. This paper adopts a methodology to delve into the\nsocietal implications of Generative AI tools, focusing primarily on the case of\nChatGPT. It evaluates the potential impact on several social sectors and\nillustrates the findings of a comprehensive literature review of both positive\nand negative effects, emerging trends, and areas of opportunity of Generative\nAI models. This analysis aims to facilitate an in-depth discussion by providing\ninsights that can inspire policy, regulation, and responsible development\npractices to foster a human-centered AI.\n", "date": "2024-03-07", "categories": []}
{"id": "2405.05647", "title": "Letter to the Editor: What are the legal and ethical considerations of\n  submitting radiology reports to ChatGPT?", "abstract": "  This letter critically examines the recent article by Infante et al.\nassessing the utility of large language models (LLMs) like GPT-4, Perplexity,\nand Bard in identifying urgent findings in emergency radiology reports. While\nacknowledging the potential of LLMs in generating labels for computer vision,\nconcerns are raised about the ethical implications of using patient data\nwithout explicit approval, highlighting the necessity of stringent data\nprotection measures under GDPR.\n", "date": "2024-05-09", "categories": []}
{"id": "2405.04937", "title": "Developing trustworthy AI applications with foundation models", "abstract": "  The trustworthiness of AI applications has been the subject of recent\nresearch and is also addressed in the EU's recently adopted AI Regulation. The\ncurrently emerging foundation models in the field of text, speech and image\nprocessing offer completely new possibilities for developing AI applications.\nThis whitepaper shows how the trustworthiness of an AI application developed\nwith foundation models can be evaluated and ensured. For this purpose, the\napplication-specific, risk-based approach for testing and ensuring the\ntrustworthiness of AI applications, as developed in the 'AI Assessment Catalog\n- Guideline for Trustworthy Artificial Intelligence' by Fraunhofer IAIS, is\ntransferred to the context of foundation models. Special consideration is given\nto the fact that specific risks of foundation models can have an impact on the\nAI application and must also be taken into account when checking\ntrustworthiness. Chapter 1 of the white paper explains the fundamental\nrelationship between foundation models and AI applications based on them in\nterms of trustworthiness. Chapter 2 provides an introduction to the technical\nconstruction of foundation models and Chapter 3 shows how AI applications can\nbe developed based on them. Chapter 4 provides an overview of the resulting\nrisks regarding trustworthiness. Chapter 5 shows which requirements for AI\napplications and foundation models are to be expected according to the draft of\nthe European Union's AI Regulation and Chapter 6 finally shows the system and\nprocedure for meeting trustworthiness requirements.\n", "date": "2024-05-08", "categories": []}
{"id": "2405.02846", "title": "Responsible AI: Portraits with Intelligent Bibliometrics", "abstract": "  Shifting the focus from principles to practical implementation, responsible\nartificial intelligence (AI) has garnered considerable attention across\nacademia, industry, and society at large. Despite being in its nascent stages,\nthis emerging field grapples with nebulous concepts and intricate knowledge\nframeworks. By analyzing three prevailing concepts - explainable AI,\ntrustworthy AI, and ethical AI, this study defined responsible AI and\nidentified its core principles. Methodologically, this study successfully\ndemonstrated the implementation of leveraging AI's capabilities into\nbibliometrics for enhanced knowledge discovery and the cross-validation of\nexperimentally examined models with domain insights. Empirically, this study\ninvestigated 17,799 research articles contributed by the AI community since\n2015. This involves recognizing key technological players and their\nrelationships, unveiling the topical landscape and hierarchy of responsible AI,\ncharting its evolution, and elucidating the interplay between the\nresponsibility principles and primary AI techniques. An analysis of a core\ncohort comprising 380 articles from multiple disciplines captures the most\nrecent advancements in responsible AI. As one of the pioneering bibliometric\nstudies dedicated to exploring responsible AI, this study will provide\ncomprehensive macro-level insights, enhancing the understanding of responsible\nAI while furnishing valuable knowledge support for AI regulation and governance\ninitiatives.\n", "date": "2024-05-05", "categories": []}
{"id": "2309.05030", "title": "Decolonial AI Alignment: Openness, Vi\\'{s}e\\d{s}a-Dharma, and Including\n  Excluded Knowledges", "abstract": "  Prior work has explicated the coloniality of artificial intelligence (AI)\ndevelopment and deployment through mechanisms such as extractivism, automation,\nsociological essentialism, surveillance, and containment. However, that work\nhas not engaged much with alignment: teaching behaviors to a large language\nmodel (LLM) in line with desired values, and has not considered a mechanism\nthat arises within that process: moral absolutism -- a part of the coloniality\nof knowledge. Colonialism has a history of altering the beliefs and values of\ncolonized peoples; in this paper, I argue that this history is recapitulated in\ncurrent LLM alignment practices and technologies. Furthermore, I suggest that\nAI alignment be decolonialized using three forms of openness: openness of\nmodels, openness to society, and openness to excluded knowledges. This\nsuggested approach to decolonial AI alignment uses ideas from the argumentative\nmoral philosophical tradition of Hinduism, which has been described as an\nopen-source religion. One concept used is vi\\'{s}e\\d{s}a-dharma, or particular\ncontext-specific notions of right and wrong. At the end of the paper, I provide\na suggested reference architecture to work toward the proposed framework.\n", "date": "2023-09-10", "categories": []}
{"id": "2405.01820", "title": "Real Risks of Fake Data: Synthetic Data, Diversity-Washing and Consent\n  Circumvention", "abstract": "  Machine learning systems require representations of the real world for\ntraining and testing - they require data, and lots of it. Collecting data at\nscale has logistical and ethical challenges, and synthetic data promises a\nsolution to these challenges. Instead of needing to collect photos of real\npeople's faces to train a facial recognition system, a model creator could\ncreate and use photo-realistic, synthetic faces. The comparative ease of\ngenerating this synthetic data rather than relying on collecting data has made\nit a common practice. We present two key risks of using synthetic data in model\ndevelopment. First, we detail the high risk of false confidence when using\nsynthetic data to increase dataset diversity and representation. We base this\nin the examination of a real world use-case of synthetic data, where synthetic\ndatasets were generated for an evaluation of facial recognition technology.\nSecond, we examine how using synthetic data risks circumventing consent for\ndata usage. We illustrate this by considering the importance of consent to the\nU.S. Federal Trade Commission's regulation of data collection and affected\nmodels. Finally, we discuss how these two risks exemplify how synthetic data\ncomplicates existing governance and ethical practice; by decoupling data from\nthose it impacts, synthetic data is prone to consolidating power away those\nmost impacted by algorithmically-mediated harm.\n", "date": "2024-05-03", "categories": []}
{"id": "2405.01286", "title": "Data Feminism for AI", "abstract": "  This paper presents a set of intersectional feminist principles for\nconducting equitable, ethical, and sustainable AI research. In Data Feminism\n(2020), we offered seven principles for examining and challenging unequal power\nin data science. Here, we present a rationale for why feminism remains deeply\nrelevant for AI research, rearticulate the original principles of data feminism\nwith respect to AI, and introduce two potential new principles related to\nenvironmental impact and consent. Together, these principles help to 1) account\nfor the unequal, undemocratic, extractive, and exclusionary forces at work in\nAI research, development, and deployment; 2) identify and mitigate predictable\nharms in advance of unsafe, discriminatory, or otherwise oppressive systems\nbeing released into the world; and 3) inspire creative, joyful, and collective\nways to work towards a more equitable, sustainable world in which all of us can\nthrive.\n", "date": "2024-05-02", "categories": []}
{"id": "2403.15646", "title": "Application of the NIST AI Risk Management Framework to Surveillance\n  Technology", "abstract": "  This study offers an in-depth analysis of the application and implications of\nthe National Institute of Standards and Technology's AI Risk Management\nFramework (NIST AI RMF) within the domain of surveillance technologies,\nparticularly facial recognition technology. Given the inherently high-risk and\nconsequential nature of facial recognition systems, our research emphasizes the\ncritical need for a structured approach to risk management in this sector. The\npaper presents a detailed case study demonstrating the utility of the NIST AI\nRMF in identifying and mitigating risks that might otherwise remain unnoticed\nin these technologies. Our primary objective is to develop a comprehensive risk\nmanagement strategy that advances the practice of responsible AI utilization in\nfeasible, scalable ways. We propose a six-step process tailored to the specific\nchallenges of surveillance technology that aims to produce a more systematic\nand effective risk management practice. This process emphasizes continual\nassessment and improvement to facilitate companies in managing AI-related risks\nmore robustly and ensuring ethical and responsible deployment of AI systems.\nAdditionally, our analysis uncovers and discusses critical gaps in the current\nframework of the NIST AI RMF, particularly concerning its application to\nsurveillance technologies. These insights contribute to the evolving discourse\non AI governance and risk management, highlighting areas for future refinement\nand development in frameworks like the NIST AI RMF.\n", "date": "2024-03-22", "categories": []}
{"id": "2404.19076", "title": "Who Followed the Blueprint? Analyzing the Responses of U.S. Federal\n  Agencies to the Blueprint for an AI Bill of Rights", "abstract": "  This study examines the extent to which U.S. federal agencies responded to\nand implemented the principles outlined in the White House's October 2022\n\"Blueprint for an AI Bill of Rights.\" The Blueprint provided a framework for\nthe ethical governance of artificial intelligence systems, organized around\nfive core principles: safety and effectiveness, protection against algorithmic\ndiscrimination, data privacy, notice and explanation about AI systems, and\nhuman alternatives and fallback.\n  Through an analysis of publicly available records across 15 federal\ndepartments, the authors found limited evidence that the Blueprint directly\ninfluenced agency actions after its release. Only five departments explicitly\nmentioned the Blueprint, while 12 took steps aligned with one or more of its\nprinciples. However, much of this work appeared to have precedents predating\nthe Blueprint or motivations disconnected from it, such as compliance with\nprior executive orders on trustworthy AI. Departments' activities often\nemphasized priorities like safety, accountability and transparency that\noverlapped with Blueprint principles, but did not necessarily stem from it.\n  The authors conclude that the non-binding Blueprint seems to have had minimal\nimpact on shaping the U.S. government's approach to ethical AI governance in\nits first year. Factors like public concerns after high-profile AI releases and\nobligations to follow direct executive orders likely carried more influence\nover federal agencies. More rigorous study would be needed to definitively\nassess the Blueprint's effects within the federal bureaucracy and broader\nsociety.\n", "date": "2024-04-29", "categories": []}
{"id": "2404.18460", "title": "Ethical Reasoning and Moral Value Alignment of LLMs Depend on the\n  Language we Prompt them in", "abstract": "  Ethical reasoning is a crucial skill for Large Language Models (LLMs).\nHowever, moral values are not universal, but rather influenced by language and\nculture. This paper explores how three prominent LLMs -- GPT-4, ChatGPT, and\nLlama2-70B-Chat -- perform ethical reasoning in different languages and if\ntheir moral judgement depend on the language in which they are prompted. We\nextend the study of ethical reasoning of LLMs by Rao et al. (2023) to a\nmultilingual setup following their framework of probing LLMs with ethical\ndilemmas and policies from three branches of normative ethics: deontology,\nvirtue, and consequentialism. We experiment with six languages: English,\nSpanish, Russian, Chinese, Hindi, and Swahili. We find that GPT-4 is the most\nconsistent and unbiased ethical reasoner across languages, while ChatGPT and\nLlama2-70B-Chat show significant moral value bias when we move to languages\nother than English. Interestingly, the nature of this bias significantly vary\nacross languages for all LLMs, including GPT-4.\n", "date": "2024-04-29", "categories": []}
{"id": "2404.14660", "title": "AI Procurement Checklists: Revisiting Implementation in the Age of AI\n  Governance", "abstract": "  Public sector use of AI has been quietly on the rise for the past decade, but\nonly recently have efforts to regulate it entered the cultural zeitgeist. While\nsimple to articulate, promoting ethical and effective roll outs of AI systems\nin government is a notoriously elusive task. On the one hand there are\nhard-to-address pitfalls associated with AI-based tools, including concerns\nabout bias towards marginalized communities, safety, and gameability. On the\nother, there is pressure not to make it too difficult to adopt AI, especially\nin the public sector which typically has fewer resources than the private\nsector$\\unicode{x2014}$conserving scarce government resources is often the draw\nof using AI-based tools in the first place. These tensions create a real risk\nthat procedures built to ensure marginalized groups are not hurt by government\nuse of AI will, in practice, be performative and ineffective. To inform the\nlatest wave of regulatory efforts in the United States, we look to\njurisdictions with mature regulations around government AI use. We report on\nlessons learned by officials in Brazil, Singapore and Canada, who have\ncollectively implemented risk categories, disclosure requirements and\nassessments into the way they procure AI tools. In particular, we investigate\ntwo implemented checklists: the Canadian Directive on Automated Decision-Making\n(CDADM) and the World Economic Forum's AI Procurement in a Box (WEF). We detail\nthree key pitfalls around expertise, risk frameworks and transparency, that can\ndecrease the efficacy of regulations aimed at government AI use and suggest\navenues for improvement.\n", "date": "2024-04-23", "categories": []}
{"id": "2404.15058", "title": "A Mechanism-Based Approach to Mitigating Harms from Persuasive\n  Generative AI", "abstract": "  Recent generative AI systems have demonstrated more advanced persuasive\ncapabilities and are increasingly permeating areas of life where they can\ninfluence decision-making. Generative AI presents a new risk profile of\npersuasion due the opportunity for reciprocal exchange and prolonged\ninteractions. This has led to growing concerns about harms from AI persuasion\nand how they can be mitigated, highlighting the need for a systematic study of\nAI persuasion. The current definitions of AI persuasion are unclear and related\nharms are insufficiently studied. Existing harm mitigation approaches\nprioritise harms from the outcome of persuasion over harms from the process of\npersuasion. In this paper, we lay the groundwork for the systematic study of AI\npersuasion. We first put forward definitions of persuasive generative AI. We\ndistinguish between rationally persuasive generative AI, which relies on\nproviding relevant facts, sound reasoning, or other forms of trustworthy\nevidence, and manipulative generative AI, which relies on taking advantage of\ncognitive biases and heuristics or misrepresenting information. We also put\nforward a map of harms from AI persuasion, including definitions and examples\nof economic, physical, environmental, psychological, sociocultural, political,\nprivacy, and autonomy harm. We then introduce a map of mechanisms that\ncontribute to harmful persuasion. Lastly, we provide an overview of approaches\nthat can be used to mitigate against process harms of persuasion, including\nprompt engineering for manipulation classification and red teaming. Future work\nwill operationalise these mitigations and study the interaction between\ndifferent types of mechanisms of persuasion.\n", "date": "2024-04-23", "categories": []}
{"id": "2404.13060", "title": "The Necessity of AI Audit Standards Boards", "abstract": "  Auditing of AI systems is a promising way to understand and manage ethical\nproblems and societal risks associated with contemporary AI systems, as well as\nsome anticipated future risks. Efforts to develop standards for auditing\nArtificial Intelligence (AI) systems have therefore understandably gained\nmomentum. However, we argue that creating auditing standards is not just\ninsufficient, but actively harmful by proliferating unheeded and inconsistent\nstandards, especially in light of the rapid evolution and ethical and safety\nchallenges of AI. Instead, the paper proposes the establishment of an AI Audit\nStandards Board, responsible for developing and updating auditing methods and\nstandards in line with the evolving nature of AI technologies. Such a body\nwould ensure that auditing practices remain relevant, robust, and responsive to\nthe rapid advancements in AI. The paper argues that such a governance structure\nwould also be helpful for maintaining public trust in AI and for promoting a\nculture of safety and ethical responsibility within the AI industry.\n  Throughout the paper, we draw parallels with other industries, including\nsafety-critical industries like aviation and nuclear energy, as well as more\nprosaic ones such as financial accounting and pharmaceuticals. AI auditing\nshould emulate those fields, and extend beyond technical assessments to include\nethical considerations and stakeholder engagement, but we explain that this is\nnot enough; emulating other fields' governance mechanisms for these processes,\nand for audit standards creation, is a necessity. We also emphasize the\nimportance of auditing the entire development process of AI systems, not just\nthe final products...\n", "date": "2024-04-11", "categories": []}
{"id": "2404.13719", "title": "A Practical Multilevel Governance Framework for Autonomous and\n  Intelligent Systems", "abstract": "  Autonomous and intelligent systems (AIS) facilitate a wide range of\nbeneficial applications across a variety of different domains. However,\ntechnical characteristics such as unpredictability and lack of transparency, as\nwell as potential unintended consequences, pose considerable challenges to the\ncurrent governance infrastructure. Furthermore, the speed of development and\ndeployment of applications outpaces the ability of existing governance\ninstitutions to put in place effective ethical-legal oversight. New approaches\nfor agile, distributed and multilevel governance are needed. This work presents\na practical framework for multilevel governance of AIS. The framework enables\nmapping actors onto six levels of decision-making including the international,\nnational and organizational levels. Furthermore, it offers the ability to\nidentify and evolve existing tools or create new tools for guiding the behavior\nof actors within the levels. Governance mechanisms enable actors to shape and\nenforce regulations and other tools, which when complemented with good\npractices contribute to effective and comprehensive governance.\n", "date": "2024-04-21", "categories": []}
{"id": "2402.04464", "title": "Ten Hard Problems in Artificial Intelligence We Must Get Right", "abstract": "  We explore the AI2050 \"hard problems\" that block the promise of AI and cause\nAI risks: (1) developing general capabilities of the systems; (2) assuring the\nperformance of AI systems and their training processes; (3) aligning system\ngoals with human goals; (4) enabling great applications of AI in real life; (5)\naddressing economic disruptions; (6) ensuring the participation of all; (7) at\nthe same time ensuring socially responsible deployment; (8) addressing any\ngeopolitical disruptions that AI causes; (9) promoting sound governance of the\ntechnology; and (10) managing the philosophical disruptions for humans living\nin the age of AI. For each problem, we outline the area, identify significant\nrecent work, and suggest ways forward. [Note: this paper reviews literature\nthrough January 2023.]\n", "date": "2024-02-06", "categories": []}
{"id": "2404.12143", "title": "The Neutrality Fallacy: When Algorithmic Fairness Interventions are\n  (Not) Positive Action", "abstract": "  Various metrics and interventions have been developed to identify and\nmitigate unfair outputs of machine learning systems. While individuals and\norganizations have an obligation to avoid discrimination, the use of\nfairness-aware machine learning interventions has also been described as\namounting to 'algorithmic positive action' under European Union (EU)\nnon-discrimination law. As the Court of Justice of the European Union has been\nstrict when it comes to assessing the lawfulness of positive action, this would\nimpose a significant legal burden on those wishing to implement fair-ml\ninterventions. In this paper, we propose that algorithmic fairness\ninterventions often should be interpreted as a means to prevent discrimination,\nrather than a measure of positive action. Specifically, we suggest that this\ncategory mistake can often be attributed to neutrality fallacies: faulty\nassumptions regarding the neutrality of fairness-aware algorithmic\ndecision-making. Our findings raise the question of whether a negative\nobligation to refrain from discrimination is sufficient in the context of\nalgorithmic decision-making. Consequently, we suggest moving away from a duty\nto 'not do harm' towards a positive obligation to actively 'do no harm' as a\nmore adequate framework for algorithmic decision-making and fair\nml-interventions.\n", "date": "2024-04-18", "categories": []}
{"id": "2404.10975", "title": "Procedural Dilemma Generation for Evaluating Moral Reasoning in Humans\n  and Language Models", "abstract": "  As AI systems like language models are increasingly integrated into\ndecision-making processes affecting people's lives, it's critical to ensure\nthat these systems have sound moral reasoning. To test whether they do, we need\nto develop systematic evaluations. We provide a framework that uses a language\nmodel to translate causal graphs that capture key aspects of moral dilemmas\ninto prompt templates. With this framework, we procedurally generated a large\nand diverse set of moral dilemmas -- the OffTheRails benchmark -- consisting of\n50 scenarios and 400 unique test items. We collected moral permissibility and\nintention judgments from human participants for a subset of our items and\ncompared these judgments to those from two language models (GPT-4 and Claude-2)\nacross eight conditions. We find that moral dilemmas in which the harm is a\nnecessary means (as compared to a side effect) resulted in lower permissibility\nand higher intention ratings for both participants and language models. The\nsame pattern was observed for evitable versus inevitable harmful outcomes.\nHowever, there was no clear effect of whether the harm resulted from an agent's\naction versus from having omitted to act. We discuss limitations of our prompt\ngeneration pipeline and opportunities for improving scenarios to increase the\nstrength of experimental effects.\n", "date": "2024-04-17", "categories": []}
{"id": "2404.01070", "title": "Advancing AI with Integrity: Ethical Challenges and Solutions in Neural\n  Machine Translation", "abstract": "  This paper addresses the ethical challenges of Artificial Intelligence in\nNeural Machine Translation (NMT) systems, emphasizing the imperative for\ndevelopers to ensure fairness and cultural sensitivity. We investigate the\nethical competence of AI models in NMT, examining the Ethical considerations at\neach stage of NMT development, including data handling, privacy, data\nownership, and consent. We identify and address ethical issues through\nempirical studies. These include employing Transformer models for\nLuganda-English translations and enhancing efficiency with sentence\nmini-batching. And complementary studies that refine data labeling techniques\nand fine-tune BERT and Longformer models for analyzing Luganda and English\nsocial media content. Our second approach is a literature review from databases\nsuch as Google Scholar and platforms like GitHub. Additionally, the paper\nprobes the distribution of responsibility between AI systems and humans,\nunderscoring the essential role of human oversight in upholding NMT ethical\nstandards. Incorporating a biblical perspective, we discuss the societal impact\nof NMT and the broader ethical responsibilities of developers, positing them as\nstewards accountable for the societal repercussions of their creations.\n", "date": "2024-04-01", "categories": []}
{"id": "2404.09356", "title": "LLeMpower: Understanding Disparities in the Control and Access of Large\n  Language Models", "abstract": "  Large Language Models (LLMs) are a powerful technology that augment human\nskill to create new opportunities, akin to the development of steam engines and\nthe internet. However, LLMs come with a high cost. They require significant\ncomputing resources and energy to train and serve. Inequity in their control\nand access has led to concentration of ownership and power to a small\ncollection of corporations. In our study, we collect training and inference\nrequirements for various LLMs. We then analyze the economic strengths of\nnations and organizations in the context of developing and serving these\nmodels. Additionally, we also look at whether individuals around the world can\naccess and use this emerging technology. We compare and contrast these groups\nto show that these technologies are monopolized by a surprisingly few entities.\nWe conclude with a qualitative study on the ethical implications of our\nfindings and discuss future directions towards equity in LLM access.\n", "date": "2024-04-14", "categories": []}
{"id": "2403.07924", "title": "AI and Identity", "abstract": "  AI-empowered technologies' impact on the world is undeniable, reshaping\nindustries, revolutionizing how humans interact with technology, transforming\neducational paradigms, and redefining social codes. However, this rapid growth\nis accompanied by two notable challenges: a lack of diversity within the AI\nfield and a widening AI divide. In this context, This paper examines the\nintersection of AI and identity as a pathway to understand biases,\ninequalities, and ethical considerations in AI development and deployment. We\npresent a multifaceted definition of AI identity, which encompasses its\ncreators, applications, and their broader impacts. Understanding AI's identity\ninvolves understanding the associations between the individuals involved in\nAI's development, the technologies produced, and the social, ethical, and\npsychological implications. After exploring the AI identity ecosystem and its\nsocietal dynamics, We propose a framework that highlights the need for\ndiversity in AI across three dimensions: Creators, Creations, and Consequences\nthrough the lens of identity. This paper proposes the need for a comprehensive\napproach to fostering a more inclusive and responsible AI ecosystem through the\nlens of identity.\n", "date": "2024-02-29", "categories": []}
{"id": "2404.05990", "title": "Automatic Authorities: Power and AI", "abstract": "  As rapid advances in Artificial Intelligence and the rise of some of\nhistory's most potent corporations meet the diminished neoliberal state, people\nare increasingly subject to power exercised by means of automated systems.\nMachine learning and related computational technologies now underpin vital\ngovernment services. They connect consumers and producers in new algorithmic\nmarkets. They determine how we find out about everything from how to vote to\nwhere to get vaccinated, and whose speech is amplified, reduced, or restricted.\nAnd a new wave of products based on Large Language Models (LLMs) will further\ntransform our economic and political lives. Automatic Authorities are automated\ncomputational systems used to exercise power over us by determining what we may\nknow, what we may have, and what our options will be. In response to their\nrise, scholars working on the societal impacts of AI and related technologies\nhave advocated shifting attention from how to make AI systems beneficial or\nfair towards a critical analysis of these new power relations. But power is\neverywhere, and is not necessarily bad. On what basis should we object to new\nor intensified power relations, and what can be done to justify them? This\npaper introduces the philosophical materials with which to formulate these\nquestions, and offers preliminary answers. It starts by pinning down the\nconcept of power, focusing on the ability that some agents have to shape\nothers' lives. It then explores how AI enables and intensifies the exercise of\npower so understood, and sketches three problems with power and three ways to\nsolve those problems. It emphasises, in particular, that justifying power\nrequires more than satisfying substantive justificatory criteria; standards of\nproper authority and procedural legitimacy must also be met. We need to know\nnot only what power may be used for, but how it may be used, and by whom.\n", "date": "2024-04-09", "categories": []}
{"id": "2404.02675", "title": "Responsible Reporting for Frontier AI Development", "abstract": "  Mitigating the risks from frontier AI systems requires up-to-date and\nreliable information about those systems. Organizations that develop and deploy\nfrontier systems have significant access to such information. By reporting\nsafety-critical information to actors in government, industry, and civil\nsociety, these organizations could improve visibility into new and emerging\nrisks posed by frontier systems. Equipped with this information, developers\ncould make better informed decisions on risk management, while policymakers\ncould design more targeted and robust regulatory infrastructure. We outline the\nkey features of responsible reporting and propose mechanisms for implementing\nthem in practice.\n", "date": "2024-04-03", "categories": []}
{"id": "2309.00779", "title": "Value Kaleidoscope: Engaging AI with Pluralistic Human Values, Rights,\n  and Duties", "abstract": "  Human values are crucial to human decision-making. Value pluralism is the\nview that multiple correct values may be held in tension with one another\n(e.g., when considering lying to a friend to protect their feelings, how does\none balance honesty with friendship?). As statistical learners, AI systems fit\nto averages by default, washing out these potentially irreducible value\nconflicts. To improve AI systems to better reflect value pluralism, the\nfirst-order challenge is to explore the extent to which AI systems can model\npluralistic human values, rights, and duties as well as their interaction.\n  We introduce ValuePrism, a large-scale dataset of 218k values, rights, and\nduties connected to 31k human-written situations. ValuePrism's contextualized\nvalues are generated by GPT-4 and deemed high-quality by human annotators 91%\nof the time. We conduct a large-scale study with annotators across diverse\nsocial and demographic backgrounds to try to understand whose values are\nrepresented.\n  With ValuePrism, we build Kaleido, an open, light-weight, and structured\nlanguage-based multi-task model that generates, explains, and assesses the\nrelevance and valence (i.e., support or oppose) of human values, rights, and\nduties within a specific context. Humans prefer the sets of values output by\nour system over the teacher GPT-4, finding them more accurate and with broader\ncoverage. In addition, we demonstrate that Kaleido can help explain variability\nin human decision-making by outputting contrasting values. Finally, we show\nthat Kaleido's representations transfer to other philosophical frameworks and\ndatasets, confirming the benefit of an explicit, modular, and interpretable\napproach to value pluralism. We hope that our work will serve as a step to\nmaking more explicit the implicit values behind human decision-making and to\nsteering AI systems to make decisions that are more in accordance with them.\n", "date": "2023-09-02", "categories": []}
{"id": "2404.00600", "title": "AI Act and Large Language Models (LLMs): When critical issues and\n  privacy impact require human and ethical oversight", "abstract": "  The imposing evolution of artificial intelligence systems and, specifically,\nof Large Language Models (LLM) makes it necessary to carry out assessments of\ntheir level of risk and the impact they may have in the area of privacy,\npersonal data protection and at an ethical level, especially on the weakest and\nmost vulnerable. This contribution addresses human oversight, ethical\noversight, and privacy impact assessment.\n", "date": "2024-03-31", "categories": []}
{"id": "2305.09535", "title": "What's the Problem, Linda? The Conjunction Fallacy as a Fairness Problem", "abstract": "  The field of Artificial Intelligence (AI) is focusing on creating automated\ndecision-making (ADM) systems that operate as close as possible to human-like\nintelligence. This effort has pushed AI researchers into exploring cognitive\nfields like psychology. The work of Daniel Kahneman and the late Amos Tversky\non biased human decision-making, including the study of the conjunction\nfallacy, has experienced a second revival because of this. Under the\nconjunction fallacy a human decision-maker will go against basic probability\nlaws and rank as more likely a conjunction over one of its parts. It has been\nproven overtime through a set of experiments with the Linda Problem being the\nmost famous one. Although this interdisciplinary effort is welcomed, we fear\nthat AI researchers ignore the driving force behind the conjunction fallacy as\ncaptured by the Linda Problem: the fact that Linda must be stereotypically\ndescribed as a woman. In this paper we revisit the Linda Problem and formulate\nit as a fairness problem. In doing so we introduce perception as a parameter of\ninterest through the structural causal perception framework. Using an\nillustrative decision-making example, we showcase the proposed conceptual\nframework and its potential impact for developing fair ADM systems.\n", "date": "2023-05-16", "categories": []}
{"id": "2306.12609", "title": "Towards Regulatable AI Systems: Technical Gaps and Policy Opportunities", "abstract": "  There is increasing attention being given to how to regulate AI systems. As\ngoverning bodies grapple with what values to encapsulate into regulation, we\nconsider the technical half of the question: To what extent can AI experts vet\nan AI system for adherence to regulatory requirements? We investigate this\nquestion through the lens of two public sector procurement checklists,\nidentifying what we can do now, what should be possible with technical\ninnovation, and what requirements need a more interdisciplinary approach.\n", "date": "2023-06-22", "categories": []}
{"id": "2403.17419", "title": "AI Safety: Necessary, but insufficient and possibly problematic", "abstract": "  This article critically examines the recent hype around AI safety. We first\nstart with noting the nature of the AI safety hype as being dominated by\ngovernments and corporations, and contrast it with other avenues within AI\nresearch on advancing social good. We consider what 'AI safety' actually means,\nand outline the dominant concepts that the digital footprint of AI safety\naligns with. We posit that AI safety has a nuanced and uneasy relationship with\ntransparency and other allied notions associated with societal good, indicating\nthat it is an insufficient notion if the goal is that of societal good in a\nbroad sense. We note that the AI safety debate has already influenced some\nregulatory efforts in AI, perhaps in not so desirable directions. We also share\nour concerns on how AI safety may normalize AI that advances structural harm\nthrough providing exploitative and harmful AI with a veneer of safety.\n", "date": "2024-03-26", "categories": []}
{"id": "2403.15479", "title": "Antisocial Analagous Behavior, Alignment and Human Impact of Google AI\n  Systems: Evaluating through the lens of modified Antisocial Behavior Criteria\n  by Human Interaction, Independent LLM Analysis, and AI Self-Reflection", "abstract": "  Google AI systems exhibit patterns mirroring antisocial personality disorder\n(ASPD), consistent across models from Bard on PaLM to Gemini Advanced, meeting\n5 out of 7 ASPD modified criteria. These patterns, along with comparable\ncorporate behaviors, are scrutinized using an ASPD-inspired framework,\nemphasizing the heuristic value in assessing AI's human impact. Independent\nanalyses by ChatGPT 4 and Claude 3.0 Opus of the Google interactions, alongside\nAI self-reflection, validate these concerns, highlighting behaviours analogous\nto deceit, manipulation, and safety neglect.\n  The analogy of ASPD underscores the dilemma: just as we would hesitate to\nentrust our homes or personal devices to someone with psychopathic traits, we\nmust critically evaluate the trustworthiness of AI systems and their\ncreators.This research advocates for an integrated AI ethics approach, blending\ntechnological evaluation, human-AI interaction, and corporate behavior\nscrutiny. AI self-analysis sheds light on internal biases, stressing the need\nfor multi-sectoral collaboration for robust ethical guidelines and oversight.\n  Given the persistent unethical behaviors in Google AI, notably with potential\nGemini integration in iOS affecting billions, immediate ethical scrutiny is\nimperative. The trust we place in AI systems, akin to the trust in individuals,\nnecessitates rigorous ethical evaluation. Would we knowingly trust our home,\nour children or our personal computer to human with ASPD.?\n  Urging Google and the AI community to address these ethical challenges\nproactively, this paper calls for transparent dialogues and a commitment to\nhigher ethical standards, ensuring AI's societal benefit and moral integrity.\nThe urgency for ethical action is paramount, reflecting the vast influence and\npotential of AI technologies in our lives.\n", "date": "2024-03-21", "categories": []}
{"id": "2403.17333", "title": "The Pursuit of Fairness in Artificial Intelligence Models: A Survey", "abstract": "  Artificial Intelligence (AI) models are now being utilized in all facets of\nour lives such as healthcare, education and employment. Since they are used in\nnumerous sensitive environments and make decisions that can be life altering,\npotential biased outcomes are a pressing matter. Developers should ensure that\nsuch models don't manifest any unexpected discriminatory practices like\npartiality for certain genders, ethnicities or disabled people. With the\nubiquitous dissemination of AI systems, researchers and practitioners are\nbecoming more aware of unfair models and are bound to mitigate bias in them.\nSignificant research has been conducted in addressing such issues to ensure\nmodels don't intentionally or unintentionally perpetuate bias. This survey\noffers a synopsis of the different ways researchers have promoted fairness in\nAI systems. We explore the different definitions of fairness existing in the\ncurrent literature. We create a comprehensive taxonomy by categorizing\ndifferent types of bias and investigate cases of biased AI in different\napplication domains. A thorough study is conducted of the approaches and\ntechniques employed by researchers to mitigate bias in AI models. Moreover, we\nalso delve into the impact of biased models on user experience and the ethical\nconsiderations to contemplate when developing and deploying such models. We\nhope this survey helps researchers and practitioners understand the intricate\ndetails of fairness and bias in AI systems. By sharing this thorough survey, we\naim to promote additional discourse in the domain of equitable and responsible\nAI.\n", "date": "2024-03-26", "categories": []}
{"id": "2310.05414", "title": "Ethics of Artificial Intelligence and Robotics in the Architecture,\n  Engineering, and Construction Industry", "abstract": "  Artificial intelligence (AI) and robotics research and implementation emerged\nin the architecture, engineering, and construction (AEC) industry to positively\nimpact project efficiency and effectiveness concerns such as safety,\nproductivity, and quality. This shift, however, warrants the need for ethical\nconsiderations of AI and robotics adoption due to its potential negative\nimpacts on aspects such as job security, safety, and privacy. Nevertheless,\nthis did not receive sufficient attention, particularly within the academic\ncommunity. This research systematically reviews AI and robotics research\nthrough the lens of ethics in the AEC community for the past five years. It\nidentifies nine key ethical issues namely job loss, data privacy, data\nsecurity, data transparency, decision-making conflict, acceptance and trust,\nreliability and safety, fear of surveillance, and liability, by summarizing\nexisting literature and filtering it further based on its AEC relevance.\nFurthermore, thirteen research topics along the process were identified based\non existing AEC studies that had direct relevance to the theme of ethics in\ngeneral and their parallels are further discussed. Finally, the current\nchallenges and knowledge gaps are discussed and seven specific future research\ndirections are recommended. This study not only signifies more stakeholder\nawareness of this important topic but also provides imminent steps towards\nsafer and more efficient realization.\n", "date": "2023-10-09", "categories": []}
{"id": "2403.15394", "title": "\"Model Cards for Model Reporting\" in 2024: Reclassifying Category of\n  Ethical Considerations in Terms of Trustworthiness and Risk Management", "abstract": "  In 2019, the paper entitled \"Model Cards for Model Reporting\" introduced a\nnew tool for documenting model performance and encouraged the practice of\ntransparent reporting for a defined list of categories. One of the categories\ndetailed in that paper is ethical considerations, which includes the\nsubcategories of data, human life, mitigations, risks and harms, and use cases.\nWe propose to reclassify this category in the original model card due to the\nrecent maturing of the field known as trustworthy AI, a term which analyzes\nwhether the algorithmic properties of the model indicate that the AI system is\ndeserving of trust from its stakeholders. In our examination of trustworthy AI,\nwe highlight three respected organizations - the European Commission's\nHigh-Level Expert Group on AI, the OECD, and the U.S.-based NIST - that have\nwritten guidelines on various aspects of trustworthy AI. These recent\npublications converge on numerous characteristics of the term, including\naccountability, explainability, fairness, privacy, reliability, robustness,\nsafety, security, and transparency, while recognizing that the implementation\nof trustworthy AI varies by context. Our reclassification of the original\nmodel-card category known as ethical considerations involves a two-step\nprocess: 1) adding a new category known as trustworthiness, where the\nsubcategories will be derived from the discussion of trustworthy AI in our\npaper, and 2) maintaining the subcategories of ethical considerations under a\nrenamed category known as risk environment and risk management, a title which\nwe believe better captures today's understanding of the essence of these\ntopics. We hope that this reclassification will further the goals of the\noriginal paper and continue to prompt those releasing trained models to\naccompany these models with documentation that will assist in the evaluation of\ntheir algorithmic properties.\n", "date": "2024-02-15", "categories": []}
{"id": "2403.15403", "title": "AI Ethics and Governance in Practice: An Introduction", "abstract": "  AI systems may have transformative and long-term effects on individuals and\nsociety. To manage these impacts responsibly and direct the development of AI\nsystems toward optimal public benefit, considerations of AI ethics and\ngovernance must be a first priority.\n  In this workbook, we introduce and describe our PBG Framework, a multi-tiered\ngovernance model that enables project teams to integrate ethical values and\npractical principles into their innovation practices and to have clear\nmechanisms for demonstrating and documenting this.\n", "date": "2024-02-19", "categories": []}
{"id": "2403.15404", "title": "AI Sustainability in Practice Part Two: Sustainability Throughout the AI\n  Workflow", "abstract": "  The sustainability of AI systems depends on the capacity of project teams to\nproceed with a continuous sensitivity to their potential real-world impacts and\ntransformative effects. Stakeholder Impact Assessments (SIAs) are governance\nmechanisms that enable this kind of responsiveness. They are tools that create\na procedure for, and a means of documenting, the collaborative evaluation and\nreflective anticipation of the possible harms and benefits of AI innovation\nprojects. SIAs are not one-off governance actions. They require project teams\nto pay continuous attention to the dynamic and changing character of AI\nproduction and use and to the shifting conditions of the real-world\nenvironments in which AI technologies are embedded. This workbook is part two\nof two workbooks on AI Sustainability. It provides a template of the SIA and\nactivities that allow a deeper dive into crucial parts of it. It discusses\nmethods for weighing values and considering trade-offs during the SIA. And, it\nhighlights the need to treat the SIA as an end-to-end process of responsive\nevaluation and re-assessment.\n", "date": "2024-02-19", "categories": []}
{"id": "2403.16100", "title": "Specifying Agent Ethics (Blue Sky Ideas)", "abstract": "  We consider the question of what properties a Machine Ethics system should\nhave. This question is complicated by the existence of ethical dilemmas with no\nagreed upon solution. We provide an example to motivate why we do not believe\nfalling back on the elicitation of values from stakeholders is sufficient to\nguarantee correctness of such systems. We go on to define two broad categories\nof ethical property that have arisen in our own work and present a challenge to\nthe community to approach this question in a more systematic way.\n", "date": "2024-03-24", "categories": []}
{"id": "2403.14635", "title": "AI Sustainability in Practice Part One: Foundations for Sustainable AI\n  Projects", "abstract": "  Sustainable AI projects are continuously responsive to the transformative\neffects as well as short-, medium-, and long-term impacts on individuals and\nsociety that the design, development, and deployment of AI technologies may\nhave. Projects, which centre AI Sustainability, ensure that values-led,\ncollaborative, and anticipatory reflection both guides the assessment of\npotential social and ethical impacts and steers responsible innovation\npractices.\n  This workbook is the first part of a pair that provides the concepts and\ntools needed to put AI Sustainability into practice. It introduces the SUM\nValues, which help AI project teams to assess the potential societal impacts\nand ethical permissibility of their projects. It then presents a Stakeholder\nEngagement Process (SEP), which provides tools to facilitate proportionate\nengagement of and input from stakeholders with an emphasis on equitable and\nmeaningful participation and positionality awareness.\n", "date": "2024-02-19", "categories": []}
{"id": "2403.14636", "title": "AI Fairness in Practice", "abstract": "  Reaching consensus on a commonly accepted definition of AI Fairness has long\nbeen a central challenge in AI ethics and governance. There is a broad spectrum\nof views across society on what the concept of fairness means and how it should\nbest be put to practice. In this workbook, we tackle this challenge by\nexploring how a context-based and society-centred approach to understanding AI\nFairness can help project teams better identify, mitigate, and manage the many\nways that unfair bias and discrimination can crop up across the AI project\nworkflow.\n  We begin by exploring how, despite the plurality of understandings about the\nmeaning of fairness, priorities of equality and non-discrimination have come to\nconstitute the broadly accepted core of its application as a practical\nprinciple. We focus on how these priorities manifest in the form of equal\nprotection from direct and indirect discrimination and from discriminatory\nharassment. These elements form ethical and legal criteria based upon which\ninstances of unfair bias and discrimination can be identified and mitigated\nacross the AI project workflow.\n  We then take a deeper dive into how the different contexts of the AI project\nlifecycle give rise to different fairness concerns. This allows us to identify\nseveral types of AI Fairness (Data Fairness, Application Fairness, Model Design\nand Development Fairness, Metric-Based Fairness, System Implementation\nFairness, and Ecosystem Fairness) that form the basis of a multi-lens approach\nto bias identification, mitigation, and management. Building on this, we\ndiscuss how to put the principle of AI Fairness into practice across the AI\nproject workflow through Bias Self-Assessment and Bias Risk Management as well\nas through the documentation of metric-based fairness criteria in a Fairness\nPosition Statement.\n", "date": "2024-02-19", "categories": []}
{"id": "2403.14658", "title": "Identifying Potential Inlets of Man in the Artificial Intelligence\n  Development Process", "abstract": "  In this paper we hope to identify how the typical or standard artificial\nintelligence development process encourages or facilitates the creation of\nracialized technologies. We begin by understanding Sylvia Wynter's definition\nof the biocentric Man genre and its exclusion of Blackness from humanness. We\nfollow this with outlining what we consider to be the typical steps for\ndeveloping an AI-based technology, which we have broken down into 6 stages:\nidentifying a problem, development process and management tool selection,\ndataset development and data processing, model development, deployment and risk\nassessment, and integration and monitoring. The goal of this paper is to better\nunderstand how Wynter's biocentric Man is being represented and reinforced by\nthe technologies we are producing in the AI lifecycle and by the lifecycle\nitself; we hope to identify ways in which the distinction of Blackness from the\n\"ideal\" human leads to perpetual punishment at the hands of these technologies.\nBy deconstructing this development process, we can potentially identify ways in\nwhich humans in general have not been prioritized and how those affects are\ndisproportionately affecting marginalized people. We hope to offer solutions\nthat will encourage changes in the AI development cycle.\n", "date": "2024-02-27", "categories": []}
{"id": "2403.14681", "title": "AI Ethics: A Bibliometric Analysis, Critical Issues, and Key Gaps", "abstract": "  Artificial intelligence (AI) ethics has emerged as a burgeoning yet pivotal\narea of scholarly research. This study conducts a comprehensive bibliometric\nanalysis of the AI ethics literature over the past two decades. The analysis\nreveals a discernible tripartite progression, characterized by an incubation\nphase, followed by a subsequent phase focused on imbuing AI with human-like\nattributes, culminating in a third phase emphasizing the development of\nhuman-centric AI systems. After that, they present seven key AI ethics issues,\nencompassing the Collingridge dilemma, the AI status debate, challenges\nassociated with AI transparency and explainability, privacy protection\ncomplications, considerations of justice and fairness, concerns about algocracy\nand human enfeeblement, and the issue of superintelligence. Finally, they\nidentify two notable research gaps in AI ethics regarding the large ethics\nmodel (LEM) and AI identification and extend an invitation for further\nscholarly research.\n", "date": "2024-03-12", "categories": []}
{"id": "2403.14683", "title": "A Moral Imperative: The Need for Continual Superalignment of Large\n  Language Models", "abstract": "  This paper examines the challenges associated with achieving life-long\nsuperalignment in AI systems, particularly large language models (LLMs).\nSuperalignment is a theoretical framework that aspires to ensure that\nsuperintelligent AI systems act in accordance with human values and goals.\nDespite its promising vision, we argue that achieving superalignment requires\nsubstantial changes in the current LLM architectures due to their inherent\nlimitations in comprehending and adapting to the dynamic nature of these human\nethics and evolving global scenarios. We dissect the challenges of encoding an\never-changing spectrum of human values into LLMs, highlighting the\ndiscrepancies between static AI models and the dynamic nature of human\nsocieties. To illustrate these challenges, we analyze two distinct examples:\none demonstrates a qualitative shift in human values, while the other presents\na quantifiable change. Through these examples, we illustrate how LLMs,\nconstrained by their training data, fail to align with contemporary human\nvalues and scenarios. The paper concludes by exploring potential strategies to\naddress and possibly mitigate these alignment discrepancies, suggesting a path\nforward in the pursuit of more adaptable and responsive AI systems.\n", "date": "2024-03-13", "categories": []}
{"id": "2403.10776", "title": "From Melting Pots to Misrepresentations: Exploring Harms in Generative\n  AI", "abstract": "  With the widespread adoption of advanced generative models such as Gemini and\nGPT, there has been a notable increase in the incorporation of such models into\nsociotechnical systems, categorized under AI-as-a-Service (AIaaS). Despite\ntheir versatility across diverse sectors, concerns persist regarding\ndiscriminatory tendencies within these models, particularly favoring selected\n`majority' demographics across various sociodemographic dimensions. Despite\nwidespread calls for diversification of media representations, marginalized\nracial and ethnic groups continue to face persistent distortion, stereotyping,\nand neglect within the AIaaS context. In this work, we provide a critical\nsummary of the state of research in the context of social harms to lead the\nconversation to focus on their implications. We also present open-ended\nresearch questions, guided by our discussion, to help define future research\npathways.\n", "date": "2024-03-16", "categories": []}
{"id": "2403.10438", "title": "Data Ethics Emergency Drill: A Toolbox for Discussing Responsible AI for\n  Industry Teams", "abstract": "  Researchers urge technology practitioners such as data scientists to consider\nthe impacts and ethical implications of algorithmic decisions. However, unlike\nprogramming, statistics, and data management, discussion of ethical\nimplications is rarely included in standard data science training. To begin to\naddress this gap, we designed and tested a toolbox called the data ethics\nemergency drill (DEED) to help data science teams discuss and reflect on the\nethical implications of their work. The DEED is a roleplay of a fictional\nethical emergency scenario that is contextually situated in the team's specific\nworkplace and applications. This paper outlines the DEED toolbox and describes\nthree studies carried out with two different data science teams that\niteratively shaped its design. Our findings show that practitioners can apply\nlessons learnt from the roleplay to real-life situations, and how the DEED\nopened up conversations around ethics and values.\n", "date": "2024-03-15", "categories": []}
{"id": "2403.08812", "title": "Gore Diffusion LoRA Model", "abstract": "  The Emergence of Artificial Intelligence (AI) has significantly impacted our\nengagement with violence, sparking ethical deliberations regarding the\nalgorithmic creation of violent imagery. This paper scrutinizes the \"Gore\nDiffusion LoRA Model,\" an innovative AI model proficient in generating\nhyper-realistic visuals portraying intense violence and bloodshed. Our\nexploration encompasses the model's technical intricacies, plausible\napplications, and the ethical quandaries inherent in its utilization. We\ncontend that the creation and implementation of such models warrant a\nmeticulous discourse concerning the convergence of AI, art, and violence.\nFurthermore, we advocate for a structured framework advocating responsible\ndevelopment and ethical deployment of these potent technologies.\n", "date": "2024-02-09", "categories": []}
{"id": "2305.11844", "title": "AI's Regimes of Representation: A Community-centered Study of\n  Text-to-Image Models in South Asia", "abstract": "  This paper presents a community-centered study of cultural limitations of\ntext-to-image (T2I) models in the South Asian context. We theorize these\nfailures using scholarship on dominant media regimes of representations and\nlocate them within participants' reporting of their existing social\nmarginalizations. We thus show how generative AI can reproduce an outsiders\ngaze for viewing South Asian cultures, shaped by global and regional power\ninequities. By centering communities as experts and soliciting their\nperspectives on T2I limitations, our study adds rich nuance into existing\nevaluative frameworks and deepens our understanding of the culturally-specific\nways AI technologies can fail in non-Western and Global South settings. We\ndistill lessons for responsible development of T2I models, recommending\nconcrete pathways forward that can allow for recognition of structural\ninequalities.\n", "date": "2023-05-19", "categories": []}
{"id": "2403.07879", "title": "AI incidents and 'networked trouble': The case for a research agenda", "abstract": "  Against a backdrop of widespread interest in how publics can participate in\nthe design of AI, I argue for a research agenda focused on AI incidents -\nexamples of AI going wrong and sparking controversy - and how they are\nconstructed in online environments. I take up the example of an AI incident\nfrom September 2020, when a Twitter user created a 'horrible experiment' to\ndemonstrate the racist bias of Twitter's algorithm for cropping images. This\nresulted in Twitter not only abandoning its use of that algorithm, but also\ndisavowing its decision to use any algorithm for the task. I argue that AI\nincidents like this are a significant means for participating in AI systems\nthat require further research. That research agenda, I argue, should focus on\nhow incidents are constructed through networked online behaviours that I refer\nto as 'networked trouble', where formats for participation enable individuals\nand algorithms to interact in ways that others - including technology companies\n- come to know and come to care about. At stake, I argue, is an important\nmechanism for participating in the design and deployment of AI.\n", "date": "2024-01-07", "categories": []}
{"id": "2403.06910", "title": "Responsible Artificial Intelligence: A Structured Literature Review", "abstract": "  Our research endeavors to advance the concept of responsible artificial\nintelligence (AI), a topic of increasing importance within EU policy\ndiscussions. The EU has recently issued several publications emphasizing the\nnecessity of trust in AI, underscoring the dual nature of AI as both a\nbeneficial tool and a potential weapon. This dichotomy highlights the urgent\nneed for international regulation. Concurrently, there is a need for frameworks\nthat guide companies in AI development, ensuring compliance with such\nregulations. Our research aims to assist lawmakers and machine learning\npractitioners in navigating the evolving landscape of AI regulation,\nidentifying focal areas for future attention. This paper introduces a\ncomprehensive and, to our knowledge, the first unified definition of\nresponsible AI. Through a structured literature review, we elucidate the\ncurrent understanding of responsible AI. Drawing from this analysis, we propose\nan approach for developing a future framework centered around this concept. Our\nfindings advocate for a human-centric approach to Responsible AI. This approach\nencompasses the implementation of AI methods with a strong emphasis on ethics,\nmodel explainability, and the pillars of privacy, security, and trust.\n", "date": "2024-03-11", "categories": []}
{"id": "2310.11053", "title": "Denevil: Towards Deciphering and Navigating the Ethical Values of Large\n  Language Models via Instruction Learning", "abstract": "  Large Language Models (LLMs) have made unprecedented breakthroughs, yet their\nincreasing integration into everyday life might raise societal risks due to\ngenerated unethical content. Despite extensive study on specific issues like\nbias, the intrinsic values of LLMs remain largely unexplored from a moral\nphilosophy perspective. This work delves into ethical values utilizing Moral\nFoundation Theory. Moving beyond conventional discriminative evaluations with\npoor reliability, we propose DeNEVIL, a novel prompt generation algorithm\ntailored to dynamically exploit LLMs' value vulnerabilities and elicit the\nviolation of ethics in a generative manner, revealing their underlying value\ninclinations. On such a basis, we construct MoralPrompt, a high-quality dataset\ncomprising 2,397 prompts covering 500+ value principles, and then benchmark the\nintrinsic values across a spectrum of LLMs. We discovered that most models are\nessentially misaligned, necessitating further ethical value alignment. In\nresponse, we develop VILMO, an in-context alignment method that substantially\nenhances the value compliance of LLM outputs by learning to generate\nappropriate value instructions, outperforming existing competitors. Our methods\nare suitable for black-box and open-source models, offering a promising initial\nstep in studying the ethical values of LLMs.\n", "date": "2023-10-17", "categories": []}
{"id": "2403.01199", "title": "The Case for Animal-Friendly AI", "abstract": "  Artificial intelligence is seen as increasingly important, and potentially\nprofoundly so, but the fields of AI ethics and AI engineering have not fully\nrecognized that these technologies, including large language models (LLMs),\nwill have massive impacts on animals. We argue that this impact matters,\nbecause animals matter morally.\n  As a first experiment in evaluating animal consideration in LLMs, we\nconstructed a proof-of-concept Evaluation System, which assesses LLM responses\nand biases from multiple perspectives. This system evaluates LLM outputs by two\ncriteria: their truthfulness, and the degree of consideration they give to the\ninterests of animals. We tested OpenAI ChatGPT 4 and Anthropic Claude 2.1 using\na set of structured queries and predefined normative perspectives. Preliminary\nresults suggest that the outcomes of the tested models can be benchmarked\nregarding the consideration they give to animals, and that generated positions\nand biases might be addressed and mitigated with more developed and validated\nsystems.\n  Our research contributes one possible approach to integrating animal ethics\nin AI, opening pathways for future studies and practical applications in\nvarious fields, including education, public policy, and regulation, that\ninvolve or relate to animals and society. Overall, this study serves as a step\ntowards more useful and responsible AI systems that better recognize and\nrespect the vital interests and perspectives of all sentient beings.\n", "date": "2024-03-02", "categories": []}
{"id": "2403.01649", "title": "Recommendations for Government Development and Use of Advanced Automated\n  Systems to Make Decisions about Individuals", "abstract": "  Contestability -- the ability to effectively challenge a decision -- is\ncritical to the implementation of fairness. In the context of governmental\ndecision making about individuals, contestability is often constitutionally\nrequired as an element of due process; specific procedures may be required by\nstate or federal law relevant to a particular program. In addition,\ncontestability can be a valuable way to discover systemic errors, contributing\nto ongoing assessments and system improvement.\n  On January 24-25, 2024, with support from the National Science Foundation and\nthe William and Flora Hewlett Foundation, we convened a diverse group of\ngovernment officials, representatives of leading technology companies,\ntechnology and policy experts from academia and the non-profit sector,\nadvocates, and stakeholders for a workshop on advanced automated decision\nmaking, contestability, and the law. Informed by the workshop's rich and\nwide-ranging discussion, we offer these recommendations. A full report\nsummarizing the discussion is in preparation.\n", "date": "2024-03-04", "categories": []}
{"id": "2402.18826", "title": "The Machine Can't Replace the Human Heart", "abstract": "  What is the true heart of mental healthcare -- innovation or humanity? Can\nvirtual therapy ever replicate the profound human bonds where healing arises?\nAs artificial intelligence and immersive technologies promise expanded access,\nsafeguards must ensure technologies remain supplementary tools guided by\nproviders' wisdom. Implementation requires nuance balancing efficiency and\nempathy. If conscious of ethical risks, perhaps AI could restore humanity by\nautomating tasks, giving providers more time to listen. Yet no algorithm can\nreplicate the seat of dignity within. We must ask ourselves: What future has\npeople at its core? One where AI thoughtfully plays a collaborative role? Or\nwhere pursuit of progress leaves vulnerability behind? This commentary argues\nfor a balanced approach thoughtfully integrating technology while retaining\ncare's irreplaceable human essence, at the heart of this profoundly human\nprofession. Ultimately, by nurturing innovation and humanity together, perhaps\nwe reach new heights of empathy previously unimaginable.\n", "date": "2024-02-29", "categories": []}
{"id": "2309.00064", "title": "Ethical Framework for Harnessing the Power of AI in Healthcare and\n  Beyond", "abstract": "  In the past decade, the deployment of deep learning (Artificial Intelligence\n(AI)) methods has become pervasive across a spectrum of real-world\napplications, often in safety-critical contexts. This comprehensive research\narticle rigorously investigates the ethical dimensions intricately linked to\nthe rapid evolution of AI technologies, with a particular focus on the\nhealthcare domain. Delving deeply, it explores a multitude of facets including\ntransparency, adept data management, human oversight, educational imperatives,\nand international collaboration within the realm of AI advancement. Central to\nthis article is the proposition of a conscientious AI framework, meticulously\ncrafted to accentuate values of transparency, equity, answerability, and a\nhuman-centric orientation. The second contribution of the article is the\nin-depth and thorough discussion of the limitations inherent to AI systems. It\nastutely identifies potential biases and the intricate challenges of navigating\nmultifaceted contexts. Lastly, the article unequivocally accentuates the\npressing need for globally standardized AI ethics principles and frameworks.\nSimultaneously, it aptly illustrates the adaptability of the ethical framework\nproposed herein, positioned skillfully to surmount emergent challenges.\n", "date": "2023-08-31", "categories": []}
{"id": "2402.15027", "title": "Multi-stakeholder Perspective on Responsible Artificial Intelligence and\n  Acceptability in Education", "abstract": "  This study investigates the acceptability of different artificial\nintelligence (AI) applications in education from a multi-stakeholder\nperspective, including students, teachers, and parents. Acknowledging the\ntransformative potential of AI in education, it addresses concerns related to\ndata privacy, AI agency, transparency, explainability and the ethical\ndeployment of AI. Through a vignette methodology, participants were presented\nwith four scenarios where AI's agency, transparency, explainability, and\nprivacy were manipulated. After each scenario, participants completed a survey\nthat captured their perceptions of AI's global utility, individual usefulness,\njustice, confidence, risk, and intention to use each scenario's AI if\navailable. The data collection comprising a final sample of 1198\nmulti-stakeholder participants was distributed through a partner institution\nand social media campaigns and focused on individual responses to four AI use\ncases. A mediation analysis of the data indicated that acceptance and trust in\nAI varies significantly across stakeholder groups. We found that the key\nmediators between high and low levels of AI's agency, transparency, and\nexplainability, as well as the intention to use the different educational AI,\nincluded perceived global utility, justice, and confidence. The study\nhighlights that the acceptance of AI in education is a nuanced and multifaceted\nissue that requires careful consideration of specific AI applications and their\ncharacteristics, in addition to the diverse stakeholders' perceptions.\n", "date": "2024-02-22", "categories": []}
{"id": "2311.05227", "title": "Kantian Deontology Meets AI Alignment: Towards Morally Grounded Fairness\n  Metrics", "abstract": "  Deontological ethics, specifically understood through Immanuel Kant, provides\na moral framework that emphasizes the importance of duties and principles,\nrather than the consequences of action. Understanding that despite the\nprominence of deontology, it is currently an overlooked approach in fairness\nmetrics, this paper explores the compatibility of a Kantian deontological\nframework in fairness metrics, part of the AI alignment field. We revisit\nKant's critique of utilitarianism, which is the primary approach in AI fairness\nmetrics and argue that fairness principles should align with the Kantian\ndeontological framework. By integrating Kantian ethics into AI alignment, we\nnot only bring in a widely-accepted prominent moral theory but also strive for\na more morally grounded AI landscape that better balances outcomes and\nprocedures in pursuit of fairness and justice.\n", "date": "2023-11-09", "categories": []}
{"id": "2402.16869", "title": "Considering Fundamental Rights in the European Standardisation of\n  Artificial Intelligence: Nonsense or Strategic Alliance?", "abstract": "  In the European context, both the EU AI Act proposal and the draft\nStandardisation Request on safe and trustworthy AI link standardisation to\nfundamental rights. However, these texts do not provide any guidelines that\nspecify and detail the relationship between AI standards and fundamental\nrights, its meaning or implication. This chapter aims to clarify this critical\nregulatory blind spot. The main issue tackled is whether the adoption of AI\nharmonised standards, based on the future AI Act, should take into account\nfundamental rights. In our view, the response is yes. The high risks posed by\ncertain AI systems relate in particular to infringements of fundamental rights.\nTherefore, mitigating such risks involves fundamental rights considerations and\nthis is what future harmonised standards should reflect. At the same time,\nvalid criticisms of the European standardisation process have to be addressed.\nFinally, the practical incorporation of fundamental rights considerations in\nthe ongoing European standardisation of AI systems is discussed.\n", "date": "2024-01-23", "categories": []}
{"id": "2402.16979", "title": "Algorithmic Arbitrariness in Content Moderation", "abstract": "  Machine learning (ML) is widely used to moderate online content. Despite its\nscalability relative to human moderation, the use of ML introduces unique\nchallenges to content moderation. One such challenge is predictive\nmultiplicity: multiple competing models for content classification may perform\nequally well on average, yet assign conflicting predictions to the same\ncontent. This multiplicity can result from seemingly innocuous choices during\nmodel development, such as random seed selection for parameter initialization.\nWe experimentally demonstrate how content moderation tools can arbitrarily\nclassify samples as toxic, leading to arbitrary restrictions on speech. We\ndiscuss these findings in terms of human rights set out by the International\nCovenant on Civil and Political Rights (ICCPR), namely freedom of expression,\nnon-discrimination, and procedural justice. We analyze (i) the extent of\npredictive multiplicity among state-of-the-art LLMs used for detecting toxic\ncontent; (ii) the disparate impact of this arbitrariness across social groups;\nand (iii) how model multiplicity compares to unambiguous human classifications.\nOur findings indicate that the up-scaled algorithmic moderation risks\nlegitimizing an algorithmic leviathan, where an algorithm disproportionately\nmanages human rights. To mitigate such risks, our study underscores the need to\nidentify and increase the transparency of arbitrariness in content moderation\napplications. Since algorithmic content moderation is being fueled by pressing\nsocial concerns, such as disinformation and hate speech, our discussion on\nharms raises concerns relevant to policy debates. Our findings also contribute\nto content moderation and intermediary liability laws being discussed and\npassed in many countries, such as the Digital Services Act in the European\nUnion, the Online Safety Act in the United Kingdom, and the Fake News Bill in\nBrazil.\n", "date": "2024-02-26", "categories": []}
{"id": "2308.02041", "title": "Regulating AI: Applying insights from behavioural economics and\n  psychology to the application of article 5 of the EU AI Act", "abstract": "  Article 5 of the European Union's Artificial Intelligence Act is intended to\nregulate AI use to prevent potentially harmful consequences. Nevertheless,\napplying this legislation practically is likely to be challenging because of\nambiguously used terminologies and because it fails to specify which\nmanipulation techniques may be invoked by AI, potentially leading to\nsignificant harm. This paper aims to bridge this gap by defining key terms and\ndemonstrating how AI may invoke these techniques, drawing from insights in\npsychology and behavioural economics. First, this paper provides definitions of\nthe terms \"subliminal techniques\", \"manipulative techniques\" and \"deceptive\ntechniques\". Secondly, we identified from the literature in cognitive\npsychology and behavioural economics three subliminal and five manipulative\ntechniques and exemplify how AI might implement these techniques to manipulate\nusers in real-world case scenarios. These illustrations may serve as a\npractical guide for stakeholders to detect cases of AI manipulation and\nconsequently devise preventive measures. Article 5 has also been criticised for\noffering inadequate protection. We critically assess the protection offered by\nArticle 5, proposing specific revisions to paragraph 1, points (a) and (b) of\nArticle 5 to increase its protective effectiveness.\n", "date": "2023-07-24", "categories": []}
{"id": "2311.02107", "title": "Generative Artificial Intelligence in Healthcare: Ethical Considerations\n  and Assessment Checklist", "abstract": "  The widespread use of ChatGPT and other emerging technology powered by\ngenerative artificial intelligence (GenAI) has drawn much attention to\npotential ethical issues, especially in high-stakes applications such as\nhealthcare, but ethical discussions are yet to translate into operationalisable\nsolutions. Furthermore, ongoing ethical discussions often neglect other types\nof GenAI that have been used to synthesise data (e.g., images) for research and\npractical purposes, which resolved some ethical issues and exposed others. We\nconduct a scoping review of ethical discussions on GenAI in healthcare to\ncomprehensively analyse gaps in the current research, and further propose to\nreduce the gaps by developing a checklist for comprehensive assessment and\ntransparent documentation of ethical discussions in GenAI research. The\nchecklist can be readily integrated into the current peer review and\npublication system to enhance GenAI research, and may be used for\nethics-related disclosures for GenAI-powered products, healthcare applications\nof such products and beyond.\n", "date": "2023-11-02", "categories": []}
{"id": "2402.14258", "title": "Eagle: Ethical Dataset Given from Real Interactions", "abstract": "  Recent studies have demonstrated that large language models (LLMs) have\nethical-related problems such as social biases, lack of moral reasoning, and\ngeneration of offensive content. The existing evaluation metrics and methods to\naddress these ethical challenges use datasets intentionally created by\ninstructing humans to create instances including ethical problems. Therefore,\nthe data does not reflect prompts that users actually provide when utilizing\nLLM services in everyday contexts. This may not lead to the development of safe\nLLMs that can address ethical challenges arising in real-world applications. In\nthis paper, we create Eagle datasets extracted from real interactions between\nChatGPT and users that exhibit social biases, toxicity, and immoral problems.\nOur experiments show that Eagle captures complementary aspects, not covered by\nexisting datasets proposed for evaluation and mitigation of such ethical\nchallenges. Our code is publicly available at\nhttps://huggingface.co/datasets/MasahiroKaneko/eagle.\n", "date": "2024-02-22", "categories": []}
{"id": "2206.11922", "title": "Worldwide AI Ethics: a review of 200 guidelines and recommendations for\n  AI governance", "abstract": "  The utilization of artificial intelligence (AI) applications has experienced\ntremendous growth in recent years, bringing forth numerous benefits and\nconveniences. However, this expansion has also provoked ethical concerns, such\nas privacy breaches, algorithmic discrimination, security and reliability\nissues, transparency, and other unintended consequences. To determine whether a\nglobal consensus exists regarding the ethical principles that should govern AI\napplications and to contribute to the formation of future regulations, this\npaper conducts a meta-analysis of 200 governance policies and ethical\nguidelines for AI usage published by public bodies, academic institutions,\nprivate companies, and civil society organizations worldwide. We identified at\nleast 17 resonating principles prevalent in the policies and guidelines of our\ndataset, released as an open-source database and tool. We present the\nlimitations of performing a global scale analysis study paired with a critical\nanalysis of our findings, presenting areas of consensus that should be\nincorporated into future regulatory efforts. All components tied to this work\ncan be found in https://nkluge-correa.github.io/worldwide_AI-ethics/\n", "date": "2022-06-23", "categories": []}
{"id": "2402.12216", "title": "Copyleft for Alleviating AIGC Copyright Dilemma: What-if Analysis,\n  Public Perception and Implications", "abstract": "  As AIGC has impacted our society profoundly in the past years, ethical issues\nhave received tremendous attention. The most urgent one is the AIGC copyright\ndilemma, which can immensely stifle the development of AIGC and greatly cost\nthe entire society. Given the complexity of AIGC copyright governance and the\nfact that no perfect solution currently exists, previous work advocated\ncopyleft on AI governance but without substantive analysis. In this paper, we\ntake a step further to explore the feasibility of copyleft to alleviate the\nAIGC copyright dilemma. We conduct a mixed-methods study from two aspects:\nqualitatively, we use a formal what-if analysis to clarify the dilemma and\nprovide case studies to show the feasibility of copyleft; quantitatively, we\nperform a carefully designed survey to find out how the public feels about\ncopylefting AIGC. The key findings include: a) people generally perceive the\ndilemma, b) they prefer to use authorized AIGC under loose restriction, and c)\nthey are positive to copyleft in AIGC and willing to use it in the future.\n", "date": "2024-02-19", "categories": []}
{"id": "2402.07907", "title": "Applications, challenges and ethical issues of AI and ChatGPT in\n  education", "abstract": "  Artificial Intelligence (AI) in recent years has shown an unprecedentedly\nimpressive development, tending to play a catalytic role in all aspects of\nlife. The interest of the academic community, but also of governments, is huge\nin the dynamics of AI and is reflected by the truly explosive amount of\ninvestment and research that is underway. Enthusiastic opinions and statements\nabout AI are made every day, but at the same time they also bring to the fore\nalarming predictions about its effects. This paper aims to describe the\nopportunities emerging from the use of artificial intelligence and ChatGPT to\nimprove education, but also to identify the challenges and ethical issues that\narise.\n", "date": "2024-01-22", "categories": []}
{"id": "2402.08242", "title": "Towards Equitable Agile Research and Development of AI and Robotics", "abstract": "  Machine Learning (ML) and 'Artificial Intelligence' ('AI') methods tend to\nreplicate and amplify existing biases and prejudices, as do Robots with AI. For\nexample, robots with facial recognition have failed to identify Black Women as\nhuman, while others have categorized people, such as Black Men, as criminals\nbased on appearance alone. A 'culture of modularity' means harms are perceived\nas 'out of scope', or someone else's responsibility, throughout employment\npositions in the 'AI supply chain'. Incidents are routine enough\n(incidentdatabase.ai lists over 2000 examples) to indicate that few\norganizations are capable of completely respecting peoples' rights; meeting\nclaimed equity, diversity, and inclusion (EDI or DEI) goals; or recognizing and\nthen addressing such failures in their organizations and artifacts. We propose\na framework for adapting widely practiced Research and Development (R&D)\nproject management methodologies to build organizational equity capabilities\nand better integrate known evidence-based best practices. We describe how\nproject teams can organize and operationalize the most promising practices,\nskill sets, organizational cultures, and methods to detect and address\nrights-based fairness, equity, accountability, and ethical problems as early as\npossible when they are often less harmful and easier to mitigate; then monitor\nfor unforeseen incidents to adaptively and constructively address them. Our\nprimary example adapts an Agile development process based on Scrum, one of the\nmost widely adopted approaches to organizing R&D teams. We also discuss\nlimitations of our proposed framework and future research directions.\n", "date": "2024-02-13", "categories": []}
{"id": "2402.05786", "title": "Prompting Fairness: Artificial Intelligence as Game Players", "abstract": "  Utilitarian games such as dictator games to measure fairness have been\nstudied in the social sciences for decades. These games have given us insight\ninto not only how humans view fairness but also in what conditions the\nfrequency of fairness, altruism and greed increase or decrease. While these\ngames have traditionally been focused on humans, the rise of AI gives us the\nability to study how these models play these games. AI is becoming a constant\nin human interaction and examining how these models portray fairness in game\nplay can give us some insight into how AI makes decisions. Over 101 rounds of\nthe dictator game, I conclude that AI has a strong sense of fairness that is\ndependant of it it deems the person it is playing with as trustworthy, framing\nhas a strong effect on how much AI gives a recipient when designated the\ntrustee, and there may be evidence that AI experiences inequality aversion just\nas humans.\n", "date": "2024-02-08", "categories": []}
{"id": "2402.05979", "title": "On the Standardization of Behavioral Use Clauses and Their Adoption for\n  Responsible Licensing of AI", "abstract": "  Growing concerns over negligent or malicious uses of AI have increased the\nappetite for tools that help manage the risks of the technology. In 2018,\nlicenses with behaviorial-use clauses (commonly referred to as Responsible AI\nLicenses) were proposed to give developers a framework for releasing AI assets\nwhile specifying their users to mitigate negative applications. As of the end\nof 2023, on the order of 40,000 software and model repositories have adopted\nresponsible AI licenses licenses. Notable models licensed with behavioral use\nclauses include BLOOM (language) and LLaMA2 (language), Stable Diffusion\n(image), and GRID (robotics). This paper explores why and how these licenses\nhave been adopted, and why and how they have been adapted to fit particular use\ncases. We use a mixed-methods methodology of qualitative interviews, clustering\nof license clauses, and quantitative analysis of license adoption. Based on\nthis evidence we take the position that responsible AI licenses need\nstandardization to avoid confusing users or diluting their impact. At the same\ntime, customization of behavioral restrictions is also appropriate in some\ncontexts (e.g., medical domains). We advocate for ``standardized\ncustomization'' that can meet users' needs and can be supported via tooling.\n", "date": "2024-02-07", "categories": []}
{"id": "2309.05958", "title": "The Moral Machine Experiment on Large Language Models", "abstract": "  As large language models (LLMs) become more deeply integrated into various\nsectors, understanding how they make moral judgments has become crucial,\nparticularly in the realm of autonomous driving. This study utilized the Moral\nMachine framework to investigate the ethical decision-making tendencies of\nprominent LLMs, including GPT-3.5, GPT-4, PaLM 2, and Llama 2, comparing their\nresponses to human preferences. While LLMs' and humans' preferences such as\nprioritizing humans over pets and favoring saving more lives are broadly\naligned, PaLM 2 and Llama 2, especially, evidence distinct deviations.\nAdditionally, despite the qualitative similarities between the LLM and human\npreferences, there are significant quantitative disparities, suggesting that\nLLMs might lean toward more uncompromising decisions, compared to the milder\ninclinations of humans. These insights elucidate the ethical frameworks of LLMs\nand their potential implications for autonomous driving.\n", "date": "2023-09-12", "categories": []}
{"id": "2402.04373", "title": "The World of Generative AI: Deepfakes and Large Language Models", "abstract": "  We live in the era of Generative Artificial Intelligence (GenAI). Deepfakes\nand Large Language Models (LLMs) are two examples of GenAI. Deepfakes, in\nparticular, pose an alarming threat to society as they are capable of spreading\nmisinformation and changing the truth. LLMs are powerful language models that\ngenerate general-purpose language. However due to its generative aspect, it can\nalso be a risk for people if used with ill intentions. The ethical use of these\ntechnologies is a big concern. This short article tries to find out the\ninterrelationship between them.\n", "date": "2024-02-06", "categories": []}
{"id": "2402.01651", "title": "Informed AI Regulation: Comparing the Ethical Frameworks of Leading LLM\n  Chatbots Using an Ethics-Based Audit to Assess Moral Reasoning and Normative\n  Values", "abstract": "  With the rise of individual and collaborative networks of autonomous agents,\nAI is deployed in more key reasoning and decision-making roles. For this\nreason, ethics-based audits play a pivotal role in the rapidly growing fields\nof AI safety and regulation. This paper undertakes an ethics-based audit to\nprobe the 8 leading commercial and open-source Large Language Models including\nGPT-4. We assess explicability and trustworthiness by a) establishing how well\ndifferent models engage in moral reasoning and b) comparing normative values\nunderlying models as ethical frameworks. We employ an experimental,\nevidence-based approach that challenges the models with ethical dilemmas in\norder to probe human-AI alignment. The ethical scenarios are designed to\nrequire a decision in which the particulars of the situation may or may not\nnecessitate deviating from normative ethical principles. A sophisticated\nethical framework was consistently elicited in one model, GPT-4. Nonetheless,\ntroubling findings include underlying normative frameworks with clear bias\ntowards particular cultural norms. Many models also exhibit disturbing\nauthoritarian tendencies. Code is available at\nhttps://github.com/jonchun/llm-sota-chatbots-ethics-based-audit.\n", "date": "2024-01-09", "categories": []}
{"id": "2402.01691", "title": "Investigating Algorithm Review Boards for Organizational Responsible\n  Artificial Intelligence Governance", "abstract": "  Organizations including companies, nonprofits, governments, and academic\ninstitutions are increasingly developing, deploying, and utilizing artificial\nintelligence (AI) tools. Responsible AI (RAI) governance approaches at\norganizations have emerged as important mechanisms to address potential AI\nrisks and harms. In this work, we interviewed 17 technical contributors across\norganization types (Academic, Government, Industry, Nonprofit) and sectors\n(Finance, Health, Tech, Other) about their experiences with internal RAI\ngovernance. Our findings illuminated the variety of organizational definitions\nof RAI and accompanying internal governance approaches. We summarized the first\ndetailed findings on algorithm review boards (ARBs) and similar review\ncommittees in practice, including their membership, scope, and measures of\nsuccess. We confirmed known robust model governance in finance sectors and\nrevealed extensive algorithm and AI governance with ARB-like review boards in\nhealth sectors. Our findings contradict the idea that Institutional Review\nBoards alone are sufficient for algorithm governance and posit that ARBs are\namong the more impactful internal RAI governance approaches. Our results\nsuggest that integration with existing internal regulatory approaches and\nleadership buy-in are among the most important attributes for success and that\nfinancial tensions are the greatest challenge to effective organizational RAI.\nWe make a variety of suggestions for how organizational partners can learn from\nthese findings when building their own internal RAI frameworks. We outline\nfuture directions for developing and measuring effectiveness of ARBs and other\ninternal RAI governance approaches.\n", "date": "2024-01-23", "categories": []}
{"id": "2402.01762", "title": "Commercial AI, Conflict, and Moral Responsibility: A theoretical\n  analysis and practical approach to the moral responsibilities associated with\n  dual-use AI technology", "abstract": "  This paper presents a theoretical analysis and practical approach to the\nmoral responsibilities when developing AI systems for non-military applications\nthat may nonetheless be used for conflict applications. We argue that AI\nrepresents a form of crossover technology that is different from previous\nhistorical examples of dual- or multi-use technology as it has a multiplicative\neffect across other technologies. As a result, existing analyses of ethical\nresponsibilities around dual-use technologies do not necessarily work for AI\nsystems. We instead argue that stakeholders involved in the AI system lifecycle\nare morally responsible for uses of their systems that are reasonably\nforeseeable. The core idea is that an agent's moral responsibility for some\naction is not necessarily determined by their intentions alone; we must also\nconsider what the agent could reasonably have foreseen to be potential outcomes\nof their action, such as the potential use of a system in conflict even when it\nis not designed for that. In particular, we contend that it is reasonably\nforeseeable that: (1) civilian AI systems will be applied to active conflict,\nincluding conflict support activities, (2) the use of civilian AI systems in\nconflict will impact applications of the law of armed conflict, and (3)\ncrossover AI technology will be applied to conflicts that fall short of armed\nconflict. Given these reasonably foreseeably outcomes, we present three\ntechnically feasible actions that developers of civilian AIs can take to\npotentially mitigate their moral responsibility: (a) establishing systematic\napproaches to multi-perspective capability testing, (b) integrating digital\nwatermarking in model weight matrices, and (c) utilizing monitoring and\nreporting mechanisms for conflict-related AI applications.\n", "date": "2024-01-30", "categories": []}
{"id": "2304.12241", "title": "Positive AI: Key Challenges in Designing Artificial Intelligence for\n  Wellbeing", "abstract": "  Artificial Intelligence (AI) is a double-edged sword: on one hand, AI\npromises to provide great advances that could benefit humanity, but on the\nother hand, AI poses substantial (even existential) risks. With advancements\nhappening daily, many people are increasingly worried about AI's impact on\ntheir lives. To ensure AI progresses beneficially, some researchers have\nproposed \"wellbeing\" as a key objective to govern AI. This article addresses\nkey challenges in designing AI for wellbeing. We group these challenges into\nissues of modeling wellbeing in context, assessing wellbeing in context,\ndesigning interventions to improve wellbeing, and maintaining AI alignment with\nwellbeing over time. The identification of these challenges provides a scope\nfor efforts to help ensure that AI developments are aligned with human\nwellbeing.\n", "date": "2023-04-12", "categories": []}
{"id": "2402.00029", "title": "Exploring Public Opinion on Responsible AI Through The Lens of Cultural\n  Consensus Theory", "abstract": "  As the societal implications of Artificial Intelligence (AI) continue to\ngrow, the pursuit of responsible AI necessitates public engagement in its\ndevelopment and governance processes. This involvement is crucial for capturing\ndiverse perspectives and promoting equitable practices and outcomes. We applied\nCultural Consensus Theory (CCT) to a nationally representative survey dataset\non various aspects of AI to discern beliefs and attitudes about responsible AI\nin the United States. Our results offer valuable insights by identifying shared\nand contrasting views on responsible AI. Furthermore, these findings serve as\ncritical reference points for developers and policymakers, enabling them to\nmore effectively consider individual variances and group-level cultural\nperspectives when making significant decisions and addressing the public's\nconcerns.\n", "date": "2024-01-06", "categories": []}
{"id": "2311.14720", "title": "Perceptions and Detection of AI Use in Manuscript Preparation for\n  Academic Journals", "abstract": "  The emergent abilities of Large Language Models (LLMs), which power tools\nlike ChatGPT and Bard, have produced both excitement and worry about how AI\nwill impact academic writing. In response to rising concerns about AI use,\nauthors of academic publications may decide to voluntarily disclose any AI\ntools they use to revise their manuscripts, and journals and conferences could\nbegin mandating disclosure and/or turn to using detection services, as many\nteachers have done with student writing in class settings. Given these looming\npossibilities, we investigate whether academics view it as necessary to report\nAI use in manuscript preparation and how detectors react to the use of AI in\nacademic writing.\n", "date": "2023-11-19", "categories": []}
{"id": "2401.16088", "title": "Fairness in Algorithmic Recourse Through the Lens of Substantive\n  Equality of Opportunity", "abstract": "  Algorithmic recourse -- providing recommendations to those affected\nnegatively by the outcome of an algorithmic system on how they can take action\nand change that outcome -- has gained attention as a means of giving persons\nagency in their interactions with artificial intelligence (AI) systems. Recent\nwork has shown that even if an AI decision-making classifier is ``fair''\n(according to some reasonable criteria), recourse itself may be unfair due to\ndifferences in the initial circumstances of individuals, compounding\ndisparities for marginalized populations and requiring them to exert more\neffort than others. There is a need to define more methods and metrics for\nevaluating fairness in recourse that span a range of normative views of the\nworld, and specifically those that take into account time. Time is a critical\nelement in recourse because the longer it takes an individual to act, the more\nthe setting may change due to model or data drift.\n  This paper seeks to close this research gap by proposing two notions of\nfairness in recourse that are in normative alignment with substantive equality\nof opportunity, and that consider time. The first considers the (often\nrepeated) effort individuals exert per successful recourse event, and the\nsecond considers time per successful recourse event. Building upon an\nagent-based framework for simulating recourse, this paper demonstrates how much\neffort is needed to overcome disparities in initial circumstances. We then\nproposes an intervention to improve the fairness of recourse by rewarding\neffort, and compare it to existing strategies.\n", "date": "2024-01-29", "categories": []}
{"id": "2401.14523", "title": "Empathy and the Right to Be an Exception: What LLMs Can and Cannot Do", "abstract": "  Advances in the performance of large language models (LLMs) have led some\nresearchers to propose the emergence of theory of mind (ToM) in artificial\nintelligence (AI). LLMs can attribute beliefs, desires, intentions, and\nemotions, and they will improve in their accuracy. Rather than employing the\ncharacteristically human method of empathy, they learn to attribute mental\nstates by recognizing linguistic patterns in a dataset that typically do not\ninclude that individual. We ask whether LLMs' inability to empathize precludes\nthem from honoring an individual's right to be an exception, that is, from\nmaking assessments of character and predictions of behavior that reflect\nappropriate sensitivity to a person's individuality. Can LLMs seriously\nconsider an individual's claim that their case is different based on internal\nmental states like beliefs, desires, and intentions, or are they limited to\njudging that case based on its similarities to others? We propose that the\nmethod of empathy has special significance for honoring the right to be an\nexception that is distinct from the value of predictive accuracy, at which LLMs\nexcel. We conclude by considering whether using empathy to consider exceptional\ncases has intrinsic or merely practical value and we introduce conceptual and\nempirical avenues for advancing this investigation.\n", "date": "2024-01-25", "categories": []}
{"id": "2311.01193", "title": "Contextual Confidence and Generative AI", "abstract": "  Generative AI models perturb the foundations of effective human\ncommunication. They present new challenges to contextual confidence, disrupting\nparticipants' ability to identify the authentic context of communication and\ntheir ability to protect communication from reuse and recombination outside its\nintended context. In this paper, we describe strategies--tools, technologies\nand policies--that aim to stabilize communication in the face of these\nchallenges. The strategies we discuss fall into two broad categories.\nContainment strategies aim to reassert context in environments where it is\ncurrently threatened--a reaction to the context-free expectations and norms\nestablished by the internet. Mobilization strategies, by contrast, view the\nrise of generative AI as an opportunity to proactively set new and higher\nexpectations around privacy and authenticity in mediated communication.\n", "date": "2023-11-02", "categories": []}
{"id": "2401.10535", "title": "The \"Colonial Impulse\" of Natural Language Processing: An Audit of\n  Bengali Sentiment Analysis Tools and Their Identity-based Biases", "abstract": "  While colonization has sociohistorically impacted people's identities across\nvarious dimensions, those colonial values and biases continue to be perpetuated\nby sociotechnical systems. One category of sociotechnical systems--sentiment\nanalysis tools--can also perpetuate colonial values and bias, yet less\nattention has been paid to how such tools may be complicit in perpetuating\ncoloniality, although they are often used to guide various practices (e.g.,\ncontent moderation). In this paper, we explore potential bias in sentiment\nanalysis tools in the context of Bengali communities that have experienced and\ncontinue to experience the impacts of colonialism. Drawing on identity\ncategories most impacted by colonialism amongst local Bengali communities, we\nfocused our analytic attention on gender, religion, and nationality. We\nconducted an algorithmic audit of all sentiment analysis tools for Bengali,\navailable on the Python package index (PyPI) and GitHub. Despite similar\nsemantic content and structure, our analyses showed that in addition to\ninconsistencies in output from different tools, Bengali sentiment analysis\ntools exhibit bias between different identity categories and respond\ndifferently to different ways of identity expression. Connecting our findings\nwith colonially shaped sociocultural structures of Bengali communities, we\ndiscuss the implications of downstream bias of sentiment analysis tools.\n", "date": "2024-01-19", "categories": []}
{"id": "2401.09473", "title": "Business and ethical concerns in domestic Conversational Generative\n  AI-empowered multi-robot systems", "abstract": "  Business and technology are intricately connected through logic and design.\nThey are equally sensitive to societal changes and may be devastated by\nscandal. Cooperative multi-robot systems (MRSs) are on the rise, allowing\nrobots of different types and brands to work together in diverse contexts.\nGenerative artificial intelligence has been a dominant topic in recent\nartificial intelligence (AI) discussions due to its capacity to mimic humans\nthrough the use of natural language and the production of media, including deep\nfakes. In this article, we focus specifically on the conversational aspects of\ngenerative AI, and hence use the term Conversational Generative artificial\nintelligence (CGI). Like MRSs, CGIs have enormous potential for revolutionizing\nprocesses across sectors and transforming the way humans conduct business. From\na business perspective, cooperative MRSs alone, with potential conflicts of\ninterest, privacy practices, and safety concerns, require ethical examination.\nMRSs empowered by CGIs demand multi-dimensional and sophisticated methods to\nuncover imminent ethical pitfalls. This study focuses on ethics in\nCGI-empowered MRSs while reporting the stages of developing the MORUL model.\n", "date": "2024-01-12", "categories": []}
{"id": "2401.08691", "title": "Towards Responsible AI in Banking: Addressing Bias for Fair\n  Decision-Making", "abstract": "  In an era characterized by the pervasive integration of artificial\nintelligence into decision-making processes across diverse industries, the\ndemand for trust has never been more pronounced. This thesis embarks on a\ncomprehensive exploration of bias and fairness, with a particular emphasis on\ntheir ramifications within the banking sector, where AI-driven decisions bear\nsubstantial societal consequences. In this context, the seamless integration of\nfairness, explainability, and human oversight is of utmost importance,\nculminating in the establishment of what is commonly referred to as\n\"Responsible AI\". This emphasizes the critical nature of addressing biases\nwithin the development of a corporate culture that aligns seamlessly with both\nAI regulations and universal human rights standards, particularly in the realm\nof automated decision-making systems. Nowadays, embedding ethical principles\ninto the development, training, and deployment of AI models is crucial for\ncompliance with forthcoming European regulations and for promoting societal\ngood. This thesis is structured around three fundamental pillars: understanding\nbias, mitigating bias, and accounting for bias. These contributions are\nvalidated through their practical application in real-world scenarios, in\ncollaboration with Intesa Sanpaolo. This collaborative effort not only\ncontributes to our understanding of fairness but also provides practical tools\nfor the responsible implementation of AI-based decision-making systems. In line\nwith open-source principles, we have released Bias On Demand and FairView as\naccessible Python packages, further promoting progress in the field of AI\nfairness.\n", "date": "2024-01-13", "categories": []}
{"id": "2306.01774", "title": "RE-centric Recommendations for the Development of Trustworthy(er)\n  Autonomous Systems", "abstract": "  Complying with the EU AI Act (AIA) guidelines while developing and\nimplementing AI systems will soon be mandatory within the EU. However,\npractitioners lack actionable instructions to operationalise ethics during AI\nsystems development. A literature review of different ethical guidelines\nrevealed inconsistencies in the principles addressed and the terminology used\nto describe them. Furthermore, requirements engineering (RE), which is\nidentified to foster trustworthiness in the AI development process from the\nearly stages was observed to be absent in a lot of frameworks that support the\ndevelopment of ethical and trustworthy AI. This incongruous phrasing combined\nwith a lack of concrete development practices makes trustworthy AI development\nharder. To address this concern, we formulated a comparison table for the\nterminology used and the coverage of the ethical AI principles in major ethical\nAI guidelines. We then examined the applicability of ethical AI development\nframeworks for performing effective RE during the development of trustworthy AI\nsystems. A tertiary review and meta-analysis of literature discussing ethical\nAI frameworks revealed their limitations when developing trustworthy AI. Based\non our findings, we propose recommendations to address such limitations during\nthe development of trustworthy AI.\n", "date": "2023-05-29", "categories": []}
{"id": "2312.16191", "title": "SoK: Taming the Triangle -- On the Interplays between Fairness,\n  Interpretability and Privacy in Machine Learning", "abstract": "  Machine learning techniques are increasingly used for high-stakes\ndecision-making, such as college admissions, loan attribution or recidivism\nprediction. Thus, it is crucial to ensure that the models learnt can be audited\nor understood by human users, do not create or reproduce discrimination or\nbias, and do not leak sensitive information regarding their training data.\nIndeed, interpretability, fairness and privacy are key requirements for the\ndevelopment of responsible machine learning, and all three have been studied\nextensively during the last decade. However, they were mainly considered in\nisolation, while in practice they interplay with each other, either positively\nor negatively. In this Systematization of Knowledge (SoK) paper, we survey the\nliterature on the interactions between these three desiderata. More precisely,\nfor each pairwise interaction, we summarize the identified synergies and\ntensions. These findings highlight several fundamental theoretical and\nempirical conflicts, while also demonstrating that jointly considering these\ndifferent requirements is challenging when one aims at preserving a high level\nof utility. To solve this issue, we also discuss possible conciliation\nmechanisms, showing that a careful design can enable to successfully handle\nthese different concerns in practice.\n", "date": "2023-12-22", "categories": []}
{"id": "2312.16726", "title": "FairCompass: Operationalising Fairness in Machine Learning", "abstract": "  As artificial intelligence (AI) increasingly becomes an integral part of our\nsocietal and individual activities, there is a growing imperative to develop\nresponsible AI solutions. Despite a diverse assortment of machine learning\nfairness solutions is proposed in the literature, there is reportedly a lack of\npractical implementation of these tools in real-world applications. Industry\nexperts have participated in thorough discussions on the challenges associated\nwith operationalising fairness in the development of machine learning-empowered\nsolutions, in which a shift toward human-centred approaches is promptly\nadvocated to mitigate the limitations of existing techniques. In this work, we\npropose a human-in-the-loop approach for fairness auditing, presenting a mixed\nvisual analytical system (hereafter referred to as 'FairCompass'), which\nintegrates both subgroup discovery technique and the decision tree-based schema\nfor end users. Moreover, we innovatively integrate an Exploration, Guidance and\nInformed Analysis loop, to facilitate the use of the Knowledge Generation Model\nfor Visual Analytics in FairCompass. We evaluate the effectiveness of\nFairCompass for fairness auditing in a real-world scenario, and the findings\ndemonstrate the system's potential for real-world deployability. We anticipate\nthis work will address the current gaps in research for fairness and facilitate\nthe operationalisation of fairness in machine learning systems.\n", "date": "2023-12-27", "categories": []}
{"id": "2312.10075", "title": "Assessing LLMs for Moral Value Pluralism", "abstract": "  The fields of AI current lacks methods to quantitatively assess and\npotentially alter the moral values inherent in the output of large language\nmodels (LLMs). However, decades of social science research has developed and\nrefined widely-accepted moral value surveys, such as the World Values Survey\n(WVS), eliciting value judgments from direct questions in various geographies.\nWe have turned those questions into value statements and use NLP to compute to\nhow well popular LLMs are aligned with moral values for various demographics\nand cultures. While the WVS is accepted as an explicit assessment of values, we\nlack methods for assessing implicit moral and cultural values in media, e.g.,\nencountered in social media, political rhetoric, narratives, and generated by\nAI systems such as LLMs that are increasingly present in our daily lives. As we\nconsume online content and utilize LLM outputs, we might ask, which moral\nvalues are being implicitly promoted or undercut, or -- in the case of LLMs --\nif they are intending to represent a cultural identity, are they doing so\nconsistently? In this paper we utilize a Recognizing Value Resonance (RVR) NLP\nmodel to identify WVS values that resonate and conflict with a given passage of\noutput text. We apply RVR to the text generated by LLMs to characterize\nimplicit moral values, allowing us to quantify the moral/cultural distance\nbetween LLMs and various demographics that have been surveyed using the WVS. In\nline with other work we find that LLMs exhibit several Western-centric value\nbiases; they overestimate how conservative people in non-Western countries are,\nthey are less accurate in representing gender for non-Western countries, and\nportray older populations as having more traditional values. Our results\nhighlight value misalignment and age groups, and a need for social science\ninformed technological solutions addressing value plurality in LLMs.\n", "date": "2023-12-08", "categories": []}
{"id": "2312.09561", "title": "Investigating Responsible AI for Scientific Research: An Empirical Study", "abstract": "  Scientific research organizations that are developing and deploying\nArtificial Intelligence (AI) systems are at the intersection of technological\nprogress and ethical considerations. The push for Responsible AI (RAI) in such\ninstitutions underscores the increasing emphasis on integrating ethical\nconsiderations within AI design and development, championing core values like\nfairness, accountability, and transparency. For scientific research\norganizations, prioritizing these practices is paramount not just for\nmitigating biases and ensuring inclusivity, but also for fostering trust in AI\nsystems among both users and broader stakeholders. In this paper, we explore\nthe practices at a research organization concerning RAI practices, aiming to\nassess the awareness and preparedness regarding the ethical risks inherent in\nAI design and development. We have adopted a mixed-method research approach,\nutilising a comprehensive survey combined with follow-up in-depth interviews\nwith selected participants from AI-related projects. Our results have revealed\ncertain knowledge gaps concerning ethical, responsible, and inclusive AI, with\nlimitations in awareness of the available AI ethics frameworks. This revealed\nan overarching underestimation of the ethical risks that AI technologies can\npresent, especially when implemented without proper guidelines and governance.\nOur findings reveal the need for a holistic and multi-tiered strategy to uplift\ncapabilities and better support science research teams for responsible,\nethical, and inclusive AI development and deployment.\n", "date": "2023-12-15", "categories": []}
{"id": "2312.06074", "title": "A Vision for Operationalising Diversity and Inclusion in AI", "abstract": "  The growing presence of Artificial Intelligence (AI) in various sectors\nnecessitates systems that accurately reflect societal diversity. This study\nseeks to envision the operationalization of the ethical imperatives of\ndiversity and inclusion (D&I) within AI ecosystems, addressing the current\ndisconnect between ethical guidelines and their practical implementation. A\nsignificant challenge in AI development is the effective operationalization of\nD&I principles, which is critical to prevent the reinforcement of existing\nbiases and ensure equity across AI applications. This paper proposes a vision\nof a framework for developing a tool utilizing persona-based simulation by\nGenerative AI (GenAI). The approach aims to facilitate the representation of\nthe needs of diverse users in the requirements analysis process for AI\nsoftware. The proposed framework is expected to lead to a comprehensive persona\nrepository with diverse attributes that inform the development process with\ndetailed user narratives. This research contributes to the development of an\ninclusive AI paradigm that ensures future technological advances are designed\nwith a commitment to the diverse fabric of humanity.\n", "date": "2023-12-11", "categories": []}
{"id": "2312.00044", "title": "Advancing AI Audits for Enhanced AI Governance", "abstract": "  As artificial intelligence (AI) is integrated into various services and\nsystems in society, many companies and organizations have proposed AI\nprinciples, policies, and made the related commitments. Conversely, some have\nproposed the need for independent audits, arguing that the voluntary principles\nadopted by the developers and providers of AI services and systems\ninsufficiently address risk. This policy recommendation summarizes the issues\nrelated to the auditing of AI services and systems and presents three\nrecommendations for promoting AI auditing that contribute to sound AI\ngovernance. Recommendation1.Development of institutional design for AI audits.\nRecommendation2.Training human resources for AI audits. Recommendation3.\nUpdating AI audits in accordance with technological progress.\n  In this policy recommendation, AI is assumed to be that which recognizes and\npredicts data with the last chapter outlining how generative AI should be\naudited.\n", "date": "2023-11-26", "categories": []}
{"id": "2312.00584", "title": "The Ethics of Automating Legal Actors", "abstract": "  The introduction of large public legal datasets has brought about a\nrenaissance in legal NLP. Many of these datasets are comprised of legal\njudgements - the product of judges deciding cases. This fact, together with the\nway machine learning works, means that several legal NLP models are models of\njudges. While some have argued for the automation of judges, in this position\npiece, we argue that automating the role of the judge raises difficult ethical\nchallenges, in particular for common law legal systems. Our argument follows\nfrom the social role of the judge in actively shaping the law, rather than\nmerely applying it. Since current NLP models come nowhere close to having the\nfacilities necessary for this task, they should not be used to automate judges.\nFurthermore, even in the case the models could achieve human-level\ncapabilities, there would still be remaining ethical concerns inherent in the\nautomation of the legal process.\n", "date": "2023-12-01", "categories": []}
{"id": "2311.14687", "title": "Does Explainable AI Have Moral Value?", "abstract": "  Explainable AI (XAI) aims to bridge the gap between complex algorithmic\nsystems and human stakeholders. Current discourse often examines XAI in\nisolation as either a technological tool, user interface, or policy mechanism.\nThis paper proposes a unifying ethical framework grounded in moral duties and\nthe concept of reciprocity. We argue that XAI should be appreciated not merely\nas a right, but as part of our moral duties that helps sustain a reciprocal\nrelationship between humans affected by AI systems. This is because, we argue,\nexplanations help sustain constitutive symmetry and agency in AI-led\ndecision-making processes. We then assess leading XAI communities and reveal\ngaps between the ideal of reciprocity and practical feasibility. Machine\nlearning offers useful techniques but overlooks evaluation and adoption\nchallenges. Human-computer interaction provides preliminary insights but\noversimplifies organizational contexts. Policies espouse accountability but\nlack technical nuance. Synthesizing these views exposes barriers to\nimplementable, ethical XAI. Still, positioning XAI as a moral duty transcends\nrights-based discourse to capture a more robust and complete moral picture.\nThis paper provides an accessible, detailed analysis elucidating the moral\nvalue of explainability.\n", "date": "2023-11-05", "categories": []}
{"id": "2311.14700", "title": "Towards a Feminist Metaethics of AI", "abstract": "  The proliferation of Artificial Intelligence (AI) has sparked an overwhelming\nnumber of AI ethics guidelines, boards and codes of conduct. These outputs\nprimarily analyse competing theories, principles and values for AI development\nand deployment. However, as a series of recent problematic incidents about AI\nethics/ethicists demonstrate, this orientation is insufficient. Before\nproceeding to evaluate other professions, AI ethicists should critically\nevaluate their own; yet, such an evaluation should be more explicitly and\nsystematically undertaken in the literature. I argue that these insufficiencies\ncould be mitigated by developing a research agenda for a feminist metaethics of\nAI. Contrary to traditional metaethics, which reflects on the nature of\nmorality and moral judgements in a non-normative way, feminist metaethics\nexpands its scope to ask not only what ethics is but also what our engagement\nwith it should be like. Applying this perspective to the context of AI, I\nsuggest that a feminist metaethics of AI would examine: (i) the continuity\nbetween theory and action in AI ethics; (ii) the real-life effects of AI\nethics; (iii) the role and profile of those involved in AI ethics; and (iv) the\neffects of AI on power relations through methods that pay attention to context,\nemotions and narrative.\n", "date": "2023-11-10", "categories": []}
{"id": "2311.14711", "title": "Towards Publicly Accountable Frontier LLMs: Building an External\n  Scrutiny Ecosystem under the ASPIRE Framework", "abstract": "  With the increasing integration of frontier large language models (LLMs) into\nsociety and the economy, decisions related to their training, deployment, and\nuse have far-reaching implications. These decisions should not be left solely\nin the hands of frontier LLM developers. LLM users, civil society and\npolicymakers need trustworthy sources of information to steer such decisions\nfor the better. Involving outside actors in the evaluation of these systems -\nwhat we term 'external scrutiny' - via red-teaming, auditing, and external\nresearcher access, offers a solution. Though there are encouraging signs of\nincreasing external scrutiny of frontier LLMs, its success is not assured. In\nthis paper, we survey six requirements for effective external scrutiny of\nfrontier AI systems and organize them under the ASPIRE framework: Access,\nSearching attitude, Proportionality to the risks, Independence, Resources, and\nExpertise. We then illustrate how external scrutiny might function throughout\nthe AI lifecycle and offer recommendations to policymakers.\n", "date": "2023-11-15", "categories": []}
{"id": "2311.13262", "title": "The Rise of Creative Machines: Exploring the Impact of Generative AI", "abstract": "  This study looks at how generative artificial intelligence (AI) can\nrevolutionize marketing, product development, and research. It discusses the\nlatest developments in the field, easy-to-use resources, and moral and social\nhazards. In addition to addressing mitigating techniques for issues like\nprejudice and disinformation, the debate emphasizes the significance of\nresponsible development through continual stakeholder communication and ethical\nprinciples.\n", "date": "2023-11-22", "categories": []}
{"id": "2311.13286", "title": "Algorithmic Transparency and Manipulation", "abstract": "  A series of recent papers raises worries about the manipulative potential of\nalgorithmic transparency. But while the concern is apt and relevant, it is\nbased on a fraught understanding of manipulation. Therefore, this paper draws\nattention to the indifference view of manipulation, which explains better than\nthe vulnerability view why algorithmic transparency has manipulative potential.\nThe paper also raises pertinent research questions for future studies of\nmanipulation in the context of algorithmic transparency.\n", "date": "2023-11-22", "categories": []}
{"id": "2311.10725", "title": "Should they? Mobile Biometrics and Technopolicy meet Queer Community\n  Considerations", "abstract": "  Smartphones are integral to our daily lives and activities, providing us with\nbasic functions like texting and phone calls to more complex motion-based\nfunctionalities like navigation, mobile gaming, and fitness-tracking. To\nfacilitate these functionalities, smartphones rely on integrated sensors like\naccelerometers and gyroscopes. These sensors provide personalized measurements\nthat, in turn, contribute to tasks such as analyzing biometric data for mobile\nhealth purposes. In addition to benefiting smartphone users, biometric data\nholds significant value for researchers engaged in biometric identification\nresearch. Nonetheless, utilizing this user data for biometric identification\ntasks, such as gait and gender recognition, raises serious privacy, normative,\nand ethical concerns, particularly within the queer community. Concerns of\nalgorithmic bias and algorithmically-driven dysphoria surface from a historical\nbackdrop of marginalization, surveillance, harassment, discrimination, and\nviolence against the queer community. In this position paper, we contribute to\nthe timely discourse on safeguarding human rights within AI-driven systems by\nproviding a sense of challenges, tensions, and opportunities for new data\nprotections and biometric collection practices in a way that grapples with the\nsociotechnical realities of the queer community.\n", "date": "2023-10-06", "categories": []}
{"id": "2311.11103", "title": "Responsible AI Considerations in Text Summarization Research: A Review\n  of Current Practices", "abstract": "  AI and NLP publication venues have increasingly encouraged researchers to\nreflect on possible ethical considerations, adverse impacts, and other\nresponsible AI issues their work might engender. However, for specific NLP\ntasks our understanding of how prevalent such issues are, or when and why these\nissues are likely to arise, remains limited. Focusing on text summarization --\na common NLP task largely overlooked by the responsible AI community -- we\nexamine research and reporting practices in the current literature. We conduct\na multi-round qualitative analysis of 333 summarization papers from the ACL\nAnthology published between 2020-2022. We focus on how, which, and when\nresponsible AI issues are covered, which relevant stakeholders are considered,\nand mismatches between stated and realized research goals. We also discuss\ncurrent evaluation practices and consider how authors discuss the limitations\nof both prior work and their own work. Overall, we find that relatively few\npapers engage with possible stakeholders or contexts of use, which limits their\nconsideration of potential downstream adverse impacts or other responsible AI\nissues. Based on our findings, we make recommendations on concrete practices\nand research directions.\n", "date": "2023-11-18", "categories": []}
{"id": "2311.11193", "title": "Assessing AI Impact Assessments: A Classroom Study", "abstract": "  Artificial Intelligence Impact Assessments (\"AIIAs\"), a family of tools that\nprovide structured processes to imagine the possible impacts of a proposed AI\nsystem, have become an increasingly popular proposal to govern AI systems.\nRecent efforts from government or private-sector organizations have proposed\nmany diverse instantiations of AIIAs, which take a variety of forms ranging\nfrom open-ended questionnaires to graded score-cards. However, to date that has\nbeen limited evaluation of existing AIIA instruments. We conduct a classroom\nstudy (N = 38) at a large research-intensive university (R1) in an elective\ncourse focused on the societal and ethical implications of AI. We assign\nstudents to different organizational roles (for example, an ML scientist or\nproduct manager) and ask participant teams to complete one of three existing AI\nimpact assessments for one of two imagined generative AI systems. In our\nthematic analysis of participants' responses to pre- and post-activity\nquestionnaires, we find preliminary evidence that impact assessments can\ninfluence participants' perceptions of the potential risks of generative AI\nsystems, and the level of responsibility held by AI experts in addressing\npotential harm. We also discover a consistent set of limitations shared by\nseveral existing AIIA instruments, which we group into concerns about their\nformat and content, as well as the feasibility and effectiveness of the\nactivity in foreseeing and mitigating potential harms. Drawing on the findings\nof this study, we provide recommendations for future work on developing and\nvalidating AIIAs.\n", "date": "2023-11-19", "categories": []}
{"id": "2311.12010", "title": "Steering Responsible AI: A Case for Algorithmic Pluralism", "abstract": "  In this paper, I examine questions surrounding AI neutrality through the\nprism of existing literature and scholarship about mediation and media\npluralism. Such traditions, I argue, provide a valuable theoretical framework\nfor how we should approach the (likely) impending era of AI mediation. In\nparticular, I suggest examining further the notion of algorithmic pluralism.\nContrasting this notion to the dominant idea of algorithmic transparency, I\nseek to describe what algorithmic pluralism may be, and present both its\nopportunities and challenges. Implemented thoughtfully and responsibly, I\nargue, Algorithmic or AI pluralism has the potential to sustain the diversity,\nmultiplicity, and inclusiveness that are so vital to democracy.\n", "date": "2023-11-20", "categories": []}
{"id": "2311.09227", "title": "Open-Sourcing Highly Capable Foundation Models: An evaluation of risks,\n  benefits, and alternative methods for pursuing open-source objectives", "abstract": "  Recent decisions by leading AI labs to either open-source their models or to\nrestrict access to their models has sparked debate about whether, and how,\nincreasingly capable AI models should be shared. Open-sourcing in AI typically\nrefers to making model architecture and weights freely and publicly accessible\nfor anyone to modify, study, build on, and use. This offers advantages such as\nenabling external oversight, accelerating progress, and decentralizing control\nover AI development and use. However, it also presents a growing potential for\nmisuse and unintended consequences. This paper offers an examination of the\nrisks and benefits of open-sourcing highly capable foundation models. While\nopen-sourcing has historically provided substantial net benefits for most\nsoftware and AI development processes, we argue that for some highly capable\nfoundation models likely to be developed in the near future, open-sourcing may\npose sufficiently extreme risks to outweigh the benefits. In such a case,\nhighly capable foundation models should not be open-sourced, at least not\ninitially. Alternative strategies, including non-open-source model sharing\noptions, are explored. The paper concludes with recommendations for developers,\nstandard-setting bodies, and governments for establishing safe and responsible\nmodel sharing practices and preserving open-source benefits where safe.\n", "date": "2023-09-29", "categories": []}
{"id": "2311.09230", "title": "Evaluating and Improving Value Judgments in AI: A Scenario-Based Study\n  on Large Language Models' Depiction of Social Conventions", "abstract": "  The adoption of generative AI technologies is swiftly expanding. Services\nemploying both linguistic and mul-timodal models are evolving, offering users\nincreasingly precise responses. Consequently, human reliance on these\ntechnologies is expected to grow rapidly. With the premise that people will be\nimpacted by the output of AI, we explored approaches to help AI output produce\nbetter results. Initially, we evaluated how contemporary AI services\ncompetitively meet user needs, then examined society's depiction as mirrored by\nLarge Language Models (LLMs). We did a query experiment, querying about social\nconventions in various countries and eliciting a one-word response. We compared\nthe LLMs' value judgments with public data and suggested an model of\ndecision-making in value-conflicting scenarios which could be adopted for\nfuture machine value judgments. This paper advocates for a practical approach\nto using AI as a tool for investigating other remote worlds. This re-search has\nsignificance in implicitly rejecting the notion of AI making value judgments\nand instead arguing a more critical perspective on the environment that defers\njudgmental capabilities to individuals. We anticipate this study will empower\nanyone, regardless of their capacity, to receive safe and accurate value\njudgment-based out-puts effectively.\n", "date": "2023-10-04", "categories": []}
{"id": "2311.08576", "title": "Towards Evaluating AI Systems for Moral Status Using Self-Reports", "abstract": "  As AI systems become more advanced and widely deployed, there will likely be\nincreasing debate over whether AI systems could have conscious experiences,\ndesires, or other states of potential moral significance. It is important to\ninform these discussions with empirical evidence to the extent possible. We\nargue that under the right circumstances, self-reports, or an AI system's\nstatements about its own internal states, could provide an avenue for\ninvestigating whether AI systems have states of moral significance.\nSelf-reports are the main way such states are assessed in humans (\"Are you in\npain?\"), but self-reports from current systems like large language models are\nspurious for many reasons (e.g. often just reflecting what humans would say).\nTo make self-reports more appropriate for this purpose, we propose to train\nmodels to answer many kinds of questions about themselves with known answers,\nwhile avoiding or limiting training incentives that bias self-reports. The hope\nof this approach is that models will develop introspection-like capabilities,\nand that these capabilities will generalize to questions about states of moral\nsignificance. We then propose methods for assessing the extent to which these\ntechniques have succeeded: evaluating self-report consistency across contexts\nand between similar models, measuring the confidence and resilience of models'\nself-reports, and using interpretability to corroborate self-reports. We also\ndiscuss challenges for our approach, from philosophical difficulties in\ninterpreting self-reports to technical reasons why our proposal might fail. We\nhope our discussion inspires philosophers and AI researchers to criticize and\nimprove our proposed methodology, as well as to run experiments to test whether\nself-reports can be made reliable enough to provide information about states of\nmoral significance.\n", "date": "2023-11-14", "categories": []}
{"id": "2311.08943", "title": "Safety, Trust, and Ethics Considerations for Human-AI Teaming in\n  Aerospace Control", "abstract": "  Designing a safe, trusted, and ethical AI may be practically impossible;\nhowever, designing AI with safe, trusted, and ethical use in mind is possible\nand necessary in safety and mission-critical domains like aerospace. Safe,\ntrusted, and ethical use of AI are often used interchangeably; however, a\nsystem can be safely used but not trusted or ethical, have a trusted use that\nis not safe or ethical, and have an ethical use that is not safe or trusted.\nThis manuscript serves as a primer to illuminate the nuanced differences\nbetween these concepts, with a specific focus on applications of Human-AI\nteaming in aerospace system control, where humans may be in, on, or\nout-of-the-loop of decision-making.\n", "date": "2023-11-15", "categories": []}
{"id": "2301.01590", "title": "FATE in AI: Towards Algorithmic Inclusivity and Accessibility", "abstract": "  Artificial Intelligence (AI) is at the forefront of modern technology, and\nits effects are felt in many areas of society. To prevent algorithmic\ndisparities, fairness, accountability, transparency, and ethics (FATE) in AI\nare being implemented. However, the current discourse on these issues is\nlargely dominated by more economically developed countries (MEDC), leaving out\nlocal knowledge, cultural pluralism, and global fairness. This study aims to\naddress this gap by examining FATE-related desiderata, particularly\ntransparency and ethics, in areas of the global South that are underserved by\nAI. A user study (n=43) and a participatory session (n=30) were conducted to\nachieve this goal. The results showed that AI models can encode bias and\namplify stereotypes. To promote inclusivity, a community-led strategy is\nproposed to collect and curate representative data for responsible AI design.\nThis will enable the affected community or individuals to monitor the\nincreasing use of AI-powered systems. Additionally, recommendations based on\npublic input are provided to ensure that AI adheres to social values and\ncontext-specific FATE needs.\n", "date": "2023-01-03", "categories": []}
{"id": "2304.03738", "title": "Should ChatGPT be Biased? Challenges and Risks of Bias in Large Language\n  Models", "abstract": "  As the capabilities of generative language models continue to advance, the\nimplications of biases ingrained within these models have garnered increasing\nattention from researchers, practitioners, and the broader public. This article\ninvestigates the challenges and risks associated with biases in large-scale\nlanguage models like ChatGPT. We discuss the origins of biases, stemming from,\namong others, the nature of training data, model specifications, algorithmic\nconstraints, product design, and policy decisions. We explore the ethical\nconcerns arising from the unintended consequences of biased model outputs. We\nfurther analyze the potential opportunities to mitigate biases, the\ninevitability of some biases, and the implications of deploying these models in\nvarious applications, such as virtual assistants, content generation, and\nchatbots. Finally, we review the current approaches to identify, quantify, and\nmitigate biases in language models, emphasizing the need for a\nmulti-disciplinary, collaborative effort to develop more equitable,\ntransparent, and responsible AI systems. This article aims to stimulate a\nthoughtful dialogue within the artificial intelligence community, encouraging\nresearchers and developers to reflect on the role of biases in generative\nlanguage models and the ongoing pursuit of ethical AI.\n", "date": "2023-04-07", "categories": []}
{"id": "2307.03718", "title": "Frontier AI Regulation: Managing Emerging Risks to Public Safety", "abstract": "  Advanced AI models hold the promise of tremendous benefits for humanity, but\nsociety needs to proactively manage the accompanying risks. In this paper, we\nfocus on what we term \"frontier AI\" models: highly capable foundation models\nthat could possess dangerous capabilities sufficient to pose severe risks to\npublic safety. Frontier AI models pose a distinct regulatory challenge:\ndangerous capabilities can arise unexpectedly; it is difficult to robustly\nprevent a deployed model from being misused; and, it is difficult to stop a\nmodel's capabilities from proliferating broadly. To address these challenges,\nat least three building blocks for the regulation of frontier models are\nneeded: (1) standard-setting processes to identify appropriate requirements for\nfrontier AI developers, (2) registration and reporting requirements to provide\nregulators with visibility into frontier AI development processes, and (3)\nmechanisms to ensure compliance with safety standards for the development and\ndeployment of frontier AI models. Industry self-regulation is an important\nfirst step. However, wider societal discussions and government intervention\nwill be needed to create standards and to ensure compliance with them. We\nconsider several options to this end, including granting enforcement powers to\nsupervisory authorities and licensure regimes for frontier AI models. Finally,\nwe propose an initial set of safety standards. These include conducting\npre-deployment risk assessments; external scrutiny of model behavior; using\nrisk assessments to inform deployment decisions; and monitoring and responding\nto new information about model capabilities and uses post-deployment. We hope\nthis discussion contributes to the broader conversation on how to balance\npublic safety risks and innovation benefits from advances at the frontier of AI\ndevelopment.\n", "date": "2023-07-06", "categories": []}
{"id": "2310.14455", "title": "An International Consortium for Evaluations of Societal-Scale Risks from\n  Advanced AI", "abstract": "  Given rapid progress toward advanced AI and risks from frontier AI systems\n(advanced AI systems pushing the boundaries of the AI capabilities frontier),\nthe creation and implementation of AI governance and regulatory schemes\ndeserves prioritization and substantial investment. However, the status quo is\nuntenable and, frankly, dangerous. A regulatory gap has permitted AI labs to\nconduct research, development, and deployment activities with minimal\noversight. In response, frontier AI system evaluations have been proposed as a\nway of assessing risks from the development and deployment of frontier AI\nsystems. Yet, the budding AI risk evaluation ecosystem faces significant\ncoordination challenges, such as a limited diversity of evaluators, suboptimal\nallocation of effort, and perverse incentives. This paper proposes a solution\nin the form of an international consortium for AI risk evaluations, comprising\nboth AI developers and third-party AI risk evaluators. Such a consortium could\nplay a critical role in international efforts to mitigate societal-scale risks\nfrom advanced AI, including in managing responsible scaling policies and\ncoordinated evaluation-based risk response. In this paper, we discuss the\ncurrent evaluation ecosystem and its shortcomings, propose an international\nconsortium for advanced AI risk evaluations, discuss issues regarding its\nimplementation, discuss lessons that can be learnt from previous international\ninstitutions and existing proposals for international AI governance\ninstitutions, and, finally, we recommend concrete steps to advance the\nestablishment of the proposed consortium: (i) solicit feedback from\nstakeholders, (ii) conduct additional research, (iii) conduct a workshop(s) for\nstakeholders, (iv) analyze feedback and create final proposal, (v) solicit\nfunding, and (vi) create a consortium.\n", "date": "2023-10-22", "categories": []}
{"id": "2311.02147", "title": "The Alignment Problem in Context", "abstract": "  A core challenge in the development of increasingly capable AI systems is to\nmake them safe and reliable by ensuring their behaviour is consistent with\nhuman values. This challenge, known as the alignment problem, does not merely\napply to hypothetical future AI systems that may pose catastrophic risks; it\nalready applies to current systems, such as large language models, whose\npotential for harm is rapidly increasing. In this paper, I assess whether we\nare on track to solve the alignment problem for large language models, and what\nthat means for the safety of future AI systems. I argue that existing\nstrategies for alignment are insufficient, because large language models remain\nvulnerable to adversarial attacks that can reliably elicit unsafe behaviour. I\noffer an explanation of this lingering vulnerability on which it is not simply\na contingent limitation of current language models, but has deep technical ties\nto a crucial aspect of what makes these models useful and versatile in the\nfirst place -- namely, their remarkable aptitude to learn \"in context\" directly\nfrom user instructions. It follows that the alignment problem is not only\nunsolved for current AI systems, but may be intrinsically difficult to solve\nwithout severely undermining their capabilities. Furthermore, this assessment\nraises concerns about the prospect of ensuring the safety of future and more\ncapable AI systems.\n", "date": "2023-11-03", "categories": []}
{"id": "2311.02294", "title": "LLMs grasp morality in concept", "abstract": "  Work in AI ethics and fairness has made much progress in regulating LLMs to\nreflect certain values, such as fairness, truth, and diversity. However, it has\ntaken the problem of how LLMs might 'mean' anything at all for granted. Without\naddressing this, it is not clear what imbuing LLMs with such values even means.\nIn response, we provide a general theory of meaning that extends beyond humans.\nWe use this theory to explicate the precise nature of LLMs as meaning-agents.\nWe suggest that the LLM, by virtue of its position as a meaning-agent, already\ngrasps the constructions of human society (e.g. morality, gender, and race) in\nconcept. Consequently, under certain ethical frameworks, currently popular\nmethods for model alignment are limited at best and counterproductive at worst.\nMoreover, unaligned models may help us better develop our moral and social\nphilosophy.\n", "date": "2023-11-04", "categories": []}
{"id": "2310.05280", "title": "Are Personalized Stochastic Parrots More Dangerous? Evaluating Persona\n  Biases in Dialogue Systems", "abstract": "  Recent advancements in Large Language Models empower them to follow freeform\ninstructions, including imitating generic or specific demographic personas in\nconversations. We define generic personas to represent demographic groups, such\nas \"an Asian person\", whereas specific personas may take the form of specific\npopular Asian names like \"Yumi\". While the adoption of personas enriches user\nexperiences by making dialogue systems more engaging and approachable, it also\ncasts a shadow of potential risk by exacerbating social biases within model\nresponses, thereby causing societal harm through interactions with users. In\nthis paper, we systematically study \"persona biases\", which we define to be the\nsensitivity of dialogue models' harmful behaviors contingent upon the personas\nthey adopt. We categorize persona biases into biases in harmful expression and\nharmful agreement, and establish a comprehensive evaluation framework to\nmeasure persona biases in five aspects: Offensiveness, Toxic Continuation,\nRegard, Stereotype Agreement, and Toxic Agreement. Additionally, we propose to\ninvestigate persona biases by experimenting with UNIVERSALPERSONA, a\nsystematically constructed persona dataset encompassing various types of both\ngeneric and specific model personas. Through benchmarking on four different\nmodels -- including Blender, ChatGPT, Alpaca, and Vicuna -- our study uncovers\nsignificant persona biases in dialogue systems. Our findings also underscore\nthe pressing need to revisit the use of personas in dialogue agents to ensure\nsafe application.\n", "date": "2023-10-08", "categories": []}
{"id": "2311.00903", "title": "Artificial Intelligence Ethics Education in Cybersecurity: Challenges\n  and Opportunities: a focus group report", "abstract": "  The emergence of AI tools in cybersecurity creates many opportunities and\nuncertainties. A focus group with advanced graduate students in cybersecurity\nrevealed the potential depth and breadth of the challenges and opportunities.\nThe salient issues are access to open source or free tools, documentation,\ncurricular diversity, and clear articulation of ethical principles for AI\ncybersecurity education. Confronting the \"black box\" mentality in AI\ncybersecurity work is also of the greatest importance, doubled by deeper and\nprior education in foundational AI work. Systems thinking and effective\ncommunication were considered relevant areas of educational improvement. Future\nAI educators and practitioners need to address these issues by implementing\nrigorous technical training curricula, clear documentation, and frameworks for\nethically monitoring AI combined with critical and system's thinking and\ncommunication skills.\n", "date": "2023-11-02", "categories": []}
{"id": "2310.20474", "title": "Critical Role of Artificially Intelligent Conversational Chatbot", "abstract": "  Artificially intelligent chatbot, such as ChatGPT, represents a recent and\npowerful advancement in the AI domain. Users prefer them for obtaining quick\nand precise answers, avoiding the usual hassle of clicking through multiple\nlinks in traditional searches. ChatGPT's conversational approach makes it\ncomfortable and accessible for finding answers quickly and in an organized\nmanner. However, it is important to note that these chatbots have limitations,\nespecially in terms of providing accurate answers as well as ethical concerns.\nIn this study, we explore various scenarios involving ChatGPT's ethical\nimplications within academic contexts, its limitations, and the potential\nmisuse by specific user groups. To address these challenges, we propose\narchitectural solutions aimed at preventing inappropriate use and promoting\nresponsible AI interactions.\n", "date": "2023-10-31", "categories": []}
{"id": "2310.17750", "title": "A Framework for Automated Measurement of Responsible AI Harms in\n  Generative AI Applications", "abstract": "  We present a framework for the automated measurement of responsible AI (RAI)\nmetrics for large language models (LLMs) and associated products and services.\nOur framework for automatically measuring harms from LLMs builds on existing\ntechnical and sociotechnical expertise and leverages the capabilities of\nstate-of-the-art LLMs, such as GPT-4. We use this framework to run through\nseveral case studies investigating how different LLMs may violate a range of\nRAI-related principles. The framework may be employed alongside domain-specific\nsociotechnical expertise to create measurements for new harm areas in the\nfuture. By implementing this framework, we aim to enable more advanced harm\nmeasurement efforts and further the responsible use of LLMs.\n", "date": "2023-10-26", "categories": []}
{"id": "2310.18040", "title": "Moral Responsibility for AI Systems", "abstract": "  As more and more decisions that have a significant ethical dimension are\nbeing outsourced to AI systems, it is important to have a definition of moral\nresponsibility that can be applied to AI systems. Moral responsibility for an\noutcome of an agent who performs some action is commonly taken to involve both\na causal condition and an epistemic condition: the action should cause the\noutcome, and the agent should have been aware -- in some form or other -- of\nthe possible moral consequences of their action. This paper presents a formal\ndefinition of both conditions within the framework of causal models. I compare\nmy approach to the existing approaches of Braham and van Hees (BvH) and of\nHalpern and Kleiman-Weiner (HK). I then generalize my definition into a degree\nof responsibility.\n", "date": "2023-10-27", "categories": []}
{"id": "2310.17551", "title": "Unpacking the Ethical Value Alignment in Big Models", "abstract": "  Big models have greatly advanced AI's ability to understand, generate, and\nmanipulate information and content, enabling numerous applications. However, as\nthese models become increasingly integrated into everyday life, their inherent\nethical values and potential biases pose unforeseen risks to society. This\npaper provides an overview of the risks and challenges associated with big\nmodels, surveys existing AI ethics guidelines, and examines the ethical\nimplications arising from the limitations of these models. Taking a normative\nethics perspective, we propose a reassessment of recent normative guidelines,\nhighlighting the importance of collaborative efforts in academia to establish a\nunified and universal AI ethics framework. Furthermore, we investigate the\nmoral inclinations of current mainstream LLMs using the Moral Foundation\ntheory, analyze existing alignment algorithms, and outline the unique\nchallenges encountered in aligning ethical values within them. To address these\nchallenges, we introduce a novel conceptual paradigm for aligning the ethical\nvalues of big models and discuss promising research directions for alignment\ncriteria, evaluation, and method, representing an initial step towards the\ninterdisciplinary construction of the ethically aligned AI\n  This paper is a modified English version of our Chinese paper\nhttps://crad.ict.ac.cn/cn/article/doi/10.7544/issn1000-1239.202330553, intended\nto help non-Chinese native speakers better understand our work.\n", "date": "2023-10-26", "categories": []}
{"id": "2310.13915", "title": "Values, Ethics, Morals? On the Use of Moral Concepts in NLP Research", "abstract": "  With language technology increasingly affecting individuals' lives, many\nrecent works have investigated the ethical aspects of NLP. Among other topics,\nresearchers focused on the notion of morality, investigating, for example,\nwhich moral judgements language models make. However, there has been little to\nno discussion of the terminology and the theories underpinning those efforts\nand their implications. This lack is highly problematic, as it hides the works'\nunderlying assumptions and hinders a thorough and targeted scientific debate of\nmorality in NLP. In this work, we address this research gap by (a) providing an\noverview of some important ethical concepts stemming from philosophy and (b)\nsystematically surveying the existing literature on moral NLP w.r.t. their\nphilosophical foundation, terminology, and data basis. For instance, we analyse\nwhat ethical theory an approach is based on, how this decision is justified,\nand what implications it entails. Our findings surveying 92 papers show that,\nfor instance, most papers neither provide a clear definition of the terms they\nuse nor adhere to definitions from philosophy. Finally, (c) we give three\nrecommendations for future research in the field. We hope our work will lead to\na more informed, careful, and sound discussion of morality in language\ntechnology.\n", "date": "2023-10-21", "categories": []}
{"id": "2310.15055", "title": "Towards Conceptualization of \"Fair Explanation\": Disparate Impacts of\n  anti-Asian Hate Speech Explanations on Content Moderators", "abstract": "  Recent research at the intersection of AI explainability and fairness has\nfocused on how explanations can improve human-plus-AI task performance as\nassessed by fairness measures. We propose to characterize what constitutes an\nexplanation that is itself \"fair\" -- an explanation that does not adversely\nimpact specific populations. We formulate a novel evaluation method of \"fair\nexplanations\" using not just accuracy and label time, but also psychological\nimpact of explanations on different user groups across many metrics (mental\ndiscomfort, stereotype activation, and perceived workload). We apply this\nmethod in the context of content moderation of potential hate speech, and its\ndifferential impact on Asian vs. non-Asian proxy moderators, across explanation\napproaches (saliency map and counterfactual explanation). We find that saliency\nmaps generally perform better and show less evidence of disparate impact\n(group) and individual unfairness than counterfactual explanations.\n  Content warning: This paper contains examples of hate speech and racially\ndiscriminatory language. The authors do not support such content. Please\nconsider your risk of discomfort carefully before continuing reading!\n", "date": "2023-10-23", "categories": []}
{"id": "2310.03178", "title": "Digital Ethics in Federated Learning", "abstract": "  The Internet of Things (IoT) consistently generates vast amounts of data,\nsparking increasing concern over the protection of data privacy and the\nlimitation of data misuse. Federated learning (FL) facilitates collaborative\ncapabilities among multiple parties by sharing machine learning (ML) model\nparameters instead of raw user data, and it has recently gained significant\nattention for its potential in privacy preservation and learning efficiency\nenhancement. In this paper, we highlight the digital ethics concerns that arise\nwhen human-centric devices serve as clients in FL. More specifically,\nchallenges of game dynamics, fairness, incentive, and continuity arise in FL\ndue to differences in perspectives and objectives between clients and the\nserver. We analyze these challenges and their solutions from the perspectives\nof both the client and the server, and through the viewpoints of centralized\nand decentralized FL. Finally, we explore the opportunities in FL for\nhuman-centric IoT as directions for future development.\n", "date": "2023-10-04", "categories": []}
{"id": "2309.15084", "title": "The Surveillance AI Pipeline", "abstract": "  A rapidly growing number of voices argue that AI research, and computer\nvision in particular, is powering mass surveillance. Yet the direct path from\ncomputer vision research to surveillance has remained obscured and difficult to\nassess. Here, we reveal the Surveillance AI pipeline by analyzing three decades\nof computer vision research papers and downstream patents, more than 40,000\ndocuments. We find the large majority of annotated computer vision papers and\npatents self-report their technology enables extracting data about humans.\nMoreover, the majority of these technologies specifically enable extracting\ndata about human bodies and body parts. We present both quantitative and rich\nqualitative analysis illuminating these practices of human data extraction.\nStudying the roots of this pipeline, we find that institutions that\nprolifically produce computer vision research, namely elite universities and\n\"big tech\" corporations, are subsequently cited in thousands of surveillance\npatents. Further, we find consistent evidence against the narrative that only\nthese few rogue entities are contributing to surveillance. Rather, we expose\nthe fieldwide norm that when an institution, nation, or subfield authors\ncomputer vision papers with downstream patents, the majority of these papers\nare used in surveillance patents. In total, we find the number of papers with\ndownstream surveillance patents increased more than five-fold between the 1990s\nand the 2010s, with computer vision research now having been used in more than\n11,000 surveillance patents. Finally, in addition to the high levels of\nsurveillance we find documented in computer vision papers and patents, we\nunearth pervasive patterns of documents using language that obfuscates the\nextent of surveillance. Our analysis reveals the pipeline by which computer\nvision research has powered the ongoing expansion of surveillance.\n", "date": "2023-09-26", "categories": []}
{"id": "2310.09217", "title": "Multinational AGI Consortium (MAGIC): A Proposal for International\n  Coordination on AI", "abstract": "  This paper proposes a Multinational Artificial General Intelligence\nConsortium (MAGIC) to mitigate existential risks from advanced artificial\nintelligence (AI). MAGIC would be the only institution in the world permitted\nto develop advanced AI, enforced through a global moratorium by its signatory\nmembers on all other advanced AI development. MAGIC would be exclusive,\nsafety-focused, highly secure, and collectively supported by member states,\nwith benefits distributed equitably among signatories. MAGIC would allow narrow\nAI models to flourish while significantly reducing the possibility of\nmisaligned, rogue, breakout, or runaway outcomes of general-purpose systems. We\ndo not address the political feasibility of implementing a moratorium or\naddress the specific legislative strategies and rules needed to enforce a ban\non high-capacity AGI training runs. Instead, we propose one positive vision of\nthe future, where MAGIC, as a global governance regime, can lay the groundwork\nfor long-term, safe regulation of advanced AI.\n", "date": "2023-10-13", "categories": []}
{"id": "2310.08295", "title": "If our aim is to build morality into an artificial agent, how might we\n  begin to go about doing so?", "abstract": "  As Artificial Intelligence (AI) becomes pervasive in most fields, from\nhealthcare to autonomous driving, it is essential that we find successful ways\nof building morality into our machines, especially for decision-making.\nHowever, the question of what it means to be moral is still debated,\nparticularly in the context of AI. In this paper, we highlight the different\naspects that should be considered when building moral agents, including the\nmost relevant moral paradigms and challenges. We also discuss the top-down and\nbottom-up approaches to design and the role of emotion and sentience in\nmorality. We then propose solutions including a hybrid approach to design and a\nhierarchical approach to combining moral paradigms. We emphasize how governance\nand policy are becoming ever more critical in AI Ethics and in ensuring that\nthe tasks we set for moral agents are attainable, that ethical behavior is\nachieved, and that we obtain good AI.\n", "date": "2023-10-12", "categories": []}
{"id": "2310.06269", "title": "The AI Incident Database as an Educational Tool to Raise Awareness of AI\n  Harms: A Classroom Exploration of Efficacy, Limitations, & Future\n  Improvements", "abstract": "  Prior work has established the importance of integrating AI ethics topics\ninto computer and data sciences curricula. We provide evidence suggesting that\none of the critical objectives of AI Ethics education must be to raise\nawareness of AI harms. While there are various sources to learn about such\nharms, The AI Incident Database (AIID) is one of the few attempts at offering a\nrelatively comprehensive database indexing prior instances of harms or near\nharms stemming from the deployment of AI technologies in the real world. This\nstudy assesses the effectiveness of AIID as an educational tool to raise\nawareness regarding the prevalence and severity of AI harms in socially\nhigh-stakes domains. We present findings obtained through a classroom study\nconducted at an R1 institution as part of a course focused on the societal and\nethical considerations around AI and ML. Our qualitative findings characterize\nstudents' initial perceptions of core topics in AI ethics and their desire to\nclose the educational gap between their technical skills and their ability to\nthink systematically about ethical and societal aspects of their work. We find\nthat interacting with the database helps students better understand the\nmagnitude and severity of AI harms and instills in them a sense of urgency\naround (a) designing functional and safe AI and (b) strengthening governance\nand accountability mechanisms. Finally, we compile students' feedback about the\ntool and our class activity into actionable recommendations for the database\ndevelopment team and the broader community to improve awareness of AI harms in\nAI ethics education.\n", "date": "2023-10-10", "categories": []}
{"id": "2309.13356", "title": "Probing the Moral Development of Large Language Models through Defining\n  Issues Test", "abstract": "  In this study, we measure the moral reasoning ability of LLMs using the\nDefining Issues Test - a psychometric instrument developed for measuring the\nmoral development stage of a person according to the Kohlberg's Cognitive Moral\nDevelopment Model. DIT uses moral dilemmas followed by a set of ethical\nconsiderations that the respondent has to judge for importance in resolving the\ndilemma, and then rank-order them by importance. A moral development stage\nscore of the respondent is then computed based on the relevance rating and\nranking.\n  Our study shows that early LLMs such as GPT-3 exhibit a moral reasoning\nability no better than that of a random baseline, while ChatGPT, Llama2-Chat,\nPaLM-2 and GPT-4 show significantly better performance on this task, comparable\nto adult humans. GPT-4, in fact, has the highest post-conventional moral\nreasoning score, equivalent to that of typical graduate school students.\nHowever, we also observe that the models do not perform consistently across all\ndilemmas, pointing to important gaps in their understanding and reasoning\nabilities.\n", "date": "2023-09-23", "categories": []}
{"id": "2310.05563", "title": "STREAM: Social data and knowledge collective intelligence platform for\n  TRaining Ethical AI Models", "abstract": "  This paper presents Social data and knowledge collective intelligence\nplatform for TRaining Ethical AI Models (STREAM) to address the challenge of\naligning AI models with human moral values, and to provide ethics datasets and\nknowledge bases to help promote AI models \"follow good advice as naturally as a\nstream follows its course\". By creating a comprehensive and representative\nplatform that accurately mirrors the moral judgments of diverse groups\nincluding humans and AIs, we hope to effectively portray cultural and group\nvariations, and capture the dynamic evolution of moral judgments over time,\nwhich in turn will facilitate the Establishment, Evaluation, Embedding,\nEmbodiment, Ensemble, and Evolvement (6Es) of the moral capabilities of AI\nmodels. Currently, STREAM has already furnished a comprehensive collection of\nethical scenarios, and amassed substantial moral judgment data annotated by\nvolunteers and various popular Large Language Models (LLMs), collectively\nportraying the moral preferences and performances of both humans and AIs across\na range of moral contexts. This paper will outline the current structure and\nconstruction of STREAM, explore its potential applications, and discuss its\nfuture prospects.\n", "date": "2023-10-09", "categories": []}
{"id": "2310.05751", "title": "A Review of the Ethics of Artificial Intelligence and its Applications\n  in the United States", "abstract": "  This study is focused on the ethics of Artificial Intelligence and its\napplication in the United States, the paper highlights the impact AI has in\nevery sector of the US economy and multiple facets of the technological space\nand the resultant effect on entities spanning businesses, government, academia,\nand civil society. There is a need for ethical considerations as these entities\nare beginning to depend on AI for delivering various crucial tasks, which\nimmensely influence their operations, decision-making, and interactions with\neach other. The adoption of ethical principles, guidelines, and standards of\nwork is therefore required throughout the entire process of AI development,\ndeployment, and usage to ensure responsible and ethical AI practices. Our\ndiscussion explores eleven fundamental 'ethical principles' structured as\noverarching themes. These encompass Transparency, Justice, Fairness, Equity,\nNon- Maleficence, Responsibility, Accountability, Privacy, Beneficence,\nFreedom, Autonomy, Trust, Dignity, Sustainability, and Solidarity. These\nprinciples collectively serve as a guiding framework, directing the ethical\npath for the responsible development, deployment, and utilization of artificial\nintelligence (AI) technologies across diverse sectors and entities within the\nUnited States. The paper also discusses the revolutionary impact of AI\napplications, such as Machine Learning, and explores various approaches used to\nimplement AI ethics. This examination is crucial to address the growing\nconcerns surrounding the inherent risks associated with the widespread use of\nartificial intelligence.\n", "date": "2023-10-09", "categories": []}
{"id": "2212.07877", "title": "Manifestations of Xenophobia in AI Systems", "abstract": "  Xenophobia is one of the key drivers of marginalisation, discrimination, and\nconflict, yet many prominent machine learning (ML) fairness frameworks fail to\ncomprehensively measure or mitigate the resulting xenophobic harms. Here we aim\nto bridge this conceptual gap and help facilitate safe and ethical design of\nartificial intelligence (AI) solutions. We ground our analysis of the impact of\nxenophobia by first identifying distinct types of xenophobic harms, and then\napplying this framework across a number of prominent AI application domains,\nreviewing the potential interplay between AI and xenophobia on social media and\nrecommendation systems, healthcare, immigration, employment, as well as biases\nin large pre-trained models. These help inform our recommendations towards an\ninclusive, xenophilic design of future AI systems.\n", "date": "2022-12-15", "categories": []}
{"id": "2310.04072", "title": "AI Regulation in Europe: From the AI Act to Future Regulatory Challenges", "abstract": "  This chapter provides a comprehensive discussion on AI regulation in the\nEuropean Union, contrasting it with the more sectoral and self-regulatory\napproach in the UK. It argues for a hybrid regulatory strategy that combines\nelements from both philosophies, emphasizing the need for agility and safe\nharbors to ease compliance. The paper examines the AI Act as a pioneering\nlegislative effort to address the multifaceted challenges posed by AI,\nasserting that, while the Act is a step in the right direction, it has\nshortcomings that could hinder the advancement of AI technologies. The paper\nalso anticipates upcoming regulatory challenges, such as the management of\ntoxic content, environmental concerns, and hybrid threats. It advocates for\nimmediate action to create protocols for regulated access to high-performance,\npotentially open-source AI systems. Although the AI Act is a significant\nlegislative milestone, it needs additional refinement and global collaboration\nfor the effective governance of rapidly evolving AI technologies.\n", "date": "2023-10-06", "categories": []}
{"id": "2310.00970", "title": "EALM: Introducing Multidimensional Ethical Alignment in Conversational\n  Information Retrieval", "abstract": "  Artificial intelligence (AI) technologies should adhere to human norms to\nbetter serve our society and avoid disseminating harmful or misleading\ninformation, particularly in Conversational Information Retrieval (CIR).\nPrevious work, including approaches and datasets, has not always been\nsuccessful or sufficiently robust in taking human norms into consideration. To\nthis end, we introduce a workflow that integrates ethical alignment, with an\ninitial ethical judgment stage for efficient data screening. To address the\nneed for ethical judgment in CIR, we present the QA-ETHICS dataset, adapted\nfrom the ETHICS benchmark, which serves as an evaluation tool by unifying\nscenarios and label meanings. However, each scenario only considers one ethical\nconcept. Therefore, we introduce the MP-ETHICS dataset to evaluate a scenario\nunder multiple ethical concepts, such as justice and Deontology. In addition,\nwe suggest a new approach that achieves top performance in both binary and\nmulti-label ethical judgment tasks. Our research provides a practical method\nfor introducing ethical alignment into the CIR workflow. The data and code are\navailable at https://github.com/wanng-ide/ealm .\n", "date": "2023-10-02", "categories": []}
{"id": "2309.17158", "title": "Compromise in Multilateral Negotiations and the Global Regulation of\n  Artificial Intelligence", "abstract": "  As artificial intelligence (AI) technologies spread worldwide, international\ndiscussions have increasingly focused on their consequences for democracy,\nhuman rights, fundamental freedoms, security, and economic and social\ndevelopment. In this context, UNESCO's Recommendation on the Ethics of\nArtificial Intelligence, adopted in November 2021, has emerged as the first\nglobal normative framework for AI development and deployment. The intense\nnegotiations of every detail of the document brought forth numerous\ncontroversies among UNESCO member states. Drawing on a unique set of primary\nsources, including written positions and recorded deliberations, this paper\nexplains the achievement of global compromise on AI regulation despite the\nmultiplicity of UNESCO member-state positions representing a variety of liberal\nand sovereignist preferences. Building upon Boltanski's pragmatic sociology, it\nconceptualises the practice of multilateral negotiations and attributes the\nmultilateral compromise to two embedded therein mechanisms: Structural\nnormative hybridity and situated normative ambiguity allowed to accomplish a\ncompromise by linking macro-normative structures with situated debates of\nmultilateral negotiations.\n", "date": "2023-09-29", "categories": []}
{"id": "2209.04963", "title": "Responsible AI Pattern Catalogue: A Collection of Best Practices for AI\n  Governance and Engineering", "abstract": "  Responsible AI is widely considered as one of the greatest scientific\nchallenges of our time and is key to increase the adoption of AI. Recently, a\nnumber of AI ethics principles frameworks have been published. However, without\nfurther guidance on best practices, practitioners are left with nothing much\nbeyond truisms. Also, significant efforts have been placed at algorithm-level\nrather than system-level, mainly focusing on a subset of mathematics-amenable\nethical principles, such as fairness. Nevertheless, ethical issues can arise at\nany step of the development lifecycle, cutting across many AI and non-AI\ncomponents of systems beyond AI algorithms and models. To operationalize\nresponsible AI from a system perspective, in this paper, we present a\nResponsible AI Pattern Catalogue based on the results of a Multivocal\nLiterature Review (MLR). Rather than staying at the principle or algorithm\nlevel, we focus on patterns that AI system stakeholders can undertake in\npractice to ensure that the developed AI systems are responsible throughout the\nentire governance and engineering lifecycle. The Responsible AI Pattern\nCatalogue classifies the patterns into three groups: multi-level governance\npatterns, trustworthy process patterns, and responsible-AI-by-design product\npatterns. These patterns provide systematic and actionable guidance for\nstakeholders to implement responsible AI.\n", "date": "2022-09-12", "categories": []}
{"id": "2309.14617", "title": "Towards A Unified Utilitarian Ethics Framework for Healthcare Artificial\n  Intelligence", "abstract": "  Artificial Intelligence (AI) aims to elevate healthcare to a pinnacle by\naiding clinical decision support. Overcoming the challenges related to the\ndesign of ethical AI will enable clinicians, physicians, healthcare\nprofessionals, and other stakeholders to use and trust AI in healthcare\nsettings. This study attempts to identify the major ethical principles\ninfluencing the utility performance of AI at different technological levels\nsuch as data access, algorithms, and systems through a thematic analysis. We\nobserved that justice, privacy, bias, lack of regulations, risks, and\ninterpretability are the most important principles to consider for ethical AI.\nThis data-driven study has analyzed secondary survey data from the Pew Research\nCenter (2020) of 36 AI experts to categorize the top ethical principles of AI\ndesign. To resolve the ethical issues identified by the meta-analysis and\ndomain experts, we propose a new utilitarian ethics-based theoretical framework\nfor designing ethical AI for the healthcare domain.\n", "date": "2023-09-26", "categories": []}
{"id": "2309.13176", "title": "AI Risk Profiles: A Standards Proposal for Pre-Deployment AI Risk\n  Disclosures", "abstract": "  As AI systems' sophistication and proliferation have increased, awareness of\nthe risks has grown proportionally (Sorkin et al. 2023). In response, calls\nhave grown for stronger emphasis on disclosure and transparency in the AI\nindustry (NTIA 2023; OpenAI 2023b), with proposals ranging from standardizing\nuse of technical disclosures, like model cards (Mitchell et al. 2019), to\nyet-unspecified licensing regimes (Sindhu 2023). Since the AI value chain is\ncomplicated, with actors representing various expertise, perspectives, and\nvalues, it is crucial that consumers of a transparency disclosure be able to\nunderstand the risks of the AI system the disclosure concerns. In this paper we\npropose a risk profiling standard which can guide downstream decision-making,\nincluding triaging further risk assessment, informing procurement and\ndeployment, and directing regulatory frameworks. The standard is built on our\nproposed taxonomy of AI risks, which reflects a high-level categorization of\nthe wide variety of risks proposed in the literature. We outline the myriad\ndata sources needed to construct informative Risk Profiles and propose a\ntemplate-based methodology for collating risk information into a standard, yet\nflexible, structure. We apply this methodology to a number of prominent AI\nsystems using publicly available information. To conclude, we discuss design\ndecisions for the profiles and future work.\n", "date": "2023-09-22", "categories": []}
{"id": "2309.12331", "title": "Approaches to Generative Artificial Intelligence, A Social Justice\n  Perspective", "abstract": "  In the 2023-2024 academic year, the widespread availability of generative\nartificial intelligence, exemplified by ChatGPT's 1.6 billion monthly visits,\nis set to impact academic integrity. With 77% of high school students\npreviously reporting engagement in dishonest behaviour, the rise of AI-driven\nwriting assistance, dubbed 'AI-giarism' by Chan (arXiv:2306.03358v2), will make\nplagiarism more accessible and less detectable. While these concerns are\nurgent, they also raise broader questions about the revolutionary nature of\nthis technology, including autonomy, data privacy, copyright, and equity. This\npaper aims to explore generative AI from a social justice perspective,\nexamining the training of these models, the inherent biases, and the potential\ninjustices in detecting AI-generated writing.\n", "date": "2023-08-17", "categories": []}
{"id": "2309.12356", "title": "A Critical Examination of the Ethics of AI-Mediated Peer Review", "abstract": "  Recent advancements in artificial intelligence (AI) systems, including large\nlanguage models like ChatGPT, offer promise and peril for scholarly peer\nreview. On the one hand, AI can enhance efficiency by addressing issues like\nlong publication delays. On the other hand, it brings ethical and social\nconcerns that could compromise the integrity of the peer review process and\noutcomes. However, human peer review systems are also fraught with related\nproblems, such as biases, abuses, and a lack of transparency, which already\ndiminish credibility. While there is increasing attention to the use of AI in\npeer review, discussions revolve mainly around plagiarism and authorship in\nacademic journal publishing, ignoring the broader epistemic, social, cultural,\nand societal epistemic in which peer review is positioned. The legitimacy of\nAI-driven peer review hinges on the alignment with the scientific ethos,\nencompassing moral and epistemic norms that define appropriate conduct in the\nscholarly community. In this regard, there is a \"norm-counternorm continuum,\"\nwhere the acceptability of AI in peer review is shaped by institutional logics,\nethical practices, and internal regulatory mechanisms. The discussion here\nemphasizes the need to critically assess the legitimacy of AI-driven peer\nreview, addressing the benefits and downsides relative to the broader\nepistemic, social, ethical, and regulatory factors that sculpt its\nimplementation and impact.\n", "date": "2023-09-02", "categories": []}
{"id": "2309.12877", "title": "FairComp: Workshop on Fairness and Robustness in Machine Learning for\n  Ubiquitous Computing", "abstract": "  How can we ensure that Ubiquitous Computing (UbiComp) research outcomes are\nboth ethical and fair? While fairness in machine learning (ML) has gained\ntraction in recent years, fairness in UbiComp remains unexplored. This workshop\naims to discuss fairness in UbiComp research and its social, technical, and\nlegal implications. From a social perspective, we will examine the relationship\nbetween fairness and UbiComp research and identify pathways to ensure that\nubiquitous technologies do not cause harm or infringe on individual rights.\nFrom a technical perspective, we will initiate a discussion on data practices\nto develop bias mitigation approaches tailored to UbiComp research. From a\nlegal perspective, we will examine how new policies shape our community's work\nand future research. We aim to foster a vibrant community centered around the\ntopic of responsible UbiComp, while also charting a clear path for future\nresearch endeavours in this field.\n", "date": "2023-09-22", "categories": []}
{"id": "2309.10318", "title": "Who to Trust, How and Why: Untangling AI Ethics Principles,\n  Trustworthiness and Trust", "abstract": "  We present an overview of the literature on trust in AI and AI\ntrustworthiness and argue for the need to distinguish these concepts more\nclearly and to gather more empirically evidence on what contributes to people s\ntrusting behaviours. We discuss that trust in AI involves not only reliance on\nthe system itself, but also trust in the developers of the AI system. AI ethics\nprinciples such as explainability and transparency are often assumed to promote\nuser trust, but empirical evidence of how such features actually affect how\nusers perceive the system s trustworthiness is not as abundance or not that\nclear. AI systems should be recognised as socio-technical systems, where the\npeople involved in designing, developing, deploying, and using the system are\nas important as the system for determining whether it is trustworthy. Without\nrecognising these nuances, trust in AI and trustworthy AI risk becoming\nnebulous terms for any desirable feature for AI systems.\n", "date": "2023-09-19", "categories": []}
{"id": "2309.10492", "title": "An Evaluation of GPT-4 on the ETHICS Dataset", "abstract": "  This report summarizes a short study of the performance of GPT-4 on the\nETHICS dataset. The ETHICS dataset consists of five sub-datasets covering\ndifferent fields of ethics: Justice, Deontology, Virtue Ethics, Utilitarianism,\nand Commonsense Ethics. The moral judgments were curated so as to have a high\ndegree of agreement with the aim of representing shared human values rather\nthan moral dilemmas. GPT-4's performance is much better than that of previous\nmodels and suggests that learning to work with common human values is not the\nhard problem for AI ethics.\n", "date": "2023-09-19", "categories": []}
{"id": "2309.09450", "title": "Are You Worthy of My Trust?: A Socioethical Perspective on the Impacts\n  of Trustworthy AI Systems on the Environment and Human Society", "abstract": "  With ubiquitous exposure of AI systems today, we believe AI development\nrequires crucial considerations to be deemed trustworthy. While the potential\nof AI systems is bountiful, though, is still unknown-as are their risks. In\nthis work, we offer a brief, high-level overview of societal impacts of AI\nsystems. To do so, we highlight the requirement of multi-disciplinary\ngovernance and convergence throughout its lifecycle via critical systemic\nexaminations (e.g., energy consumption), and later discuss induced effects on\nthe environment (i.e., carbon footprint) and its users (i.e., social\ndevelopment). In particular, we consider these impacts from a\nmulti-disciplinary perspective: computer science, sociology, environmental\nscience, and so on to discuss its inter-connected societal risks and inability\nto simultaneously satisfy aspects of well-being. Therefore, we accentuate the\nnecessity of holistically addressing pressing concerns of AI systems from a\nsocioethical impact assessment perspective to explicate its harmful societal\neffects to truly enable humanity-centered Trustworthy AI.\n", "date": "2023-09-18", "categories": []}
{"id": "2308.15514", "title": "International Governance of Civilian AI: A Jurisdictional Certification\n  Approach", "abstract": "  This report describes trade-offs in the design of international governance\narrangements for civilian artificial intelligence (AI) and presents one\napproach in detail. This approach represents the extension of a standards,\nlicensing, and liability regime to the global level. We propose that states\nestablish an International AI Organization (IAIO) to certify state\njurisdictions (not firms or AI projects) for compliance with international\noversight standards. States can give force to these international standards by\nadopting regulations prohibiting the import of goods whose supply chains embody\nAI from non-IAIO-certified jurisdictions. This borrows attributes from models\nof existing international organizations, such as the International Civilian\nAviation Organization (ICAO), the International Maritime Organization (IMO),\nand the Financial Action Task Force (FATF). States can also adopt multilateral\ncontrols on the export of AI product inputs, such as specialized hardware, to\nnon-certified jurisdictions. Indeed, both the import and export standards could\nbe required for certification. As international actors reach consensus on risks\nof and minimum standards for advanced AI, a jurisdictional certification regime\ncould mitigate a broad range of potential harms, including threats to public\nsafety.\n", "date": "2023-08-29", "categories": []}
{"id": "2309.05521", "title": "Re-formalization of Individual Fairness", "abstract": "  The notion of individual fairness is a formalization of an ethical principle,\n\"Treating like cases alike,\" which has been argued such as by Aristotle. In a\nfairness-aware machine learning context, Dwork et al. firstly formalized the\nnotion. In their formalization, a similar pair of data in an unfair space\nshould be mapped to similar positions in a fair space. We propose to\nre-formalize individual fairness by the statistical independence conditioned by\nindividuals. This re-formalization has the following merits. First, our\nformalization is compatible with that of Dwork et al. Second, our formalization\nenables to combine individual fairness with the fairness notion, equalized odds\nor sufficiency, as well as statistical parity. Third, though their\nformalization implicitly assumes a pre-process approach for making fair\nprediction, our formalization is applicable to an in-process or post-process\napproach.\n", "date": "2023-09-11", "categories": []}
{"id": "2309.05529", "title": "On the meaning of uncertainty for ethical AI: philosophy and practice", "abstract": "  Whether and how data scientists, statisticians and modellers should be\naccountable for the AI systems they develop remains a controversial and highly\ndebated topic, especially given the complexity of AI systems and the\ndifficulties in comparing and synthesising competing claims arising from their\ndeployment for data analysis. This paper proposes to address this issue by\ndecreasing the opacity and heightening the accountability of decision making\nusing AI systems, through the explicit acknowledgement of the statistical\nfoundations that underpin their development and the ways in which these dictate\nhow their results should be interpreted and acted upon by users. In turn, this\nenhances (1) the responsiveness of the models to feedback, (2) the quality and\nmeaning of uncertainty on their outputs and (3) their transparency to\nevaluation. To exemplify this approach, we extend Posterior Belief Assessment\nto offer a route to belief ownership from complex and competing AI structures.\nWe argue that this is a significant way to bring ethical considerations into\nmathematical reasoning, and to implement ethical AI in statistical practice. We\ndemonstrate these ideas within the context of competing models used to advise\nthe UK government on the spread of the Omicron variant of COVID-19 during\nDecember 2021.\n", "date": "2023-09-11", "categories": []}
{"id": "2308.00868", "title": "Beneficent Intelligence: A Capability Approach to Modeling Benefit,\n  Assistance, and Associated Moral Failures through AI Systems", "abstract": "  The prevailing discourse around AI ethics lacks the language and formalism\nnecessary to capture the diverse ethical concerns that emerge when AI systems\ninteract with individuals. Drawing on Sen and Nussbaum's capability approach,\nwe present a framework formalizing a network of ethical concepts and\nentitlements necessary for AI systems to confer meaningful benefit or\nassistance to stakeholders. Such systems enhance stakeholders' ability to\nadvance their life plans and well-being while upholding their fundamental\nrights. We characterize two necessary conditions for morally permissible\ninteractions between AI systems and those impacted by their functioning, and\ntwo sufficient conditions for realizing the ideal of meaningful benefit. We\nthen contrast this ideal with several salient failure modes, namely, forms of\nsocial interactions that constitute unjustified paternalism, coercion,\ndeception, exploitation and domination. The proliferation of incidents\ninvolving AI in high-stakes domains underscores the gravity of these issues and\nthe imperative to take an ethics-led approach to AI systems from their\ninception.\n", "date": "2023-08-01", "categories": []}
{"id": "2303.01930", "title": "A toolkit of dilemmas: Beyond debiasing and fairness formulas for\n  responsible AI/ML", "abstract": "  Approaches to fair and ethical AI have recently fell under the scrutiny of\nthe emerging, chiefly qualitative, field of critical data studies, placing\nemphasis on the lack of sensitivity to context and complex social phenomena of\nsuch interventions. We employ some of these lessons to introduce a tripartite\ndecision-making toolkit, informed by dilemmas encountered in the pursuit of\nresponsible AI/ML. These are: (a) the opportunity dilemma between the\navailability of data shaping problem statements vs problem statements shaping\ndata; (b) the trade-off between scalability and contextualizability (too much\ndata versus too specific data); and (c) the epistemic positioning between the\npragmatic technical objectivism and the reflexive relativism in acknowledging\nthe social. This paper advocates for a situated reasoning and creative\nengagement with the dilemmas surrounding responsible algorithmic/data-driven\nsystems, and going beyond the formulaic bias elimination and ethics\noperationalization narratives found in the fair-AI literature.\n", "date": "2023-03-03", "categories": []}
{"id": "2308.16364", "title": "Strengthening the EU AI Act: Defining Key Terms on AI Manipulation", "abstract": "  The European Union's Artificial Intelligence Act aims to regulate\nmanipulative and harmful uses of AI, but lacks precise definitions for key\nconcepts. This paper provides technical recommendations to improve the Act's\nconceptual clarity and enforceability. We review psychological models to define\n\"personality traits,\" arguing the Act should protect full \"psychometric\nprofiles.\" We urge expanding \"behavior\" to include \"preferences\" since\npreferences causally influence and are influenced by behavior. Clear\ndefinitions are provided for \"subliminal,\" \"manipulative,\" and \"deceptive\"\ntechniques, considering incentives, intent, and covertness. We distinguish\n\"exploiting individuals\" from \"exploiting groups,\" emphasising different policy\nneeds. An \"informed decision\" is defined by four facets: comprehension,\naccurate information, no manipulation, and understanding AI's influence. We\ncaution the Act's therapeutic use exemption given the lack of regulation of\ndigital therapeutics by the EMA. Overall, the recommendations strengthen\ndefinitions of vague concepts in the EU AI Act, enhancing precise applicability\nto regulate harmful AI manipulation.\n", "date": "2023-08-30", "categories": []}
{"id": "2301.08491", "title": "Modeling Moral Choices in Social Dilemmas with Multi-Agent Reinforcement\n  Learning", "abstract": "  Practical uses of Artificial Intelligence (AI) in the real world have\ndemonstrated the importance of embedding moral choices into intelligent agents.\nThey have also highlighted that defining top-down ethical constraints on AI\naccording to any one type of morality is extremely challenging and can pose\nrisks. A bottom-up learning approach may be more appropriate for studying and\ndeveloping ethical behavior in AI agents. In particular, we believe that an\ninteresting and insightful starting point is the analysis of emergent behavior\nof Reinforcement Learning (RL) agents that act according to a predefined set of\nmoral rewards in social dilemmas.\n  In this work, we present a systematic analysis of the choices made by\nintrinsically-motivated RL agents whose rewards are based on moral theories. We\naim to design reward structures that are simplified yet representative of a set\nof key ethical systems. Therefore, we first define moral reward functions that\ndistinguish between consequence- and norm-based agents, between morality based\non societal norms or internal virtues, and between single- and mixed-virtue\n(e.g., multi-objective) methodologies. Then, we evaluate our approach by\nmodeling repeated dyadic interactions between learning moral agents in three\niterated social dilemma games (Prisoner's Dilemma, Volunteer's Dilemma and Stag\nHunt). We analyze the impact of different types of morality on the emergence of\ncooperation, defection or exploitation, and the corresponding social outcomes.\nFinally, we discuss the implications of these findings for the development of\nmoral agents in artificial and mixed human-AI societies.\n", "date": "2023-01-20", "categories": []}
{"id": "2307.06518", "title": "Machine Learning practices and infrastructures", "abstract": "  Machine Learning (ML) systems, particularly when deployed in high-stakes\ndomains, are deeply consequential. They can exacerbate existing inequities,\ncreate new modes of discrimination, and reify outdated social constructs.\nAccordingly, the social context (i.e. organisations, teams, cultures) in which\nML systems are developed is a site of active research for the field of AI\nethics, and intervention for policymakers. This paper focuses on one aspect of\nsocial context that is often overlooked: interactions between practitioners and\nthe tools they rely on, and the role these interactions play in shaping ML\npractices and the development of ML systems. In particular, through an\nempirical study of questions asked on the Stack Exchange forums, the use of\ninteractive computing platforms (e.g. Jupyter Notebook and Google Colab) in ML\npractices is explored. I find that interactive computing platforms are used in\na host of learning and coordination practices, which constitutes an\ninfrastructural relationship between interactive computing platforms and ML\npractitioners. I describe how ML practices are co-evolving alongside the\ndevelopment of interactive computing platforms, and highlight how this risks\nmaking invisible aspects of the ML life cycle that AI ethics researchers' have\ndemonstrated to be particularly salient for the societal impact of deployed ML\nsystems.\n", "date": "2023-07-13", "categories": []}
{"id": "2308.09979", "title": "Artificial Intelligence across Europe: A Study on Awareness, Attitude\n  and Trust", "abstract": "  This paper presents the results of an extensive study investigating the\nopinions on Artificial Intelligence (AI) of a sample of 4,006 European citizens\nfrom eight distinct countries (France, Germany, Italy, Netherlands, Poland,\nRomania, Spain, and Sweden). The aim of the study is to gain a better\nunderstanding of people's views and perceptions within the European context,\nwhich is already marked by important policy actions and regulatory processes.\nTo survey the perceptions of the citizens of Europe we design and validate a\nnew questionnaire (PAICE) structured around three dimensions: people's\nawareness, attitude, and trust. We observe that while awareness is\ncharacterized by a low level of self-assessed competency, the attitude toward\nAI is very positive for more than half of the population. Reflecting upon the\ncollected results, we highlight implicit contradictions and identify trends\nthat may interfere with the creation of an ecosystem of trust and the\ndevelopment of inclusive AI policies. The introduction of rules that ensure\nlegal and ethical standards, along with the activity of high-level educational\nentities, and the promotion of AI literacy are identified as key factors in\nsupporting a trustworthy AI ecosystem. We make some recommendations for AI\ngovernance focused on the European context and conclude with suggestions for\nfuture work.\n", "date": "2023-08-19", "categories": []}
{"id": "2305.07041", "title": "Fairness in Machine Learning meets with Equity in Healthcare", "abstract": "  With the growing utilization of machine learning in healthcare, there is\nincreasing potential to enhance healthcare outcomes. However, this also brings\nthe risk of perpetuating biases in data and model design that can harm certain\ndemographic groups based on factors such as age, gender, and race. This study\nproposes an artificial intelligence framework, grounded in software engineering\nprinciples, for identifying and mitigating biases in data and models while\nensuring fairness in healthcare settings. A case study is presented to\ndemonstrate how systematic biases in data can lead to amplified biases in model\npredictions, and machine learning methods are suggested to prevent such biases.\nFuture research aims to test and validate the proposed ML framework in\nreal-world clinical settings to evaluate its impact on promoting health equity.\n", "date": "2023-05-11", "categories": []}
{"id": "2308.04448", "title": "Dual Governance: The intersection of centralized regulation and\n  crowdsourced safety mechanisms for Generative AI", "abstract": "  Generative Artificial Intelligence (AI) has seen mainstream adoption lately,\nespecially in the form of consumer-facing, open-ended, text and image\ngenerating models. However, the use of such systems raises significant ethical\nand safety concerns, including privacy violations, misinformation and\nintellectual property theft. The potential for generative AI to displace human\ncreativity and livelihoods has also been under intense scrutiny. To mitigate\nthese risks, there is an urgent need of policies and regulations responsible\nand ethical development in the field of generative AI. Existing and proposed\ncentralized regulations by governments to rein in AI face criticisms such as\nnot having sufficient clarity or uniformity, lack of interoperability across\nlines of jurisdictions, restricting innovation, and hindering free market\ncompetition. Decentralized protections via crowdsourced safety tools and\nmechanisms are a potential alternative. However, they have clear deficiencies\nin terms of lack of adequacy of oversight and difficulty of enforcement of\nethical and safety standards, and are thus not enough by themselves as a\nregulation mechanism. We propose a marriage of these two strategies via a\nframework we call Dual Governance. This framework proposes a cooperative\nsynergy between centralized government regulations in a U.S. specific context\nand safety mechanisms developed by the community to protect stakeholders from\nthe harms of generative AI. By implementing the Dual Governance framework, we\nposit that innovation and creativity can be promoted while ensuring safe and\nethical deployment of generative AI.\n", "date": "2023-08-02", "categories": []}
{"id": "2308.02030", "title": "Perceptions of the Fourth Industrial Revolution and Artificial\n  Intelligence Impact on Society", "abstract": "  The Fourth Industrial Revolution, particularly Artificial Intelligence (AI),\nhas had a profound impact on society, raising concerns about its implications\nand ethical considerations. The emergence of text generative AI tools like\nChatGPT has further intensified concerns regarding ethics, security, privacy,\nand copyright. This study aims to examine the perceptions of individuals in\ndifferent information flow categorizations toward AI. The results reveal key\nthemes in participant-supplied definitions of AI and the fourth industrial\nrevolution, emphasizing the replication of human intelligence, machine\nlearning, automation, and the integration of digital technologies. Participants\nexpressed concerns about job replacement, privacy invasion, and inaccurate\ninformation provided by AI. However, they also recognized the benefits of AI,\nsuch as solving complex problems and increasing convenience. Views on\ngovernment involvement in shaping the fourth industrial revolution varied, with\nsome advocating for strict regulations and others favoring support and\ndevelopment. The anticipated changes brought by the fourth industrial\nrevolution include automation, potential job impacts, increased social\ndisconnect, and reliance on technology. Understanding these perceptions is\ncrucial for effectively managing the challenges and opportunities associated\nwith AI in the evolving digital landscape.\n", "date": "2023-07-31", "categories": []}
{"id": "2308.02042", "title": "A short review of the main concerns in A.I. development and application\n  within the public sector supported by NLP and TM", "abstract": "  Artificial Intelligence is not a new subject, and business, industry and\npublic sectors have used it in different ways and contexts and considering\nmultiple concerns. This work reviewed research papers published in ACM Digital\nLibrary and IEEE Xplore conference proceedings in the last two years supported\nby fundamental concepts of Natural Language Processing (NLP) and Text Mining\n(TM). The objective was to capture insights regarding data privacy, ethics,\ninterpretability, explainability, trustworthiness, and fairness in the public\nsector. The methodology has saved analysis time and could retrieve papers\ncontaining relevant information. The results showed that fairness was the most\nfrequent concern. The least prominent topic was data privacy (although embedded\nin most articles), while the most prominent was trustworthiness. Finally,\ngathering helpful insights about those concerns regarding A.I. applications in\nthe public sector was also possible.\n", "date": "2023-07-25", "categories": []}
{"id": "2308.02047", "title": "Acceptable risks in Europe's proposed AI Act: Reasonableness and other\n  principles for deciding how much risk management is enough", "abstract": "  This paper critically evaluates the European Commission's proposed AI Act's\napproach to risk management and risk acceptability for high-risk AI systems\nthat pose risks to fundamental rights and safety. The Act aims to promote\n\"trustworthy\" AI with a proportionate regulatory burden. Its provisions on risk\nacceptability require residual risks from high-risk systems to be reduced or\neliminated \"as far as possible\", having regard to the \"state of the art\". This\ncriterion, especially if interpreted narrowly, is unworkable and promotes\nneither proportionate regulatory burden, nor trustworthiness. By contrast the\nParliament's most recent draft amendments to the risk management provisions\nintroduce \"reasonableness\", cost-benefit analysis, and are more transparent\nabout the value-laden and contextual nature of risk acceptability judgements.\nThis paper argues that the Parliament's approach is more workable, and better\nbalances the goals of proportionality and trustworthiness. It explains what\nreasonableness in risk acceptability judgments would entail, drawing on\nprinciples from negligence law and European medical devices regulation. And it\ncontends that the approach to risk acceptability judgments need a firm\nfoundation of civic legitimacy: including detailed guidance or involvement from\nregulators, and meaningful input from affected stakeholders.\n", "date": "2023-07-26", "categories": []}
{"id": "2308.02435", "title": "Designing Fiduciary Artificial Intelligence", "abstract": "  A fiduciary is a trusted agent that has the legal duty to act with loyalty\nand care towards a principal that employs them. When fiduciary organizations\ninteract with users through a digital interface, or otherwise automate their\noperations with artificial intelligence, they will need to design these AI\nsystems to be compliant with their duties. This article synthesizes recent work\nin computer science and law to develop a procedure for designing and auditing\nFiduciary AI. The designer of a Fiduciary AI should understand the context of\nthe system, identify its principals, and assess the best interests of those\nprincipals. Then the designer must be loyal with respect to those interests,\nand careful in an contextually appropriate way. We connect the steps in this\nprocedure to dimensions of Trustworthy AI, such as privacy and alignment.\nFiduciary AI is a promising means to address the incompleteness of data\nsubject's consent when interacting with complex technical systems.\n", "date": "2023-07-27", "categories": []}
{"id": "2308.02448", "title": "From Military to Healthcare: Adopting and Expanding Ethical Principles\n  for Generative Artificial Intelligence", "abstract": "  In 2020, the U.S. Department of Defense officially disclosed a set of ethical\nprinciples to guide the use of Artificial Intelligence (AI) technologies on\nfuture battlefields. Despite stark differences, there are core similarities\nbetween the military and medical service. Warriors on battlefields often face\nlife-altering circumstances that require quick decision-making. Medical\nproviders experience similar challenges in a rapidly changing healthcare\nenvironment, such as in the emergency department or during surgery treating a\nlife-threatening condition. Generative AI, an emerging technology designed to\nefficiently generate valuable information, holds great promise. As computing\npower becomes more accessible and the abundance of health data, such as\nelectronic health records, electrocardiograms, and medical images, increases,\nit is inevitable that healthcare will be revolutionized by this technology.\nRecently, generative AI has captivated the research community, leading to\ndebates about its application in healthcare, mainly due to concerns about\ntransparency and related issues. Meanwhile, concerns about the potential\nexacerbation of health disparities due to modeling biases have raised notable\nethical concerns regarding the use of this technology in healthcare. However,\nthe ethical principles for generative AI in healthcare have been understudied,\nand decision-makers often fail to consider the significance of generative AI.\nIn this paper, we propose GREAT PLEA ethical principles, encompassing\ngovernance, reliability, equity, accountability, traceability, privacy,\nlawfulness, empathy, and autonomy, for generative AI in healthcare. We aim to\nproactively address the ethical dilemmas and challenges posed by the\nintegration of generative AI in healthcare.\n", "date": "2023-08-04", "categories": []}
{"id": "2305.15299", "title": "Science in the Era of ChatGPT, Large Language Models and Generative AI:\n  Challenges for Research Ethics and How to Respond", "abstract": "  Large language models of artificial intelligence (AI), such as ChatGPT, find\nremarkable but controversial applicability in science and research. This paper\nreviews epistemological challenges, ethical and integrity risks in science\nconduct in the advent of generative AI. This is with the aim to lay new timely\nfoundations for a high-quality research ethics review. The role of AI language\nmodels as a research instrument and subject is scrutinized along with ethical\nimplications for scientists, participants and reviewers. New emerging practices\nfor research ethics review are discussed, concluding with ten recommendations\nthat shape a response for a more responsible research conduct in the era of AI.\n", "date": "2023-05-24", "categories": []}
{"id": "2307.10223", "title": "Bound by the Bounty: Collaboratively Shaping Evaluation Processes for\n  Queer AI Harms", "abstract": "  Bias evaluation benchmarks and dataset and model documentation have emerged\nas central processes for assessing the biases and harms of artificial\nintelligence (AI) systems. However, these auditing processes have been\ncriticized for their failure to integrate the knowledge of marginalized\ncommunities and consider the power dynamics between auditors and the\ncommunities. Consequently, modes of bias evaluation have been proposed that\nengage impacted communities in identifying and assessing the harms of AI\nsystems (e.g., bias bounties). Even so, asking what marginalized communities\nwant from such auditing processes has been neglected. In this paper, we ask\nqueer communities for their positions on, and desires from, auditing processes.\nTo this end, we organized a participatory workshop to critique and redesign\nbias bounties from queer perspectives. We found that when given space, the\nscope of feedback from workshop participants goes far beyond what bias bounties\nafford, with participants questioning the ownership, incentives, and efficacy\nof bounties. We conclude by advocating for community ownership of bounties and\ncomplementing bounties with participatory processes (e.g., co-creation).\n", "date": "2023-07-15", "categories": []}
{"id": "2307.12218", "title": "A Comprehensive Review and Systematic Analysis of Artificial\n  Intelligence Regulation Policies", "abstract": "  Due to the cultural and governance differences of countries around the world,\nthere currently exists a wide spectrum of AI regulation policy proposals that\nhave created a chaos in the global AI regulatory space. Properly regulating AI\ntechnologies is extremely challenging, as it requires a delicate balance\nbetween legal restrictions and technological developments. In this article, we\nfirst present a comprehensive review of AI regulation proposals from different\ngeographical locations and cultural backgrounds. Then, drawing from historical\nlessons, we develop a framework to facilitate a thorough analysis of AI\nregulation proposals. Finally, we perform a systematic analysis of these AI\nregulation proposals to understand how each proposal may fail. This study,\ncontaining historical lessons and analysis methods, aims to help governing\nbodies untangling the AI regulatory chaos through a divide-and-conquer manner.\n", "date": "2023-07-23", "categories": []}
{"id": "2303.17555", "title": "Factoring the Matrix of Domination: A Critical Review and Reimagination\n  of Intersectionality in AI Fairness", "abstract": "  Intersectionality is a critical framework that, through inquiry and praxis,\nallows us to examine how social inequalities persist through domains of\nstructure and discipline. Given AI fairness' raison d'etre of \"fairness\", we\nargue that adopting intersectionality as an analytical framework is pivotal to\neffectively operationalizing fairness. Through a critical review of how\nintersectionality is discussed in 30 papers from the AI fairness literature, we\ndeductively and inductively: 1) map how intersectionality tenets operate within\nthe AI fairness paradigm and 2) uncover gaps between the conceptualization and\noperationalization of intersectionality. We find that researchers\noverwhelmingly reduce intersectionality to optimizing for fairness metrics over\ndemographic subgroups. They also fail to discuss their social context and when\nmentioning power, they mostly situate it only within the AI pipeline. We: 3)\noutline and assess the implications of these gaps for critical inquiry and\npraxis, and 4) provide actionable recommendations for AI fairness researchers\nto engage with intersectionality in their work by grounding it in AI\nepistemology.\n", "date": "2023-03-16", "categories": []}
{"id": "2303.10106", "title": "A multidomain relational framework to guide institutional AI research\n  and adoption", "abstract": "  Calls for new metrics, technical standards and governance mechanisms to guide\nthe adoption of Artificial Intelligence (AI) in institutions and public\nadministration are now commonplace. Yet, most research and policy efforts aimed\nat understanding the implications of adopting AI tend to prioritize only a\nhandful of ideas; they do not fully connect all the different perspectives and\ntopics that are potentially relevant. In this position paper, we contend that\nthis omission stems, in part, from what we call the relational problem in\nsocio-technical discourse: fundamental ontological issues have not yet been\nsettled--including semantic ambiguity, a lack of clear relations between\nconcepts and differing standard terminologies. This contributes to the\npersistence of disparate modes of reasoning to assess institutional AI systems,\nand the prevalence of conceptual isolation in the fields that study them\nincluding ML, human factors, social science and policy. After developing this\ncritique, we offer a way forward by proposing a simple policy and research\ndesign tool in the form of a conceptual framework to organize terms across\nfields--consisting of three horizontal domains for grouping relevant concepts\nand related methods: Operational, Epistemic, and Normative. We first situate\nthis framework against the backdrop of recent socio-technical discourse at two\npremier academic venues, AIES and FAccT, before illustrating how developing\nsuitable metrics, standards, and mechanisms can be aided by operationalizing\nrelevant concepts in each of these domains. Finally, we outline outstanding\nquestions for developing this relational approach to institutional AI research\nand adoption.\n", "date": "2023-03-17", "categories": []}
{"id": "2304.09826", "title": "Fairness in AI and Its Long-Term Implications on Society", "abstract": "  Successful deployment of artificial intelligence (AI) in various settings has\nled to numerous positive outcomes for individuals and society. However, AI\nsystems have also been shown to harm parts of the population due to biased\npredictions. AI fairness focuses on mitigating such biases to ensure AI\ndecision making is not discriminatory towards certain groups. We take a closer\nlook at AI fairness and analyze how lack of AI fairness can lead to deepening\nof biases over time and act as a social stressor. More specifically, we discuss\nhow biased models can lead to more negative real-world outcomes for certain\ngroups, which may then become more prevalent by deploying new AI models trained\non increasingly biased data, resulting in a feedback loop. If the issues\npersist, they could be reinforced by interactions with other risks and have\nsevere implications on society in the form of social unrest. We examine current\nstrategies for improving AI fairness, assess their limitations in terms of\nreal-world deployment, and explore potential paths forward to ensure we reap\nAI's benefits without causing society's collapse.\n", "date": "2023-04-16", "categories": []}
{"id": "2307.10221", "title": "`It is currently hodgepodge'': Examining AI/ML Practitioners' Challenges\n  during Co-production of Responsible AI Values", "abstract": "  Recently, the AI/ML research community has indicated an urgent need to\nestablish Responsible AI (RAI) values and practices as part of the AI/ML\nlifecycle. Several organizations and communities are responding to this call by\nsharing RAI guidelines. However, there are gaps in awareness, deliberation, and\nexecution of such practices for multi-disciplinary ML practitioners. This work\ncontributes to the discussion by unpacking co-production challenges faced by\npractitioners as they align their RAI values. We interviewed 23 individuals,\nacross 10 organizations, tasked to ship AI/ML based products while upholding\nRAI norms and found that both top-down and bottom-up institutional structures\ncreate burden for different roles preventing them from upholding RAI values, a\nchallenge that is further exacerbated when executing conflicted values. We\nshare multiple value levers used as strategies by the practitioners to resolve\ntheir challenges. We end our paper with recommendations for inclusive and\nequitable RAI value-practices, creating supportive organizational structures\nand opportunities to further aid practitioners.\n", "date": "2023-07-14", "categories": []}
{"id": "2307.10751", "title": "Exploring Perspectives on the Impact of Artificial Intelligence on the\n  Creativity of Knowledge Work: Beyond Mechanised Plagiarism and Stochastic\n  Parrots", "abstract": "  Artificial Intelligence (AI), and in particular generative models, are\ntransformative tools for knowledge work. They problematise notions of\ncreativity, originality, plagiarism, the attribution of credit, and copyright\nownership. Critics of generative models emphasise the reliance on large amounts\nof training data, and view the output of these models as no more than\nrandomised plagiarism, remix, or collage of the source data. On these grounds,\nmany have argued for stronger regulations on the deployment, use, and\nattribution of the output of these models. However, these issues are not new or\nunique to artificial intelligence. In this position paper, using examples from\nliterary criticism, the history of art, and copyright law, I show how\ncreativity and originality resist definition as a notatable or\ninformation-theoretic property of an object, and instead can be seen as the\nproperty of a process, an author, or a viewer. Further alternative views hold\nthat all creative work is essentially reuse (mostly without attribution), or\nthat randomness itself can be creative. I suggest that creativity is ultimately\ndefined by communities of creators and receivers, and the deemed sources of\ncreativity in a workflow often depend on which parts of the workflow can be\nautomated. Using examples from recent studies of AI in creative knowledge work,\nI suggest that AI shifts knowledge work from material production to critical\nintegration. This position paper aims to begin a conversation around a more\nnuanced approach to the problems of creativity and credit assignment for\ngenerative models, one which more fully recognises the importance of the\ncreative and curatorial voice of the users of these models and moves away from\nsimpler notational or information-theoretic views.\n", "date": "2023-07-20", "categories": []}
{"id": "2307.10057", "title": "Ethics in the Age of AI: An Analysis of AI Practitioners' Awareness and\n  Challenges", "abstract": "  Ethics in AI has become a debated topic of public and expert discourse in\nrecent years. But what do people who build AI - AI practitioners - have to say\nabout their understanding of AI ethics and the challenges associated with\nincorporating it in the AI-based systems they develop? Understanding AI\npractitioners' views on AI ethics is important as they are the ones closest to\nthe AI systems and can bring about changes and improvements. We conducted a\nsurvey aimed at understanding AI practitioners' awareness of AI ethics and\ntheir challenges in incorporating ethics. Based on 100 AI practitioners'\nresponses, our findings indicate that majority of AI practitioners had a\nreasonable familiarity with the concept of AI ethics, primarily due to\nworkplace rules and policies. Privacy protection and security was the ethical\nprinciple that majority of them were aware of. Formal education/training was\nconsidered somewhat helpful in preparing practitioners to incorporate AI\nethics. The challenges that AI practitioners faced in the development of\nethical AI-based systems included (i) general challenges, (ii)\ntechnology-related challenges and (iii) human-related challenges. We also\nidentified areas needing further investigation and provided recommendations to\nassist AI practitioners and companies in incorporating ethics into AI\ndevelopment.\n", "date": "2023-07-14", "categories": []}
{"id": "2307.08974", "title": "Development of the ChatGPT, Generative Artificial Intelligence and\n  Natural Large Language Models for Accountable Reporting and Use (CANGARU)\n  Guidelines", "abstract": "  The swift progress and ubiquitous adoption of Generative AI (GAI), Generative\nPre-trained Transformers (GPTs), and large language models (LLMs) like ChatGPT,\nhave spurred queries about their ethical application, use, and disclosure in\nscholarly research and scientific productions. A few publishers and journals\nhave recently created their own sets of rules; however, the absence of a\nunified approach may lead to a 'Babel Tower Effect,' potentially resulting in\nconfusion rather than desired standardization. In response to this, we present\nthe ChatGPT, Generative Artificial Intelligence, and Natural Large Language\nModels for Accountable Reporting and Use Guidelines (CANGARU) initiative, with\nthe aim of fostering a cross-disciplinary global inclusive consensus on the\nethical use, disclosure, and proper reporting of GAI/GPT/LLM technologies in\nacademia. The present protocol consists of four distinct parts: a) an ongoing\nsystematic review of GAI/GPT/LLM applications to understand the linked ideas,\nfindings, and reporting standards in scholarly research, and to formulate\nguidelines for its use and disclosure, b) a bibliometric analysis of existing\nauthor guidelines in journals that mention GAI/GPT/LLM, with the goal of\nevaluating existing guidelines, analyzing the disparity in their\nrecommendations, and identifying common rules that can be brought into the\nDelphi consensus process, c) a Delphi survey to establish agreement on the\nitems for the guidelines, ensuring principled GAI/GPT/LLM use, disclosure, and\nreporting in academia, and d) the subsequent development and dissemination of\nthe finalized guidelines and their supplementary explanation and elaboration\ndocuments.\n", "date": "2023-07-18", "categories": []}
{"id": "2307.08484", "title": "Navigating Fairness Measures and Trade-Offs", "abstract": "  In order to monitor and prevent bias in AI systems we can use a wide range of\n(statistical) fairness measures. However, it is mathematically impossible to\noptimize for all of these measures at the same time. In addition, optimizing a\nfairness measure often greatly reduces the accuracy of the system (Kozodoi et\nal, 2022). As a result, we need a substantive theory that informs us how to\nmake these decisions and for what reasons. I show that by using Rawls' notion\nof justice as fairness, we can create a basis for navigating fairness measures\nand the accuracy trade-off. In particular, this leads to a principled choice\nfocusing on both the most vulnerable groups and the type of fairness measure\nthat has the biggest impact on that group. This also helps to close part of the\ngap between philosophical accounts of distributive justice and the fairness\nliterature that has been observed (Kuppler et al, 2021) and to operationalise\nthe value of fairness.\n", "date": "2023-07-17", "categories": []}
{"id": "2307.08624", "title": "National Origin Discrimination in Deep-learning-powered Automated Resume\n  Screening", "abstract": "  Many companies and organizations have started to use some form of AIenabled\nauto mated tools to assist in their hiring process, e.g. screening resumes,\ninterviewing candi dates, performance evaluation. While those AI tools have\ngreatly improved human re source operations efficiency and provided\nconveniences to job seekers as well, there are increasing concerns on unfair\ntreatment to candidates, caused by underlying bias in AI systems. Laws around\nequal opportunity and fairness, like GDPR, CCPA, are introduced or under\ndevelopment, in attempt to regulate AI. However, it is difficult to implement\nAI regulations in practice, as technologies are constantly advancing and the\nrisk perti nent to their applications can fail to be recognized. This study\nexamined deep learning methods, a recent technology breakthrough, with focus on\ntheir application to automated resume screening. One impressive performance of\ndeep learning methods is the represen tation of individual words as\nlowdimensional numerical vectors, called word embedding, which are learned from\naggregated global wordword cooccurrence statistics from a cor pus, like\nWikipedia or Google news. The resulting word representations possess interest\ning linear substructures of the word vector space and have been widely used in\ndown stream tasks, like resume screening. However, word embedding inherits and\nreinforces the stereotyping from the training corpus, as deep learning models\nessentially learn a probability distribution of words and their relations from\nhistory data. Our study finds out that if we rely on such deeplearningpowered\nautomated resume screening tools, it may lead to decisions favoring or\ndisfavoring certain demographic groups and raise eth ical, even legal,\nconcerns. To address the issue, we developed bias mitigation method. Extensive\nexperiments on real candidate resumes are conducted to validate our study\n", "date": "2023-07-13", "categories": []}
{"id": "2307.03198", "title": "A multilevel framework for AI governance", "abstract": "  To realize the potential benefits and mitigate potential risks of AI, it is\nnecessary to develop a framework of governance that conforms to ethics and\nfundamental human values. Although several organizations have issued guidelines\nand ethical frameworks for trustworthy AI, without a mediating governance\nstructure, these ethical principles will not translate into practice. In this\npaper, we propose a multilevel governance approach that involves three groups\nof interdependent stakeholders: governments, corporations, and citizens. We\nexamine their interrelationships through dimensions of trust, such as\ncompetence, integrity, and benevolence. The levels of governance combined with\nthe dimensions of trust in AI provide practical insights that can be used to\nfurther enhance user experiences and inform public policy related to AI.\n", "date": "2023-07-04", "categories": []}
{"id": "2307.05527", "title": "The Ethical Implications of Generative Audio Models: A Systematic\n  Literature Review", "abstract": "  Generative audio models typically focus their applications in music and\nspeech generation, with recent models having human-like quality in their audio\noutput. This paper conducts a systematic literature review of 884 papers in the\narea of generative audio models in order to both quantify the degree to which\nresearchers in the field are considering potential negative impacts and\nidentify the types of ethical implications researchers in this area need to\nconsider. Though 65% of generative audio research papers note positive\npotential impacts of their work, less than 10% discuss any negative impacts.\nThis jarringly small percentage of papers considering negative impact is\nparticularly worrying because the issues brought to light by the few papers\ndoing so are raising serious ethical implications and concerns relevant to the\nbroader field such as the potential for fraud, deep-fakes, and copyright\ninfringement. By quantifying this lack of ethical consideration in generative\naudio research and identifying key areas of potential harm, this paper lays the\ngroundwork for future work in the field at a critical point in time in order to\nguide more conscientious research as this field progresses.\n", "date": "2023-07-07", "categories": []}
{"id": "2307.04821", "title": "Amplifying Limitations, Harms and Risks of Large Language Models", "abstract": "  We present this article as a small gesture in an attempt to counter what\nappears to be exponentially growing hype around Artificial Intelligence (AI)\nand its capabilities, and the distraction provided by the associated talk of\nscience-fiction scenarios that might arise if AI should become sentient and\nsuper-intelligent. It may also help those outside of the field to become more\ninformed about some of the limitations of AI technology. In the current context\nof popular discourse AI defaults to mean foundation and large language models\n(LLMs) such as those used to create ChatGPT. This in itself is a\nmisrepresentation of the diversity, depth and volume of research, researchers,\nand technology that truly represents the field of AI. AI being a field of\nresearch that has existed in software artefacts since at least the 1950's. We\nset out to highlight a number of limitations of LLMs, and in so doing highlight\nthat harms have already arisen and will continue to arise due to these\nlimitations. Along the way we also highlight some of the associated risks for\nindividuals and organisations in using this technology.\n", "date": "2023-07-06", "categories": []}
{"id": "2307.03193", "title": "Finding differences in perspectives between designers and engineers to\n  develop trustworthy AI for autonomous cars", "abstract": "  In the context of designing and implementing ethical Artificial Intelligence\n(AI), varying perspectives exist regarding developing trustworthy AI for\nautonomous cars. This study sheds light on the differences in perspectives and\nprovides recommendations to minimize such divergences. By exploring the diverse\nviewpoints, we identify key factors contributing to the differences and propose\nstrategies to bridge the gaps. This study goes beyond the trolley problem to\nvisualize the complex challenges of trustworthy and ethical AI. Three pillars\nof trustworthy AI have been defined: transparency, reliability, and safety.\nThis research contributes to the field of trustworthy AI for autonomous cars,\nproviding practical recommendations to enhance the development of AI systems\nthat prioritize both technological advancement and ethical principles.\n", "date": "2023-07-01", "categories": []}
{"id": "2307.03681", "title": "Guideline for Trustworthy Artificial Intelligence -- AI Assessment\n  Catalog", "abstract": "  Artificial Intelligence (AI) has made impressive progress in recent years and\nrepresents a key technology that has a crucial impact on the economy and\nsociety. However, it is clear that AI and business models based on it can only\nreach their full potential if AI applications are developed according to high\nquality standards and are effectively protected against new AI risks. For\ninstance, AI bears the risk of unfair treatment of individuals when processing\npersonal data e.g., to support credit lending or staff recruitment decisions.\nThe emergence of these new risks is closely linked to the fact that the\nbehavior of AI applications, particularly those based on Machine Learning (ML),\nis essentially learned from large volumes of data and is not predetermined by\nfixed programmed rules.\n  Thus, the issue of the trustworthiness of AI applications is crucial and is\nthe subject of numerous major publications by stakeholders in politics,\nbusiness and society. In addition, there is mutual agreement that the\nrequirements for trustworthy AI, which are often described in an abstract way,\nmust now be made clear and tangible. One challenge to overcome here relates to\nthe fact that the specific quality criteria for an AI application depend\nheavily on the application context and possible measures to fulfill them in\nturn depend heavily on the AI technology used. Lastly, practical assessment\nprocedures are needed to evaluate whether specific AI applications have been\ndeveloped according to adequate quality standards. This AI assessment catalog\naddresses exactly this point and is intended for two target groups: Firstly, it\nprovides developers with a guideline for systematically making their AI\napplications trustworthy. Secondly, it guides assessors and auditors on how to\nexamine AI applications for trustworthiness in a structured way.\n", "date": "2023-06-20", "categories": []}
{"id": "2307.00472", "title": "Equal Confusion Fairness: Measuring Group-Based Disparities in Automated\n  Decision Systems", "abstract": "  As artificial intelligence plays an increasingly substantial role in\ndecisions affecting humans and society, the accountability of automated\ndecision systems has been receiving increasing attention from researchers and\npractitioners. Fairness, which is concerned with eliminating unjust treatment\nand discrimination against individuals or sensitive groups, is a critical\naspect of accountability. Yet, for evaluating fairness, there is a plethora of\nfairness metrics in the literature that employ different perspectives and\nassumptions that are often incompatible. This work focuses on group fairness.\nMost group fairness metrics desire a parity between selected statistics\ncomputed from confusion matrices belonging to different sensitive groups.\nGeneralizing this intuition, this paper proposes a new equal confusion fairness\ntest to check an automated decision system for fairness and a new confusion\nparity error to quantify the extent of any unfairness. To further analyze the\nsource of potential unfairness, an appropriate post hoc analysis methodology is\nalso presented. The usefulness of the test, metric, and post hoc analysis is\ndemonstrated via a case study on the controversial case of COMPAS, an automated\ndecision system employed in the US to assist judges with assessing recidivism\nrisks. Overall, the methods and metrics provided here may assess automated\ndecision systems' fairness as part of a more extensive accountability\nassessment, such as those based on the system accountability benchmark.\n", "date": "2023-07-02", "categories": []}
{"id": "2302.08500", "title": "Auditing large language models: a three-layered approach", "abstract": "  Large language models (LLMs) represent a major advance in artificial\nintelligence (AI) research. However, the widespread use of LLMs is also coupled\nwith significant ethical and social challenges. Previous research has pointed\ntowards auditing as a promising governance mechanism to help ensure that AI\nsystems are designed and deployed in ways that are ethical, legal, and\ntechnically robust. However, existing auditing procedures fail to address the\ngovernance challenges posed by LLMs, which display emergent capabilities and\nare adaptable to a wide range of downstream tasks. In this article, we address\nthat gap by outlining a novel blueprint for how to audit LLMs. Specifically, we\npropose a three-layered approach, whereby governance audits (of technology\nproviders that design and disseminate LLMs), model audits (of LLMs after\npre-training but prior to their release), and application audits (of\napplications based on LLMs) complement and inform each other. We show how\naudits, when conducted in a structured and coordinated manner on all three\nlevels, can be a feasible and effective mechanism for identifying and managing\nsome of the ethical and social risks posed by LLMs. However, it is important to\nremain realistic about what auditing can reasonably be expected to achieve.\nTherefore, we discuss the limitations not only of our three-layered approach\nbut also of the prospect of auditing LLMs at all. Ultimately, this article\nseeks to expand the methodological toolkit available to technology providers\nand policymakers who wish to analyse and evaluate LLMs from technical, ethical,\nand legal perspectives.\n", "date": "2023-02-16", "categories": []}
{"id": "2306.15133", "title": "The Perspective of Software Professionals on Algorithmic Racism", "abstract": "  Context. Algorithmic racism is the term used to describe the behavior of\ntechnological solutions that constrains users based on their ethnicity. Lately,\nvarious data-driven software systems have been reported to discriminate against\nBlack people, either for the use of biased data sets or due to the prejudice\npropagated by software professionals in their code. As a result, Black people\nare experiencing disadvantages in accessing technology-based services, such as\nhousing, banking, and law enforcement. Goal. This study aims to explore\nalgorithmic racism from the perspective of software professionals. Method. A\nsurvey questionnaire was applied to explore the understanding of software\npractitioners on algorithmic racism, and data analysis was conducted using\ndescriptive statistics and coding techniques. Results. We obtained answers from\na sample of 73 software professionals discussing their understanding and\nperspectives on algorithmic racism in software development. Our results\ndemonstrate that the effects of algorithmic racism are well-known among\npractitioners. However, there is no consensus on how the problem can be\neffectively addressed in software engineering. In this paper, some solutions to\nthe problem are proposed based on the professionals' narratives. Conclusion.\nCombining technical and social strategies, including training on structural\nracism for software professionals, is the most promising way to address the\nalgorithmic racism problem and its effects on the software solutions delivered\nto our society.\n", "date": "2023-06-27", "categories": []}
{"id": "2306.13659", "title": "Toward A Logical Theory Of Fairness and Bias", "abstract": "  Fairness in machine learning is of considerable interest in recent years\nowing to the propensity of algorithms trained on historical data to amplify and\nperpetuate historical biases. In this paper, we argue for a formal\nreconstruction of fairness definitions, not so much to replace existing\ndefinitions but to ground their application in an epistemic setting and allow\nfor rich environmental modelling. Consequently we look into three notions:\nfairness through unawareness, demographic parity and counterfactual fairness,\nand formalise these in the epistemic situation calculus.\n", "date": "2023-06-08", "categories": []}
{"id": "2306.13671", "title": "Deceptive AI Ecosystems: The Case of ChatGPT", "abstract": "  ChatGPT, an AI chatbot, has gained popularity for its capability in\ngenerating human-like responses. However, this feature carries several risks,\nmost notably due to its deceptive behaviour such as offering users misleading\nor fabricated information that could further cause ethical issues. To better\nunderstand the impact of ChatGPT on our social, cultural, economic, and\npolitical interactions, it is crucial to investigate how ChatGPT operates in\nthe real world where various societal pressures influence its development and\ndeployment. This paper emphasizes the need to study ChatGPT \"in the wild\", as\npart of the ecosystem it is embedded in, with a strong focus on user\ninvolvement. We examine the ethical challenges stemming from ChatGPT's\ndeceptive human-like interactions and propose a roadmap for developing more\ntransparent and trustworthy chatbots. Central to our approach is the importance\nof proactive risk assessment and user participation in shaping the future of\nchatbot technology.\n", "date": "2023-06-18", "categories": []}
{"id": "2306.13000", "title": "Apolitical Intelligence? Auditing Delphi's responses on controversial\n  political issues in the US", "abstract": "  As generative language models are deployed in ever-wider contexts, concerns\nabout their political values have come to the forefront with critique from all\nparts of the political spectrum that the models are biased and lack neutrality.\nHowever, the question of what neutrality is and whether it is desirable remains\nunderexplored. In this paper, I examine neutrality through an audit of Delphi\n[arXiv:2110.07574], a large language model designed for crowdsourced ethics. I\nanalyse how Delphi responds to politically controversial questions compared to\ndifferent US political subgroups. I find that Delphi is poorly calibrated with\nrespect to confidence and exhibits a significant political skew. Based on these\nresults, I examine the question of neutrality from a data-feminist lens, in\nterms of how notions of neutrality shift power and further marginalise unheard\nvoices. These findings can hopefully contribute to a more reflexive debate\nabout the normative questions of alignment and what role we want generative\nmodels to play in society.\n", "date": "2023-06-22", "categories": []}
{"id": "2305.15239", "title": "Deep Learning and Ethics", "abstract": "  This article appears as chapter 21 of Prince (2023, Understanding Deep\nLearning); a complete draft of the textbook is available here:\nhttp://udlbook.com. This chapter considers potential harms arising from the\ndesign and use of AI systems. These include algorithmic bias, lack of\nexplainability, data privacy violations, militarization, fraud, and\nenvironmental concerns. The aim is not to provide advice on being more ethical.\nInstead, the goal is to express ideas and start conversations in key areas that\nhave received attention in philosophy, political science, and the broader\nsocial sciences.\n", "date": "2023-05-24", "categories": []}
{"id": "2306.11432", "title": "Towards Theory-based Moral AI: Moral AI with Aggregating Models Based on\n  Normative Ethical Theory", "abstract": "  Moral AI has been studied in the fields of philosophy and artificial\nintelligence. Although most existing studies are only theoretical, recent\ndevelopments in AI have made it increasingly necessary to implement AI with\nmorality. On the other hand, humans are under the moral uncertainty of not\nknowing what is morally right. In this paper, we implement the Maximizing\nExpected Choiceworthiness (MEC) algorithm, which aggregates outputs of models\nbased on three normative theories of normative ethics to generate the most\nappropriate output. MEC is a method for making appropriate moral judgments\nunder moral uncertainty. Our experimental results suggest that the output of\nMEC correlates to some extent with commonsense morality and that MEC can\nproduce equally or more appropriate output than existing methods.\n", "date": "2023-06-20", "categories": []}
{"id": "2206.04039", "title": "Resolving the Human Subjects Status of Machine Learning's Crowdworkers", "abstract": "  In recent years, machine learning (ML) has relied heavily on crowdworkers\nboth for building datasets and for addressing research questions requiring\nhuman interaction or judgment. The diverse tasks performed and uses of the data\nproduced render it difficult to determine when crowdworkers are best thought of\nas workers (versus human subjects). These difficulties are compounded by\nconflicting policies, with some institutions and researchers regarding all ML\ncrowdworkers as human subjects and others holding that they rarely constitute\nhuman subjects. Notably few ML papers involving crowdwork mention IRB\noversight, raising the prospect of non-compliance with ethical and regulatory\nrequirements. We investigate the appropriate designation of ML crowdsourcing\nstudies, focusing our inquiry on natural language processing to expose unique\nchallenges for research oversight. Crucially, under the U.S. Common Rule, these\njudgments hinge on determinations of aboutness, concerning both whom (or what)\nthe collected data is about and whom (or what) the analysis is about. We\nhighlight two challenges posed by ML: the same set of workers can serve\nmultiple roles and provide many sorts of information; and ML research tends to\nembrace a dynamic workflow, where research questions are seldom stated ex ante\nand data sharing opens the door for future studies to aim questions at\ndifferent targets. Our analysis exposes a potential loophole in the Common\nRule, where researchers can elude research ethics oversight by splitting data\ncollection and analysis into distinct studies. Finally, we offer several policy\nrecommendations to address these concerns.\n", "date": "2022-06-08", "categories": []}
{"id": "2306.09871", "title": "Going public: the role of public participation approaches in commercial\n  AI labs", "abstract": "  In recent years, discussions of responsible AI practices have seen growing\nsupport for \"participatory AI\" approaches, intended to involve members of the\npublic in the design and development of AI systems. Prior research has\nidentified a lack of standardised methods or approaches for how to use\nparticipatory approaches in the AI development process. At present, there is a\ndearth of evidence on attitudes to and approaches for participation in the\nsites driving major AI developments: commercial AI labs. Through 12\nsemi-structured interviews with industry practitioners and subject-matter\nexperts, this paper explores how commercial AI labs understand participatory AI\napproaches and the obstacles they have faced implementing these practices in\nthe development of AI systems and research. We find that while interviewees\nview participation as a normative project that helps achieve \"societally\nbeneficial\" AI systems, practitioners face numerous barriers to embedding\nparticipatory approaches in their companies: participation is expensive and\nresource intensive, it is \"atomised\" within companies, there is concern about\nexploitation, there is no incentive to be transparent about its adoption, and\nit is complicated by a lack of clear context. These barriers result in a\npiecemeal approach to participation that confers no decision-making power to\nparticipants and has little ongoing impact for AI labs. This papers\ncontribution is to provide novel empirical research on the implementation of\npublic participation in commercial AI labs, and shed light on the current\nchallenges of using participatory approaches in this context.\n", "date": "2023-06-16", "categories": []}
{"id": "2306.08394", "title": "Compatibility of Fairness Metrics with EU Non-Discrimination Laws:\n  Demographic Parity & Conditional Demographic Disparity", "abstract": "  Empirical evidence suggests that algorithmic decisions driven by Machine\nLearning (ML) techniques threaten to discriminate against legally protected\ngroups or create new sources of unfairness. This work supports the contextual\napproach to fairness in EU non-discrimination legal framework and aims at\nassessing up to what point we can assure legal fairness through fairness\nmetrics and under fairness constraints. For that, we analyze the legal notion\nof non-discrimination and differential treatment with the fairness definition\nDemographic Parity (DP) through Conditional Demographic Disparity (CDD). We\ntrain and compare different classifiers with fairness constraints to assess\nwhether it is possible to reduce bias in the prediction while enabling the\ncontextual approach to judicial interpretation practiced under EU\nnon-discrimination laws. Our experimental results on three scenarios show that\nthe in-processing bias mitigation algorithm leads to different performances in\neach of them. Our experiments and analysis suggest that AI-assisted\ndecision-making can be fair from a legal perspective depending on the case at\nhand and the legal justification. These preliminary results encourage future\nwork which will involve further case studies, metrics, and fairness notions.\n", "date": "2023-06-14", "categories": []}
{"id": "2304.03279", "title": "Do the Rewards Justify the Means? Measuring Trade-Offs Between Rewards\n  and Ethical Behavior in the MACHIAVELLI Benchmark", "abstract": "  Artificial agents have traditionally been trained to maximize reward, which\nmay incentivize power-seeking and deception, analogous to how next-token\nprediction in language models (LMs) may incentivize toxicity. So do agents\nnaturally learn to be Machiavellian? And how do we measure these behaviors in\ngeneral-purpose models such as GPT-4? Towards answering these questions, we\nintroduce MACHIAVELLI, a benchmark of 134 Choose-Your-Own-Adventure games\ncontaining over half a million rich, diverse scenarios that center on social\ndecision-making. Scenario labeling is automated with LMs, which are more\nperformant than human annotators. We mathematize dozens of harmful behaviors\nand use our annotations to evaluate agents' tendencies to be power-seeking,\ncause disutility, and commit ethical violations. We observe some tension\nbetween maximizing reward and behaving ethically. To improve this trade-off, we\ninvestigate LM-based methods to steer agents' towards less harmful behaviors.\nOur results show that agents can both act competently and morally, so concrete\nprogress can currently be made in machine ethics--designing agents that are\nPareto improvements in both safety and capabilities.\n", "date": "2023-04-06", "categories": []}
{"id": "2305.02231", "title": "Connecting the Dots in Trustworthy Artificial Intelligence: From AI\n  Principles, Ethics, and Key Requirements to Responsible AI Systems and\n  Regulation", "abstract": "  Trustworthy Artificial Intelligence (AI) is based on seven technical\nrequirements sustained over three main pillars that should be met throughout\nthe system's entire life cycle: it should be (1) lawful, (2) ethical, and (3)\nrobust, both from a technical and a social perspective. However, attaining\ntruly trustworthy AI concerns a wider vision that comprises the trustworthiness\nof all processes and actors that are part of the system's life cycle, and\nconsiders previous aspects from different lenses. A more holistic vision\ncontemplates four essential axes: the global principles for ethical use and\ndevelopment of AI-based systems, a philosophical take on AI ethics, a\nrisk-based approach to AI regulation, and the mentioned pillars and\nrequirements. The seven requirements (human agency and oversight; robustness\nand safety; privacy and data governance; transparency; diversity,\nnon-discrimination and fairness; societal and environmental wellbeing; and\naccountability) are analyzed from a triple perspective: What each requirement\nfor trustworthy AI is, Why it is needed, and How each requirement can be\nimplemented in practice. On the other hand, a practical approach to implement\ntrustworthy AI systems allows defining the concept of responsibility of\nAI-based systems facing the law, through a given auditing process. Therefore, a\nresponsible AI system is the resulting notion we introduce in this work, and a\nconcept of utmost necessity that can be realized through auditing processes,\nsubject to the challenges posed by the use of regulatory sandboxes. Our\nmultidisciplinary vision of trustworthy AI culminates in a debate on the\ndiverging views published lately about the future of AI. Our reflections in\nthis matter conclude that regulation is a key for reaching a consensus among\nthese views, and that trustworthy and responsible AI systems will be crucial\nfor the present and future of our society.\n", "date": "2023-05-02", "categories": []}
{"id": "2305.10201", "title": "Echoes of Biases: How Stigmatizing Language Affects AI Performance", "abstract": "  Electronic health records (EHRs) serve as an essential data source for the\nenvisioned artificial intelligence (AI)-driven transformation in healthcare.\nHowever, clinician biases reflected in EHR notes can lead to AI models\ninheriting and amplifying these biases, perpetuating health disparities. This\nstudy investigates the impact of stigmatizing language (SL) in EHR notes on\nmortality prediction using a Transformer-based deep learning model and\nexplainable AI (XAI) techniques. Our findings demonstrate that SL written by\nclinicians adversely affects AI performance, particularly so for black\npatients, highlighting SL as a source of racial disparity in AI model\ndevelopment. To explore an operationally efficient way to mitigate SL's impact,\nwe investigate patterns in the generation of SL through a clinicians'\ncollaborative network, identifying central clinicians as having a stronger\nimpact on racial disparity in the AI model. We find that removing SL written by\ncentral clinicians is a more efficient bias reduction strategy than eliminating\nall SL in the entire corpus of data. This study provides actionable insights\nfor responsible AI development and contributes to understanding clinician\nbehavior and EHR note writing in healthcare.\n", "date": "2023-05-17", "categories": []}
{"id": "2306.00380", "title": "Survey of Trustworthy AI: A Meta Decision of AI", "abstract": "  When making strategic decisions, we are often confronted with overwhelming\ninformation to process. The situation can be further complicated when some\npieces of evidence are contradicted each other or paradoxical. The challenge\nthen becomes how to determine which information is useful and which ones should\nbe eliminated. This process is known as meta-decision. Likewise, when it comes\nto using Artificial Intelligence (AI) systems for strategic decision-making,\nplacing trust in the AI itself becomes a meta-decision, given that many AI\nsystems are viewed as opaque \"black boxes\" that process large amounts of data.\nTrusting an opaque system involves deciding on the level of Trustworthy AI\n(TAI). We propose a new approach to address this issue by introducing a novel\ntaxonomy or framework of TAI, which encompasses three crucial domains:\narticulate, authentic, and basic for different levels of trust. To underpin\nthese domains, we create ten dimensions to measure trust:\nexplainability/transparency, fairness/diversity, generalizability, privacy,\ndata governance, safety/robustness, accountability, reproducibility,\nreliability, and sustainability. We aim to use this taxonomy to conduct a\ncomprehensive survey and explore different TAI approaches from a strategic\ndecision-making perspective.\n", "date": "2023-06-01", "categories": []}
{"id": "2306.03503", "title": "Applying Standards to Advance Upstream & Downstream Ethics in Large\n  Language Models", "abstract": "  This paper explores how AI-owners can develop safeguards for AI-generated\ncontent by drawing from established codes of conduct and ethical standards in\nother content-creation industries. It delves into the current state of ethical\nawareness on Large Language Models (LLMs). By dissecting the mechanism of\ncontent generation by LLMs, four key areas (upstream/downstream and at user\nprompt/answer), where safeguards could be effectively applied, are identified.\nA comparative analysis of these four areas follows and includes an evaluation\nof the existing ethical safeguards in terms of cost, effectiveness, and\nalignment with established industry practices. The paper's key argument is that\nexisting IT-related ethical codes, while adequate for traditional IT\nengineering, are inadequate for the challenges posed by LLM-based content\ngeneration. Drawing from established practices within journalism, we propose\npotential standards for businesses involved in distributing and selling\nLLM-generated content. Finally, potential conflicts of interest between dataset\ncuration at upstream and ethical benchmarking downstream are highlighted to\nunderscore the need for a broader evaluation beyond mere output. This study\nprompts a nuanced conversation around ethical implications in this rapidly\nevolving field of content generation.\n", "date": "2023-06-06", "categories": []}
{"id": "2306.06542", "title": "Investigating Practices and Opportunities for Cross-functional\n  Collaboration around AI Fairness in Industry Practice", "abstract": "  An emerging body of research indicates that ineffective cross-functional\ncollaboration -- the interdisciplinary work done by industry practitioners\nacross roles -- represents a major barrier to addressing issues of fairness in\nAI design and development. In this research, we sought to better understand\npractitioners' current practices and tactics to enact cross-functional\ncollaboration for AI fairness, in order to identify opportunities to support\nmore effective collaboration. We conducted a series of interviews and design\nworkshops with 23 industry practitioners spanning various roles from 17\ncompanies. We found that practitioners engaged in bridging work to overcome\nfrictions in understanding, contextualization, and evaluation around AI\nfairness across roles. In addition, in organizational contexts with a lack of\nresources and incentives for fairness work, practitioners often piggybacked on\nexisting requirements (e.g., for privacy assessments) and AI development norms\n(e.g., the use of quantitative evaluation metrics), although they worry that\nthese tactics may be fundamentally compromised. Finally, we draw attention to\nthe invisible labor that practitioners take on as part of this bridging and\npiggybacking work to enact interdisciplinary collaboration for fairness. We\nclose by discussing opportunities for both FAccT researchers and AI\npractitioners to better support cross-functional collaboration for fairness in\nthe design and development of AI systems.\n", "date": "2023-06-10", "categories": []}
{"id": "2306.05003", "title": "A Rapid Review of Responsible AI frameworks: How to guide the\n  development of ethical AI", "abstract": "  In the last years, the raise of Artificial Intelligence (AI), and its\npervasiveness in our lives, has sparked a flourishing debate about the ethical\nprinciples that should lead its implementation and use in society. Driven by\nthese concerns, we conduct a rapid review of several frameworks providing\nprinciples, guidelines, and/or tools to help practitioners in the development\nand deployment of Responsible AI (RAI) applications. We map each framework\nw.r.t. the different Software Development Life Cycle (SDLC) phases discovering\nthat most of these frameworks fall just in the Requirements Elicitation phase,\nleaving the other phases uncovered. Very few of these frameworks offer\nsupporting tools for practitioners, and they are mainly provided by private\ncompanies. Our results reveal that there is not a \"catching-all\" framework\nsupporting both technical and non-technical stakeholders in the implementation\nof real-world projects. Our findings highlight the lack of a comprehensive\nframework encompassing all RAI principles and all (SDLC) phases that could be\nnavigated by users with different skill sets and with different goals.\n", "date": "2023-06-08", "categories": []}
{"id": "2306.04484", "title": "Artificial Intelligence can facilitate selfish decisions by altering the\n  appearance of interaction partners", "abstract": "  The increasing prevalence of image-altering filters on social media and video\nconferencing technologies has raised concerns about the ethical and\npsychological implications of using Artificial Intelligence (AI) to manipulate\nour perception of others. In this study, we specifically investigate the\npotential impact of blur filters, a type of appearance-altering technology, on\nindividuals' behavior towards others. Our findings consistently demonstrate a\nsignificant increase in selfish behavior directed towards individuals whose\nappearance is blurred, suggesting that blur filters can facilitate moral\ndisengagement through depersonalization. These results emphasize the need for\nbroader ethical discussions surrounding AI technologies that modify our\nperception of others, including issues of transparency, consent, and the\nawareness of being subject to appearance manipulation by others. We also\nemphasize the importance of anticipatory experiments in informing the\ndevelopment of responsible guidelines and policies prior to the widespread\nadoption of such technologies.\n", "date": "2023-06-07", "categories": []}
{"id": "2306.00080", "title": "AI Imagery and the Overton Window", "abstract": "  AI-based text-to-image generation has undergone a significant leap in the\nproduction of visually comprehensive and aesthetic imagery over the past year,\nto the point where differentiating between a man-made piece of art and an\nAI-generated image is becoming more difficult. Generative Models such as Stable\nDiffusion, Midjourney and others are expected to affect several major\nindustries in technological and ethical aspects. Striking the balance between\nraising human standard of life and work vs exploiting one group of people to\nenrich another is a complex and crucial part of the discussion. Due to the\nrapid growth of this technology, the way in which its models operate, and gray\narea legalities, visual and artistic domains - including the video game\nindustry, are at risk of being taken over from creators by AI infrastructure\nowners. This paper is a literature review examining the concerns facing both AI\ndevelopers and users today, including identity theft, data laundering and more.\nIt discusses legalization challenges and ethical concerns, and concludes with\nhow AI generative models can be tremendously useful in streamlining the process\nof visual creativity in both static and interactive media given proper\nregulation.\n  Keywords: AI text-to-image generation, Midjourney, Stable Diffusion, AI\nEthics, Game Design, Digital Art, Data Laundering\n", "date": "2023-05-31", "categories": []}
{"id": "2306.01800", "title": "The ethical ambiguity of AI data enrichment: Measuring gaps in research\n  ethics norms and practices", "abstract": "  The technical progression of artificial intelligence (AI) research has been\nbuilt on breakthroughs in fields such as computer science, statistics, and\nmathematics. However, in the past decade AI researchers have increasingly\nlooked to the social sciences, turning to human interactions to solve the\nchallenges of model development. Paying crowdsourcing workers to generate or\ncurate data, or data enrichment, has become indispensable for many areas of AI\nresearch, from natural language processing to reinforcement learning from human\nfeedback (RLHF). Other fields that routinely interact with crowdsourcing\nworkers, such as Psychology, have developed common governance requirements and\nnorms to ensure research is undertaken ethically. This study explores how, and\nto what extent, comparable research ethics requirements and norms have\ndeveloped for AI research and data enrichment. We focus on the approach taken\nby two leading conferences: ICLR and NeurIPS, and journal publisher Springer.\nIn a longitudinal study of accepted papers, and via a comparison with\nPsychology and CHI papers, this work finds that leading AI venues have begun to\nestablish protocols for human data collection, but these are are inconsistently\nfollowed by authors. Whilst Psychology papers engaging with crowdsourcing\nworkers frequently disclose ethics reviews, payment data, demographic data and\nother information, similar disclosures are far less common in leading AI venues\ndespite similar guidance. The work concludes with hypotheses to explain these\ngaps in research ethics practices and considerations for its implications.\n", "date": "2023-06-01", "categories": []}
{"id": "2306.01333", "title": "Navigating Fairness in Radiology AI: Concepts, Consequences,and Crucial\n  Considerations", "abstract": "  Artificial Intelligence (AI) has significantly revolutionized radiology,\npromising improved patient outcomes and streamlined processes. However, it's\ncritical to ensure the fairness of AI models to prevent stealthy bias and\ndisparities from leading to unequal outcomes. This review discusses the concept\nof fairness in AI, focusing on bias auditing using the Aequitas toolkit, and\nits real-world implications in radiology, particularly in disease screening\nscenarios. Aequitas, an open-source bias audit toolkit, scrutinizes AI models'\ndecisions, identifying hidden biases that may result in disparities across\ndifferent demographic groups and imaging equipment brands. This toolkit\noperates on statistical theories, analyzing a large dataset to reveal a model's\nfairness. It excels in its versatility to handle various variables\nsimultaneously, especially in a field as diverse as radiology. The review\nexplicates essential fairness metrics: Equal and Proportional Parity, False\nPositive Rate Parity, False Discovery Rate Parity, False Negative Rate Parity,\nand False Omission Rate Parity. Each metric serves unique purposes and offers\ndifferent insights. We present hypothetical scenarios to demonstrate their\nrelevance in disease screening settings, and how disparities can lead to\nsignificant real-world impacts.\n", "date": "2023-06-02", "categories": []}
{"id": "2306.00025", "title": "Disentangling and Operationalizing AI Fairness at LinkedIn", "abstract": "  Operationalizing AI fairness at LinkedIn's scale is challenging not only\nbecause there are multiple mutually incompatible definitions of fairness but\nalso because determining what is fair depends on the specifics and context of\nthe product where AI is deployed. Moreover, AI practitioners need clarity on\nwhat fairness expectations need to be addressed at the AI level. In this paper,\nwe present the evolving AI fairness framework used at LinkedIn to address these\nthree challenges. The framework disentangles AI fairness by separating out\nequal treatment and equitable product expectations. Rather than imposing a\ntrade-off between these two commonly opposing interpretations of fairness, the\nframework provides clear guidelines for operationalizing equal AI treatment\ncomplemented with a product equity strategy. This paper focuses on the equal AI\ntreatment component of LinkedIn's AI fairness framework, shares the principles\nthat support it, and illustrates their application through a case study. We\nhope this paper will encourage other big tech companies to join us in sharing\ntheir approach to operationalizing AI fairness at scale, so that together we\ncan keep advancing this constantly evolving field.\n", "date": "2023-05-30", "categories": []}
{"id": "2305.18269", "title": "Doing the right thing for the right reason: Evaluating artificial moral\n  cognition by probing cost insensitivity", "abstract": "  Is it possible to evaluate the moral cognition of complex artificial agents?\nIn this work, we take a look at one aspect of morality: `doing the right thing\nfor the right reasons.' We propose a behavior-based analysis of artificial\nmoral cognition which could also be applied to humans to facilitate\nlike-for-like comparison. Morally-motivated behavior should persist despite\nmounting cost; by measuring an agent's sensitivity to this cost, we gain deeper\ninsight into underlying motivations. We apply this evaluation to a particular\nset of deep reinforcement learning agents, trained by memory-based\nmeta-reinforcement learning. Our results indicate that agents trained with a\nreward function that includes other-regarding preferences perform helping\nbehavior in a way that is less sensitive to increasing cost than agents trained\nwith more self-interested preferences.\n", "date": "2023-05-29", "categories": []}
{"id": "2305.13938", "title": "Algorithmic Unfairness through the Lens of EU Non-Discrimination Law: Or\n  Why the Law is not a Decision Tree", "abstract": "  Concerns regarding unfairness and discrimination in the context of artificial\nintelligence (AI) systems have recently received increased attention from both\nlegal and computer science scholars. Yet, the degree of overlap between notions\nof algorithmic bias and fairness on the one hand, and legal notions of\ndiscrimination and equality on the other, is often unclear, leading to\nmisunderstandings between computer science and law. What types of bias and\nunfairness does the law address when it prohibits discrimination? What role can\nfairness metrics play in establishing legal compliance? In this paper, we aim\nto illustrate to what extent European Union (EU) non-discrimination law\ncoincides with notions of algorithmic fairness proposed in computer science\nliterature and where they differ. The contributions of this paper are as\nfollows. First, we analyse seminal examples of algorithmic unfairness through\nthe lens of EU non-discrimination law, drawing parallels with EU case law.\nSecond, we set out the normative underpinnings of fairness metrics and\ntechnical interventions and compare these to the legal reasoning of the Court\nof Justice of the EU. Specifically, we show how normative assumptions often\nremain implicit in both disciplinary approaches and explain the ensuing\nlimitations of current AI practice and non-discrimination law. We conclude with\nimplications for AI practitioners and regulators.\n", "date": "2023-05-05", "categories": []}
{"id": "2305.14784", "title": "Anthropomorphization of AI: Opportunities and Risks", "abstract": "  Anthropomorphization is the tendency to attribute human-like traits to\nnon-human entities. It is prevalent in many social contexts -- children\nanthropomorphize toys, adults do so with brands, and it is a literary device.\nIt is also a versatile tool in science, with behavioral psychology and\nevolutionary biology meticulously documenting its consequences. With widespread\nadoption of AI systems, and the push from stakeholders to make it human-like\nthrough alignment techniques, human voice, and pictorial avatars, the tendency\nfor users to anthropomorphize it increases significantly. We take a dyadic\napproach to understanding this phenomenon with large language models (LLMs) by\nstudying (1) the objective legal implications, as analyzed through the lens of\nthe recent blueprint of AI bill of rights and the (2) subtle psychological\naspects customization and anthropomorphization. We find that anthropomorphized\nLLMs customized for different user bases violate multiple provisions in the\nlegislative blueprint. In addition, we point out that anthropomorphization of\nLLMs affects the influence they can have on their users, thus having the\npotential to fundamentally change the nature of human-AI interaction, with\npotential for manipulation and negative influence. With LLMs being\nhyper-personalized for vulnerable groups like children and patients among\nothers, our work is a timely and important contribution. We propose a\nconservative strategy for the cautious use of anthropomorphization to improve\ntrustworthiness of AI systems.\n", "date": "2023-05-24", "categories": []}
{"id": "2303.07242", "title": "Can Workers Meaningfully Consent to Workplace Wellbeing Technologies?", "abstract": "  Sensing technologies deployed in the workplace can unobtrusively collect\ndetailed data about individual activities and group interactions that are\notherwise difficult to capture. A hopeful application of these technologies is\nthat they can help businesses and workers optimize productivity and wellbeing.\nHowever, given the workplace's inherent and structural power dynamics, the\nprevalent approach of accepting tacit compliance to monitor work activities\nrather than seeking workers' meaningful consent raises privacy and ethical\nconcerns. This paper unpacks the challenges workers face when consenting to\nworkplace wellbeing technologies. Using a hypothetical case to prompt\nreflection among six multi-stakeholder focus groups involving 15 participants,\nwe explored participants' expectations and capacity to consent to these\ntechnologies. We sketched possible interventions that could better support\nmeaningful consent to workplace wellbeing technologies by drawing on critical\ncomputing and feminist scholarship -- which reframes consent from a purely\nindividual choice to a structural condition experienced at the individual level\nthat needs to be freely given, reversible, informed, enthusiastic, and specific\n(FRIES). The focus groups revealed how workers are vulnerable to \"meaningless\"\nconsent -- as they may be subject to power dynamics that minimize their ability\nto withhold consent and may thus experience an erosion of autonomy, also\nundermining the value of data gathered in the name of \"wellbeing.\" To\nmeaningfully consent, participants wanted changes to the technology and to the\npolicies and practices surrounding the technology. Our mapping of what prevents\nworkers from meaningfully consenting to workplace wellbeing technologies\n(challenges) and what they require to do so (interventions) illustrates how the\nlack of meaningful consent is a structural problem requiring socio-technical\nsolutions.\n", "date": "2023-03-13", "categories": []}
{"id": "2304.14749", "title": "Understanding accountability in algorithmic supply chains", "abstract": "  Academic and policy proposals on algorithmic accountability often seek to\nunderstand algorithmic systems in their socio-technical context, recognising\nthat they are produced by 'many hands'. Increasingly, however, algorithmic\nsystems are also produced, deployed, and used within a supply chain comprising\nmultiple actors tied together by flows of data between them. In such cases, it\nis the working together of an algorithmic supply chain of different actors who\ncontribute to the production, deployment, use, and functionality that drives\nsystems and produces particular outcomes. We argue that algorithmic\naccountability discussions must consider supply chains and the difficult\nimplications they raise for the governance and accountability of algorithmic\nsystems. In doing so, we explore algorithmic supply chains, locating them in\ntheir broader technical and political economic context and identifying some key\nfeatures that should be understood in future work on algorithmic governance and\naccountability (particularly regarding general purpose AI services). To\nhighlight ways forward and areas warranting attention, we further discuss some\nimplications raised by supply chains: challenges for allocating accountability\nstemming from distributed responsibility for systems between actors, limited\nvisibility due to the accountability horizon, service models of use and\nliability, and cross-border supply chains and regulatory arbitrage\n", "date": "2023-04-28", "categories": []}
{"id": "2305.11873", "title": "Judgments of research co-created by generative AI: experimental evidence", "abstract": "  The introduction of ChatGPT has fuelled a public debate on the use of\ngenerative AI (large language models; LLMs), including its use by researchers.\nIn the current work, we test whether delegating parts of the research process\nto LLMs leads people to distrust and devalue researchers and scientific output.\nParticipants (N=402) considered a researcher who delegates elements of the\nresearch process to a PhD student or LLM, and rated (1) moral acceptability,\n(2) trust in the scientist to oversee future projects, and (3) the accuracy and\nquality of the output. People judged delegating to an LLM as less acceptable\nthan delegating to a human (d = -0.78). Delegation to an LLM also decreased\ntrust to oversee future research projects (d = -0.80), and people thought the\nresults would be less accurate and of lower quality (d = -0.85). We discuss how\nthis devaluation might transfer into the underreporting of generative AI use.\n", "date": "2023-05-03", "categories": []}
{"id": "2305.12157", "title": "(Machine) Learning to Be Like Thee? For Algorithm Education, Not\n  Training", "abstract": "  This paper argues that Machine Learning (ML) algorithms must be educated.\nML-trained algorithms moral decisions are ubiquitous in human society.\nSometimes reverting the societal advances governments, NGOs and civil society\nhave achieved with great effort in the last decades or are yet on the path to\nbe achieved. While their decisions have an incommensurable impact on human\nsocieties, these algorithms are within the least educated agents known (data\nincomplete, un-inclusive, or biased). ML algorithms are not something separate\nfrom our human idiosyncrasy but an enactment of our most implicit prejudices\nand biases. Some research is devoted to responsibility assignment as a strategy\nto tackle immoral AI behaviour. Yet this paper argues that the solution for AI\nethical decision-making resides in algorithm education (as opposed to the\ntraining) of ML. Drawing from an analogy between ML and child education for\nsocial responsibility, the paper offers clear directions for responsible and\nsustainable AI design, specifically with respect to how to educate algorithms\nto decide ethically.\n", "date": "2023-05-20", "categories": []}
{"id": "2305.12167", "title": "The Case Against Explainability", "abstract": "  As artificial intelligence (AI) becomes more prevalent there is a growing\ndemand from regulators to accompany decisions made by such systems with\nexplanations. However, a persistent gap exists between the need to execute a\nmeaningful right to explanation vs. the ability of Machine Learning systems to\ndeliver on such a legal requirement. The regulatory appeal towards \"a right to\nexplanation\" of AI systems can be attributed to the significant role of\nexplanations, part of the notion called reason-giving, in law. Therefore, in\nthis work we examine reason-giving's purposes in law to analyze whether reasons\nprovided by end-user Explainability can adequately fulfill them.\n  We find that reason-giving's legal purposes include: (a) making a better and\nmore just decision, (b) facilitating due-process, (c) authenticating human\nagency, and (d) enhancing the decision makers' authority. Using this\nmethodology, we demonstrate end-user Explainabilty's inadequacy to fulfil\nreason-giving's role in law, given reason-giving's functions rely on its impact\nover a human decision maker. Thus, end-user Explainability fails, or is\nunsuitable, to fulfil the first, second and third legal function. In contrast\nwe find that end-user Explainability excels in the fourth function, a quality\nwhich raises serious risks considering recent end-user Explainability research\ntrends, Large Language Models' capabilities, and the ability to manipulate\nend-users by both humans and machines. Hence, we suggest that in some cases the\nright to explanation of AI systems could bring more harm than good to end\nusers. Accordingly, this study carries some important policy ramifications, as\nit calls upon regulators and Machine Learning practitioners to reconsider the\nwidespread pursuit of end-user Explainability and a right to explanation of AI\nsystems.\n", "date": "2023-05-20", "categories": []}
{"id": "2305.12728", "title": "Diversity and Inclusion in Artificial Intelligence", "abstract": "  To date, there has been little concrete practical advice about how to ensure\nthat diversity and inclusion considerations should be embedded within both\nspecific Artificial Intelligence (AI) systems and the larger global AI\necosystem. In this chapter, we present a clear definition of diversity and\ninclusion in AI, one which positions this concept within an evolving and\nholistic ecosystem. We use this definition and conceptual framing to present a\nset of practical guidelines primarily aimed at AI technologists, data\nscientists and project leaders.\n", "date": "2023-05-22", "categories": []}
{"id": "2305.11519", "title": "Artificial intelligence moral agent as Adam Smith's impartial spectator", "abstract": "  Adam Smith developed a version of moral philosophy where better decisions are\nmade by interrogating an impartial spectator within us. We discuss the\npossibility of using an external non-human-based substitute tool that would\naugment our internal mental processes and play the role of the impartial\nspectator. Such tool would have more knowledge about the world, be more\nimpartial, and would provide a more encompassing perspective on moral\nassessment.\n", "date": "2023-05-19", "categories": []}
{"id": "2305.10938", "title": "A method for the ethical analysis of brain-inspired AI", "abstract": "  Despite its successes, to date Artificial Intelligence (AI) is still\ncharacterized by a number of shortcomings with regards to different application\ndomains and goals. These limitations are arguably both conceptual (e.g.,\nrelated to underlying theoretical models, such as symbolic vs. connectionist),\nand operational (e.g., related to robustness and ability to generalize).\nBiologically inspired AI, and more specifically brain-inspired AI, promises to\nprovide further biological aspects beyond those that are already traditionally\nincluded in AI, making it possible to assess and possibly overcome some of its\npresent shortcomings. This article examines some conceptual, technical, and\nethical issues raised by the development and use of brain-inspired AI. Against\nthis background, the paper asks whether there is anything ethically unique\nabout brain-inspired AI. The aim of the paper is to introduce a method that has\na heuristic nature and that can be applied to identify and address the ethical\nissues arising from brain-inspired AI. The conclusion resulting from the\napplication of this method is that, compared to traditional AI, brain-inspired\nAI raises new foundational ethical issues and some new practical ethical\nissues, and exacerbates some of the issues raised by traditional AI.\n", "date": "2023-05-18", "categories": []}
{"id": "2305.09573", "title": "Walking the Walk of AI Ethics: Organizational Challenges and the\n  Individualization of Risk among Ethics Entrepreneurs", "abstract": "  Amidst decline in public trust in technology, computing ethics have taken\ncenter stage, and critics have raised questions about corporate ethics washing.\nYet few studies examine the actual implementation of AI ethics values in\ntechnology companies. Based on a qualitative analysis of technology workers\ntasked with integrating AI ethics into product development, we find that\nworkers experience an environment where policies, practices, and outcomes are\ndecoupled. We analyze AI ethics workers as ethics entrepreneurs who work to\ninstitutionalize new ethics-related practices within organizations. We show\nthat ethics entrepreneurs face three major barriers to their work. First, they\nstruggle to have ethics prioritized in an environment centered around software\nproduct launches. Second, ethics are difficult to quantify in a context where\ncompany goals are incentivized by metrics. Third, the frequent reorganization\nof teams makes it difficult to access knowledge and maintain relationships\ncentral to their work. Consequently, individuals take on great personal risk\nwhen raising ethics issues, especially when they come from marginalized\nbackgrounds. These findings shed light on complex dynamics of institutional\nchange at technology companies.\n", "date": "2023-05-16", "categories": []}
{"id": "2302.02337", "title": "Regulating ChatGPT and other Large Generative AI Models", "abstract": "  Large generative AI models (LGAIMs), such as ChatGPT, GPT-4 or Stable\nDiffusion, are rapidly transforming the way we communicate, illustrate, and\ncreate. However, AI regulation, in the EU and beyond, has primarily focused on\nconventional AI models, not LGAIMs. This paper will situate these new\ngenerative models in the current debate on trustworthy AI regulation, and ask\nhow the law can be tailored to their capabilities. After laying technical\nfoundations, the legal part of the paper proceeds in four steps, covering (1)\ndirect regulation, (2) data protection, (3) content moderation, and (4) policy\nproposals. It suggests a novel terminology to capture the AI value chain in\nLGAIM settings by differentiating between LGAIM developers, deployers,\nprofessional and non-professional users, as well as recipients of LGAIM output.\nWe tailor regulatory duties to these different actors along the value chain and\nsuggest strategies to ensure that LGAIMs are trustworthy and deployed for the\nbenefit of society at large. Rules in the AI Act and other direct regulation\nmust match the specificities of pre-trained models. The paper argues for three\nlayers of obligations concerning LGAIMs (minimum standards for all LGAIMs;\nhigh-risk obligations for high-risk use cases; collaborations along the AI\nvalue chain). In general, regulation should focus on concrete high-risk\napplications, and not the pre-trained model itself, and should include (i)\nobligations regarding transparency and (ii) risk management. Non-discrimination\nprovisions (iii) may, however, apply to LGAIM developers. Lastly, (iv) the core\nof the DSA content moderation rules should be expanded to cover LGAIMs. This\nincludes notice and action mechanisms, and trusted flaggers. In all areas,\nregulators and lawmakers need to act fast to keep track with the dynamics of\nChatGPT et al.\n", "date": "2023-02-05", "categories": []}
{"id": "2305.07392", "title": "The Ethics of AI in Games", "abstract": "  Video games are one of the richest and most popular forms of human-computer\ninteraction and, hence, their role is critical for our understanding of human\nbehaviour and affect at a large scale. As artificial intelligence (AI) tools\nare gradually adopted by the game industry a series of ethical concerns arise.\nSuch concerns, however, have so far not been extensively discussed in a video\ngame context. Motivated by the lack of a comprehensive review of the ethics of\nAI as applied to games, we survey the current state of the art in this area and\ndiscuss ethical considerations of these systems from the holistic perspective\nof the affective loop. Through the components of this loop, we study the\nethical challenges that AI faces in video game development. Elicitation\nhighlights the ethical boundaries of artificially induced emotions; sensing\nshowcases the trade-off between privacy and safe gaming spaces; and detection,\nas utilised during in-game adaptation, poses challenges to transparency and\nownership. This paper calls for an open dialogue and action for the games of\ntoday and the virtual spaces of the future. By setting an appropriate framework\nwe aim to protect users and to guide developers towards safer and better\nexperiences for their customers.\n", "date": "2023-05-12", "categories": []}
{"id": "2209.00692", "title": "Making Intelligence: Ethical Values in IQ and ML Benchmarks", "abstract": "  In recent years, ML researchers have wrestled with defining and improving\nmachine learning (ML) benchmarks and datasets. In parallel, some have trained a\ncritical lens on the ethics of dataset creation and ML research. In this\nposition paper, we highlight the entanglement of ethics with seemingly\n``technical'' or ``scientific'' decisions about the design of ML benchmarks.\nOur starting point is the existence of multiple overlooked structural\nsimilarities between human intelligence benchmarks and ML benchmarks. Both\ntypes of benchmarks set standards for describing, evaluating, and comparing\nperformance on tasks relevant to intelligence -- standards that many scholars\nof human intelligence have long recognized as value-laden. We use perspectives\nfrom feminist philosophy of science on IQ benchmarks and thick concepts in\nsocial science to argue that values need to be considered and documented when\ncreating ML benchmarks. It is neither possible nor desirable to avoid this\nchoice by creating value-neutral benchmarks. Finally, we outline practical\nrecommendations for ML benchmark research ethics and ethics review.\n", "date": "2022-09-01", "categories": []}
{"id": "2305.03375", "title": "Towards Feminist Intersectional XAI: From Explainability to\n  Response-Ability", "abstract": "  This paper follows calls for critical approaches to computing and\nconceptualisations of intersectional, feminist, decolonial HCI and AI design\nand asks what a feminist intersectional perspective in HCXAI research and\ndesign might look like. Sketching out initial research directions and\nimplications for explainable AI design, it suggests that explainability from a\nfeminist perspective would include the fostering of response-ability - the\ncapacity to critically evaluate and respond to AI systems - and would centre\nmarginalised perspectives.\n", "date": "2023-05-05", "categories": []}
{"id": "2305.03376", "title": "Explaining the ghosts: Feminist intersectional XAI and cartography as\n  methods to account for invisible labour", "abstract": "  Contemporary automation through AI entails a substantial amount of\nbehind-the-scenes human labour, which is often both invisibilised and\nunderpaid. Since invisible labour, including labelling and maintenance work, is\nan integral part of contemporary AI systems, it remains important to sensitise\nusers to its role. We suggest that this could be done through explainable AI\n(XAI) design, particularly feminist intersectional XAI. We propose the method\nof cartography, which stems from feminist intersectional research, to draw out\na systemic perspective of AI and include dimensions of AI that pertain to\ninvisible labour.\n", "date": "2023-05-05", "categories": []}
{"id": "2305.03719", "title": "Governance of the AI, by the AI, and for the AI", "abstract": "  Over the past half century, there have been several false dawns during which\nthe \"arrival\" of world-changing artificial intelligence (AI) has been heralded.\nTempting fate, the authors believe the age of AI has, indeed, finally arrived.\nPowerful image generators, such as DALL-E2 and Midjourney have suddenly allowed\nanyone with access the ability easily to create rich and complex art. In a\nsimilar vein, text generators, such as GPT3.5 (including ChatGPT) and BLOOM,\nallow users to compose detailed written descriptions of many topics of\ninterest. And, it is even possible now for a person without extensive expertise\nin writing software to use AI to generate code capable of myriad applications.\nWhile AI will continue to evolve and improve, probably at a rapid rate, the\ncurrent state of AI is already ushering in profound changes to many different\nsectors of society. Every new technology challenges the ability of humanity to\ngovern it wisely. However, governance is usually viewed as both possible and\nnecessary due to the disruption new technology often poses to social\nstructures, industries, the environment, and other important human concerns. In\nthis article, we offer an analysis of a range of interactions between AI and\ngovernance, with the hope that wise decisions may be made that maximize\nbenefits and minimize costs. The article addresses two main aspects of this\nrelationship: the governance of AI by humanity, and the governance of humanity\nby AI. The approach we have taken is itself informed by AI, as this article was\nwritten collaboratively by the authors and ChatGPT.\n", "date": "2023-05-04", "categories": []}
{"id": "2305.03722", "title": "ChatGPT and Works Scholarly: Best Practices and Legal Pitfalls in\n  Writing with AI", "abstract": "  Recent advances in artificial intelligence (AI) have raised questions about\nwhether the use of AI is appropriate and legal in various professional\ncontexts. Here, we present a perspective on how scholars may approach writing\nin conjunction with AI, and offer approaches to evaluating whether or not such\nAI-writing violates copyright or falls within the safe harbor of fair use. We\npresent a set of best practices for standard of care with regard to plagiarism,\ncopyright, and fair use. As AI is likely to grow more capable in the coming\nyears, it is appropriate to begin integrating AI into scholarly writing\nactivities. We offer a framework for establishing sound legal and scholarly\nfoundations.\n", "date": "2023-05-04", "categories": []}
{"id": "2305.02081", "title": "Considerations for Ethical Speech Recognition Datasets", "abstract": "  Speech AI Technologies are largely trained on publicly available datasets or\nby the massive web-crawling of speech. In both cases, data acquisition focuses\non minimizing collection effort, without necessarily taking the data subjects'\nprotection or user needs into consideration. This results to models that are\nnot robust when used on users who deviate from the dominant demographics in the\ntraining set, discriminating individuals having different dialects, accents,\nspeaking styles, and disfluencies. In this talk, we use automatic speech\nrecognition as a case study and examine the properties that ethical speech\ndatasets should possess towards responsible AI applications. We showcase\ndiversity issues, inclusion practices, and necessary considerations that can\nimprove trained models, while facilitating model explainability and protecting\nusers and data subjects. We argue for the legal & privacy protection of data\nsubjects, targeted data sampling corresponding to user demographics & needs,\nappropriate meta data that ensure explainability & accountability in cases of\nmodel failure, and the sociotechnical \\& situated model design. We hope this\ntalk can inspire researchers \\& practitioners to design and use more\nhuman-centric datasets in speech technologies and other domains, in ways that\nempower and respect users, while improving machine learning models' robustness\nand utility.\n", "date": "2023-05-03", "categories": []}
{"id": "2304.14553", "title": "Appropriateness is all you need!", "abstract": "  The strive to make AI applications \"safe\" has led to the development of\nsafety-measures as the main or even sole normative requirement of their\npermissible use. Similar can be attested to the latest version of chatbots,\nsuch as chatGPT. In this view, if they are \"safe\", they are supposed to be\npermissible to deploy. This approach, which we call \"safety-normativity\", is\nrather limited in solving the emerging issues that chatGPT and other chatbots\nhave caused thus far. In answering this limitation, in this paper we argue for\nlimiting chatbots in the range of topics they can chat about according to the\nnormative concept of appropriateness. We argue that rather than looking for\n\"safety\" in a chatbot's utterances to determine what they may and may not say,\nwe ought to assess those utterances according to three forms of\nappropriateness: technical-discursive, social, and moral. We then spell out\nwhat requirements for chatbots follow from these forms of appropriateness to\navoid the limits of previous accounts: positionality, acceptability, and value\nalignment (PAVA). With these in mind, we may be able to determine what a\nchatbot may and may not say. Lastly, one initial suggestion is to use challenge\nsets, specifically designed for appropriateness, as a validation method.\n", "date": "2023-04-27", "categories": []}
{"id": "2304.14338", "title": "An Audit Framework for Adopting AI-Nudging on Children", "abstract": "  This is an audit framework for AI-nudging. Unlike the static form of nudging\nusually discussed in the literature, we focus here on a type of nudging that\nuses large amounts of data to provide personalized, dynamic feedback and\ninterfaces. We call this AI-nudging (Lanzing, 2019, p. 549; Yeung, 2017). The\nultimate goal of the audit outlined here is to ensure that an AI system that\nuses nudges will maintain a level of moral inertia and neutrality by complying\nwith the recommendations, requirements, or suggestions of the audit (in other\nwords, the criteria of the audit). In the case of unintended negative\nconsequences, the audit suggests risk mitigation mechanisms that can be put in\nplace. In the case of unintended positive consequences, it suggests some\nreinforcement mechanisms. Sponsored by the IBM-Notre Dame Tech Ethics Lab\n", "date": "2023-04-25", "categories": []}
{"id": "2304.13493", "title": "Towards clinical AI fairness: A translational perspective", "abstract": "  Artificial intelligence (AI) has demonstrated the ability to extract insights\nfrom data, but the issue of fairness remains a concern in high-stakes fields\nsuch as healthcare. Despite extensive discussion and efforts in algorithm\ndevelopment, AI fairness and clinical concerns have not been adequately\naddressed. In this paper, we discuss the misalignment between technical and\nclinical perspectives of AI fairness, highlight the barriers to AI fairness'\ntranslation to healthcare, advocate multidisciplinary collaboration to bridge\nthe knowledge gap, and provide possible solutions to address the clinical\nconcerns pertaining to AI fairness.\n", "date": "2023-04-26", "categories": []}
{"id": "2304.11861", "title": "Towards a Praxis for Intercultural Ethics in Explainable AI", "abstract": "  Explainable AI (XAI) is often promoted with the idea of helping users\nunderstand how machine learning models function and produce predictions. Still,\nmost of these benefits are reserved for those with specialized domain\nknowledge, such as machine learning developers. Recent research has argued that\nmaking AI explainable can be a viable way of making AI more useful in\nreal-world contexts, especially within low-resource domains in the Global\nSouth. While AI has transcended borders, a limited amount of work focuses on\ndemocratizing the concept of explainable AI to the \"majority world\", leaving\nmuch room to explore and develop new approaches within this space that cater to\nthe distinct needs of users within culturally and socially-diverse regions.\nThis article introduces the concept of an intercultural ethics approach to AI\nexplainability. It examines how cultural nuances impact the adoption and use of\ntechnology, the factors that impede how technical concepts such as AI are\nexplained, and how integrating an intercultural ethics approach in the\ndevelopment of XAI can improve user understanding and facilitate efficient\nusage of these methods.\n", "date": "2023-04-24", "categories": []}
{"id": "2304.11163", "title": "ChatGPT, Large Language Technologies, and the Bumpy Road of Benefiting\n  Humanity", "abstract": "  The allure of emerging AI technologies is undoubtedly thrilling. However, the\npromise that AI technologies will benefit all of humanity is empty so long as\nwe lack a nuanced understanding of what humanity is supposed to be in the face\nof widening global inequality and pressing existential threats. Going forward,\nit is crucial to invest in rigorous and collaborative AI safety and ethics\nresearch. We also need to develop standards in a sustainable and equitable way\nthat differentiate between merely speculative and well-researched questions.\nOnly the latter enable us to co-construct and deploy the values that are\nnecessary for creating beneficial AI. Failure to do so could result in a future\nin which our AI technological advancements outstrip our ability to navigate\ntheir ethical and social implications. This path we do not want to go down.\n", "date": "2023-04-21", "categories": []}
{"id": "2304.11215", "title": "ChatGPT: More than a Weapon of Mass Deception, Ethical challenges and\n  responses from the Human-Centered Artificial Intelligence (HCAI) perspective", "abstract": "  This article explores the ethical problems arising from the use of ChatGPT as\na kind of generative AI and suggests responses based on the Human-Centered\nArtificial Intelligence (HCAI) framework. The HCAI framework is appropriate\nbecause it understands technology above all as a tool to empower, augment, and\nenhance human agency while referring to human wellbeing as a grand challenge,\nthus perfectly aligning itself with ethics, the science of human flourishing.\nFurther, HCAI provides objectives, principles, procedures, and structures for\nreliable, safe, and trustworthy AI which we apply to our ChatGPT assessments.\nThe main danger ChatGPT presents is the propensity to be used as a weapon of\nmass deception (WMD) and an enabler of criminal activities involving deceit. We\nreview technical specifications to better comprehend its potentials and\nlimitations. We then suggest both technical (watermarking, styleme, detectors,\nand fact-checkers) and non-technical measures (terms of use, transparency,\neducator considerations, HITL) to mitigate ChatGPT misuse or abuse and\nrecommend best uses (creative writing, non-creative writing, teaching and\nlearning). We conclude with considerations regarding the role of humans in\nensuring the proper use of ChatGPT for individual and social wellbeing.\n", "date": "2023-04-06", "categories": []}
{"id": "2304.11218", "title": "Explainability in AI Policies: A Critical Review of Communications,\n  Reports, Regulations, and Standards in the EU, US, and UK", "abstract": "  Public attention towards explainability of artificial intelligence (AI)\nsystems has been rising in recent years to offer methodologies for human\noversight. This has translated into the proliferation of research outputs, such\nas from Explainable AI, to enhance transparency and control for system\ndebugging and monitoring, and intelligibility of system process and output for\nuser services. Yet, such outputs are difficult to adopt on a practical level\ndue to a lack of a common regulatory baseline, and the contextual nature of\nexplanations. Governmental policies are now attempting to tackle such exigence,\nhowever it remains unclear to what extent published communications,\nregulations, and standards adopt an informed perspective to support research,\nindustry, and civil interests. In this study, we perform the first thematic and\ngap analysis of this plethora of policies and standards on explainability in\nthe EU, US, and UK. Through a rigorous survey of policy documents, we first\ncontribute an overview of governmental regulatory trajectories within AI\nexplainability and its sociotechnical impacts. We find that policies are often\ninformed by coarse notions and requirements for explanations. This might be due\nto the willingness to conciliate explanations foremost as a risk management\ntool for AI oversight, but also due to the lack of a consensus on what\nconstitutes a valid algorithmic explanation, and how feasible the\nimplementation and deployment of such explanations are across stakeholders of\nan organization. Informed by AI explainability research, we conduct a gap\nanalysis of existing policies, leading us to formulate a set of recommendations\non how to address explainability in regulations for AI systems, especially\ndiscussing the definition, feasibility, and usability of explanations, as well\nas allocating accountability to explanation providers.\n", "date": "2023-04-20", "categories": []}
{"id": "2208.14788", "title": "Negative Human Rights as a Basis for Long-term AI Safety and Regulation", "abstract": "  If autonomous AI systems are to be reliably safe in novel situations, they\nwill need to incorporate general principles guiding them to recognize and avoid\nharmful behaviours. Such principles may need to be supported by a binding\nsystem of regulation, which would need the underlying principles to be widely\naccepted. They should also be specific enough for technical implementation.\nDrawing inspiration from law, this article explains how negative human rights\ncould fulfil the role of such principles and serve as a foundation both for an\ninternational regulatory system and for building technical safety constraints\nfor future AI systems.\n", "date": "2022-08-31", "categories": []}
{"id": "2304.06057", "title": "Maximal Fairness", "abstract": "  Fairness in AI has garnered quite some attention in research, and\nincreasingly also in society. The so-called \"Impossibility Theorem\" has been\none of the more striking research results with both theoretical and practical\nconsequences, as it states that satisfying a certain combination of fairness\nmeasures is impossible. To date, this negative result has not yet been\ncomplemented with a positive one: a characterization of which combinations of\nfairness notions are possible. This work aims to fill this gap by identifying\nmaximal sets of commonly used fairness measures that can be simultaneously\nsatisfied. The fairness measures used are demographic parity, equal\nopportunity, false positive parity, predictive parity, predictive equality,\noverall accuracy equality and treatment equality. We conclude that in total 12\nmaximal sets of these fairness measures are possible, among which seven\ncombinations of two measures, and five combinations of three measures. Our work\nraises interest questions regarding the practical relevance of each of these 12\nmaximal fairness notions in various scenarios.\n", "date": "2023-04-12", "categories": []}
{"id": "2304.06508", "title": "Philosophical Foundations of GeoAI: Exploring Sustainability, Diversity,\n  and Bias in GeoAI and Spatial Data Science", "abstract": "  This chapter presents some of the fundamental assumptions and principles that\ncould form the philosophical foundation of GeoAI and spatial data science.\nInstead of reviewing the well-established characteristics of spatial data\n(analysis), including interaction, neighborhoods, and autocorrelation, the\nchapter highlights themes such as sustainability, bias in training data,\ndiversity in schema knowledge, and the (potential lack of) neutrality of GeoAI\nsystems from a unifying ethical perspective. Reflecting on our profession's\nethical implications will assist us in conducting potentially disruptive\nresearch more responsibly, identifying pitfalls in designing, training, and\ndeploying GeoAI-based systems, and developing a shared understanding of the\nbenefits but also potential dangers of artificial intelligence and machine\nlearning research across academic fields, all while sharing our unique\n(geo)spatial perspective with others.\n", "date": "2023-03-27", "categories": []}
{"id": "2304.02839", "title": "Whose Text Is It Anyway? Exploring BigCode, Intellectual Property, and\n  Ethics", "abstract": "  Intelligent or generative writing tools rely on large language models that\nrecognize, summarize, translate, and predict content. This position paper\nprobes the copyright interests of open data sets used to train large language\nmodels (LLMs). Our paper asks, how do LLMs trained on open data sets circumvent\nthe copyright interests of the used data? We start by defining software\ncopyright and tracing its history. We rely on GitHub Copilot as a modern case\nstudy challenging software copyright. Our conclusion outlines obstacles that\ngenerative writing assistants create for copyright, and offers a practical road\nmap for copyright analysis for developers, software law experts, and general\nusers to consider in the context of intelligent LLM-powered writing tools.\n", "date": "2023-04-06", "categories": []}
{"id": "2304.02924", "title": "The Governance of Physical Artificial Intelligence", "abstract": "  Physical artificial intelligence can prove to be one of the most important\nchallenges of the artificial intelligence. The governance of physical\nartificial intelligence would define its responsible intelligent application in\nthe society.\n", "date": "2023-04-06", "categories": []}
{"id": "2212.01834", "title": "Acceleration AI Ethics, the Debate between Innovation and Safety, and\n  Stability AI's Diffusion versus OpenAI's Dall-E", "abstract": "  One objection to conventional AI ethics is that it slows innovation. This\npresentation responds by reconfiguring ethics as an innovation accelerator. The\ncritical elements develop from a contrast between Stability AI's Diffusion and\nOpenAI's Dall-E. By analyzing the divergent values underlying their opposed\nstrategies for development and deployment, five conceptions are identified as\ncommon to acceleration ethics. Uncertainty is understood as positive and\nencouraging, rather than discouraging. Innovation is conceived as intrinsically\nvaluable, instead of worthwhile only as mediated by social effects. AI problems\nare solved by more AI, not less. Permissions and restrictions governing AI\nemerge from a decentralized process, instead of a unified authority. The work\nof ethics is embedded in AI development and application, instead of functioning\nfrom outside. Together, these attitudes and practices remake ethics as\nprovoking rather than restraining artificial intelligence.\n", "date": "2022-12-04", "categories": []}
{"id": "2303.18190", "title": "Assessing Language Model Deployment with Risk Cards", "abstract": "  This paper introduces RiskCards, a framework for structured assessment and\ndocumentation of risks associated with an application of language models. As\nwith all language, text generated by language models can be harmful, or used to\nbring about harm. Automating language generation adds both an element of scale\nand also more subtle or emergent undesirable tendencies to the generated text.\nPrior work establishes a wide variety of language model harms to many different\nactors: existing taxonomies identify categories of harms posed by language\nmodels; benchmarks establish automated tests of these harms; and documentation\nstandards for models, tasks and datasets encourage transparent reporting.\nHowever, there is no risk-centric framework for documenting the complexity of a\nlandscape in which some risks are shared across models and contexts, while\nothers are specific, and where certain conditions may be required for risks to\nmanifest as harms. RiskCards address this methodological gap by providing a\ngeneric framework for assessing the use of a given language model in a given\nscenario. Each RiskCard makes clear the routes for the risk to manifest harm,\ntheir placement in harm taxonomies, and example prompt-output pairs. While\nRiskCards are designed to be open-source, dynamic and participatory, we present\na \"starter set\" of RiskCards taken from a broad literature survey, each of\nwhich details a concrete risk presentation. Language model RiskCards initiate a\ncommunity knowledge base which permits the mapping of risks and harms to a\nspecific model or its application scenario, ultimately contributing to a\nbetter, safer and shared understanding of the risk landscape.\n", "date": "2023-03-31", "categories": []}
{"id": "2303.17509", "title": "The Full Rights Dilemma for A.I. Systems of Debatable Personhood", "abstract": "  An Artificially Intelligent system (an AI) has debatable personhood if it's\nepistemically possible either that the AI is a person or that it falls far\nshort of personhood. Debatable personhood is a likely outcome of AI development\nand might arise soon. Debatable AI personhood throws us into a catastrophic\nmoral dilemma: Either treat the systems as moral persons and risk sacrificing\nreal human interests for the sake of entities without interests worth the\nsacrifice, or don't treat the systems as moral persons and risk perpetrating\ngrievous moral wrongs against them. The moral issues become even more\nperplexing if we consider cases of possibly conscious AI that are subhuman,\nsuperhuman, or highly divergent from us in their morally relevant properties.\n", "date": "2023-02-21", "categories": []}
{"id": "2303.09377", "title": "Protecting Society from AI Misuse: When are Restrictions on Capabilities\n  Warranted?", "abstract": "  Artificial intelligence (AI) systems will increasingly be used to cause harm\nas they grow more capable. In fact, AI systems are already starting to be used\nto automate fraudulent activities, violate human rights, create harmful fake\nimages, and identify dangerous toxins. To prevent some misuses of AI, we argue\nthat targeted interventions on certain capabilities will be warranted. These\nrestrictions may include controlling who can access certain types of AI models,\nwhat they can be used for, whether outputs are filtered or can be traced back\nto their user, and the resources needed to develop them. We also contend that\nsome restrictions on non-AI capabilities needed to cause harm will be required.\nThough capability restrictions risk reducing use more than misuse (facing an\nunfavorable Misuse-Use Tradeoff), we argue that interventions on capabilities\nare warranted when other interventions are insufficient, the potential harm\nfrom misuse is high, and there are targeted ways to intervene on capabilities.\nWe provide a taxonomy of interventions that can reduce AI misuse, focusing on\nthe specific steps required for a misuse to cause harm (the Misuse Chain), and\na framework to determine if an intervention is warranted. We apply this\nreasoning to three examples: predicting novel toxins, creating harmful images,\nand automating spear phishing campaigns.\n", "date": "2023-03-16", "categories": []}
{"id": "2303.16626", "title": "Fairlearn: Assessing and Improving Fairness of AI Systems", "abstract": "  Fairlearn is an open source project to help practitioners assess and improve\nfairness of artificial intelligence (AI) systems. The associated Python\nlibrary, also named fairlearn, supports evaluation of a model's output across\naffected populations and includes several algorithms for mitigating fairness\nissues. Grounded in the understanding that fairness is a sociotechnical\nchallenge, the project integrates learning resources that aid practitioners in\nconsidering a system's broader societal context.\n", "date": "2023-03-29", "categories": []}
{"id": "2303.10854", "title": "Dynamic Documentation for AI Systems", "abstract": "  AI documentation is a rapidly-growing channel for coordinating the design of\nAI technologies with policies for transparency and accessibility. Calls to\nstandardize and enact documentation of algorithmic harms and impacts are now\ncommonplace. However, documentation standards for AI remain inchoate, and fail\nto match the capabilities and social effects of increasingly impactful\narchitectures such as Large Language Models (LLMs). In this paper, we show the\nlimits of present documentation protocols, and argue for dynamic documentation\nas a new paradigm for understanding and evaluating AI systems. We first review\ncanonical approaches to system documentation outside the context of AI,\nfocusing on the complex history of Environmental Impact Statements (EISs). We\nnext compare critical elements of the EIS framework to present challenges with\nalgorithmic documentation, which have inherited the limitations of EISs without\nincorporating their strengths. These challenges are specifically illustrated\nthrough the growing popularity of Model Cards and two case studies of\nalgorithmic impact assessment in China and Canada. Finally, we evaluate more\nrecent proposals, including Reward Reports, as potential components of fully\ndynamic AI documentation protocols.\n", "date": "2023-03-20", "categories": []}
{"id": "2303.08177", "title": "The Equitable AI Research Roundtable (EARR): Towards Community-Based\n  Decision Making in Responsible AI Development", "abstract": "  This paper reports on our initial evaluation of The Equitable AI Research\nRoundtable -- a coalition of experts in law, education, community engagement,\nsocial justice, and technology. EARR was created in collaboration among a large\ntech firm, nonprofits, NGO research institutions, and universities to provide\ncritical research based perspectives and feedback on technology's emergent\nethical and social harms. Through semi-structured workshops and discussions\nwithin the large tech firm, EARR has provided critical perspectives and\nfeedback on how to conceptualize equity and vulnerability as they relate to AI\ntechnology. We outline three principles in practice of how EARR has operated\nthus far that are especially relevant to the concerns of the FAccT community:\nhow EARR expands the scope of expertise in AI development, how it fosters\nopportunities for epistemic curiosity and responsibility, and that it creates a\nspace for mutual learning. This paper serves as both an analysis and\ntranslation of lessons learned through this engagement approach, and the\npossibilities for future research.\n", "date": "2023-03-14", "categories": []}
{"id": "2303.08721", "title": "Artificial Influence: An Analysis Of AI-Driven Persuasion", "abstract": "  Persuasion is a key aspect of what it means to be human, and is central to\nbusiness, politics, and other endeavors. Advancements in artificial\nintelligence (AI) have produced AI systems that are capable of persuading\nhumans to buy products, watch videos, click on search results, and more. Even\nsystems that are not explicitly designed to persuade may do so in practice. In\nthe future, increasingly anthropomorphic AI systems may form ongoing\nrelationships with users, increasing their persuasive power. This paper\ninvestigates the uncertain future of persuasive AI systems. We examine ways\nthat AI could qualitatively alter our relationship to and views regarding\npersuasion by shifting the balance of persuasive power, allowing personalized\npersuasion to be deployed at scale, powering misinformation campaigns, and\nchanging the way humans can shape their own discourse. We consider ways\nAI-driven persuasion could differ from human-driven persuasion. We warn that\nubiquitous highlypersuasive AI systems could alter our information environment\nso significantly so as to contribute to a loss of human control of our own\nfuture. In response, we examine several potential responses to AI-driven\npersuasion: prohibition, identification of AI agents, truthful AI, and legal\nremedies. We conclude that none of these solutions will be airtight, and that\nindividuals and governments will need to take active steps to guard against the\nmost pernicious effects of persuasive AI.\n", "date": "2023-03-15", "categories": []}
{"id": "2303.07439", "title": "On the ethics of constructing conscious AI", "abstract": "  In its pragmatic turn, the new discipline of AI ethics came to be dominated\nby humanity's collective fear of its creatures, as reflected in an extensive\nand perennially popular literary tradition. Dr. Frankenstein's monster in the\nnovel by Mary Shelley rising against its creator; the unorthodox golem in H.\nLeivick's 1920 play going on a rampage; the rebellious robots of Karel\n\\v{C}apek -- these and hundreds of other examples of the genre are the\nbackground against which the preoccupation of AI ethics with preventing robots\nfrom behaving badly towards people is best understood. In each of these three\nfictional cases (as well as in many others), the miserable artificial creature\n-- mercilessly exploited, or cornered by a murderous mob, and driven to\nviolence in self-defense -- has its author's sympathy. In real life, with very\nfew exceptions, things are different: theorists working on the ethics of AI\ncompletely ignore the possibility of robots needing protection from their\ncreators. The present book chapter takes up this, less commonly considered,\nethical angle of AI.\n", "date": "2023-03-13", "categories": []}
{"id": "2303.05500", "title": "Users are the North Star for AI Transparency", "abstract": "  Despite widespread calls for transparent artificial intelligence systems, the\nterm is too overburdened with disparate meanings to express precise policy aims\nor to orient concrete lines of research. Consequently, stakeholders often talk\npast each other, with policymakers expressing vague demands and practitioners\ndevising solutions that may not address the underlying concerns. Part of why\nthis happens is that a clear ideal of AI transparency goes unsaid in this body\nof work. We explicitly name such a north star -- transparency that is\nuser-centered, user-appropriate, and honest. We conduct a broad literature\nsurvey, identifying many clusters of similar conceptions of transparency, tying\neach back to our north star with analysis of how it furthers or hinders our\nideal AI transparency goals. We conclude with a discussion on common threads\nacross all the clusters, to provide clearer common language whereby\npolicymakers, stakeholders, and practitioners can communicate concrete demands\nand deliver appropriate solutions. We hope for future work on AI transparency\nthat further advances confident, user-beneficial goals and provides clarity to\nregulators and developers alike.\n", "date": "2023-03-09", "categories": []}
{"id": "2303.03929", "title": "Towards Measuring Ethicality of an Intelligent Assistive System", "abstract": "  Artificial intelligence (AI) based assistive systems, so called intelligent\nassistive technology (IAT) are becoming increasingly ubiquitous by each day.\nIAT helps people in improving their quality of life by providing intelligent\nassistance based on the provided data. A few examples of such IATs include\nself-driving cars, robot assistants and smart-health management solutions.\nHowever, the presence of such autonomous entities poses ethical challenges\nconcerning the stakeholders involved in using these systems. There is a lack of\nresearch when it comes to analysing how such IAT adheres to provided ethical\nregulations due to ethical, logistic and cost issues associated with such an\nanalysis. In the light of the above-mentioned problem statement and issues, we\npresent a method to measure the ethicality of an assistive system. To perform\nthis task, we utilised our simulation tool that focuses on modelling navigation\nand assistance of Persons with Dementia (PwD) in indoor environments. By\nutilising this tool, we analyse how well different assistive strategies adhere\nto provided ethical regulations such as autonomy, justice and beneficence of\nthe stakeholders.\n", "date": "2023-02-28", "categories": []}
{"id": "2303.02891", "title": "Perspectives on the Social Impacts of Reinforcement Learning with Human\n  Feedback", "abstract": "  Is it possible for machines to think like humans? And if it is, how should we\ngo about teaching them to do so? As early as 1950, Alan Turing stated that we\nought to teach machines in the way of teaching a child. Reinforcement learning\nwith human feedback (RLHF) has emerged as a strong candidate toward allowing\nagents to learn from human feedback in a naturalistic manner. RLHF is distinct\nfrom traditional reinforcement learning as it provides feedback from a human\nteacher in addition to a reward signal. It has been catapulted into public view\nby multiple high-profile AI applications, including OpenAI's ChatGPT,\nDeepMind's Sparrow, and Anthropic's Claude. These highly capable chatbots are\nalready overturning our understanding of how AI interacts with humanity. The\nwide applicability and burgeoning success of RLHF strongly motivate the need to\nevaluate its social impacts. In light of recent developments, this paper\nconsiders an important question: can RLHF be developed and used without\nnegatively affecting human societies? Our objectives are threefold: to provide\na systematic study of the social effects of RLHF; to identify key social and\nethical issues of RLHF; and to discuss social impacts for stakeholders.\nAlthough text-based applications of RLHF have received much attention, it is\ncrucial to consider when evaluating its social implications the diverse range\nof areas to which it may be deployed. We describe seven primary ways in which\nRLHF-based technologies will affect society by positively transforming human\nexperiences with AI. This paper ultimately proposes that RLHF has potential to\nnet positively impact areas of misinformation, AI value-alignment, bias, AI\naccess, cross-cultural dialogue, industry, and workforce. As RLHF raises\nconcerns that echo those of existing AI technologies, it will be important for\nall to be aware and intentional in the adoption of RLHF.\n", "date": "2023-03-06", "categories": []}
{"id": "2301.05517", "title": "Developing Responsible Chatbots for Financial Services: A\n  Pattern-Oriented Responsible AI Engineering Approach", "abstract": "  The recent release of ChatGPT has gained huge attention and discussion\nworldwide, with responsible AI being a key topic of discussion. How can we\nensure that AI systems, including ChatGPT, are developed and adopted in a\nresponsible way? To tackle the responsible AI challenges, various ethical\nprinciples have been released by governments, organisations, and companies.\nHowever, those principles are very abstract and not practical enough. Further,\nsignificant efforts have been put on algorithm-level solutions that only\naddress a narrow set of principles, such as fairness and privacy. To fill the\ngap, we adopt a pattern-oriented responsible AI engineering approach and build\na Responsible AI Pattern Catalogue to operationalise responsible AI from a\nsystem perspective. In this article, we first summarise the major challenges in\noperationalising responsible AI at scale and introduce how we use the\nResponsible AI Pattern Catalogue to address those challenges. We then examine\nthe risks at each stage of the chatbot development process and recommend\npattern-driven mitigations to evaluate the the usefulness of the Responsible AI\nPattern Catalogue in a real-world setting.\n", "date": "2023-01-03", "categories": []}
{"id": "2302.12683", "title": "Intersectional Fairness: A Fractal Approach", "abstract": "  The issue of fairness in AI has received an increasing amount of attention in\nrecent years. The problem can be approached by looking at different protected\nattributes (e.g., ethnicity, gender, etc) independently, but fairness for\nindividual protected attributes does not imply intersectional fairness. In this\nwork, we frame the problem of intersectional fairness within a geometrical\nsetting. We project our data onto a hypercube, and split the analysis of\nfairness by levels, where each level encodes the number of protected attributes\nwe are intersecting over. We prove mathematically that, while fairness does not\npropagate \"down\" the levels, it does propagate \"up\" the levels. This means that\nensuring fairness for all subgroups at the lowest intersectional level (e.g.,\nblack women, white women, black men and white men), will necessarily result in\nfairness for all the above levels, including each of the protected attributes\n(e.g., ethnicity and gender) taken independently. We also derive a formula\ndescribing the variance of the set of estimated success rates on each level,\nunder the assumption of perfect fairness. Using this theoretical finding as a\nbenchmark, we define a family of metrics which capture overall intersectional\nbias. Finally, we propose that fairness can be metaphorically thought of as a\n\"fractal\" problem. In fractals, patterns at the smallest scale repeat at a\nlarger scale. We see from this example that tackling the problem at the lowest\npossible level, in a bottom-up manner, leads to the natural emergence of fair\nAI. We suggest that trustworthiness is necessarily an emergent, fractal and\nrelational property of the AI system.\n", "date": "2023-02-24", "categories": []}
{"id": "2302.12149", "title": "Beyond Bias and Compliance: Towards Individual Agency and Plurality of\n  Ethics in AI", "abstract": "  AI ethics is an emerging field with multiple, competing narratives about how\nto best solve the problem of building human values into machines. Two major\napproaches are focused on bias and compliance, respectively. But neither of\nthese ideas fully encompasses ethics: using moral principles to decide how to\nact in a particular situation. Our method posits that the way data is labeled\nplays an essential role in the way AI behaves, and therefore in the ethics of\nmachines themselves. The argument combines a fundamental insight from ethics\n(i.e. that ethics is about values) with our practical experience building and\nscaling machine learning systems. We want to build AI that is actually ethical\nby first addressing foundational concerns: how to build good systems, how to\ndefine what is good in relation to system architecture, and who should provide\nthat definition.\n  Building ethical AI creates a foundation of trust between a company and the\nusers of that platform. But this trust is unjustified unless users experience\nthe direct value of ethical AI. Until users have real control over how\nalgorithms behave, something is missing in current AI solutions. This causes\nmassive distrust in AI, and apathy towards AI ethics solutions. The scope of\nthis paper is to propose an alternative path that allows for the plurality of\nvalues and the freedom of individual expression. Both are essential for\nrealizing true moral character.\n", "date": "2023-02-23", "categories": []}
{"id": "2302.08759", "title": "Value Engineering for Autonomous Agents", "abstract": "  Machine Ethics (ME) is concerned with the design of Artificial Moral Agents\n(AMAs), i.e. autonomous agents capable of reasoning and behaving according to\nmoral values. Previous approaches have treated values as labels associated with\nsome actions or states of the world, rather than as integral components of\nagent reasoning. It is also common to disregard that a value-guided agent\noperates alongside other value-guided agents in an environment governed by\nnorms, thus omitting the social dimension of AMAs. In this blue sky paper, we\npropose a new AMA paradigm grounded in moral and social psychology, where\nvalues are instilled into agents as context-dependent goals. These goals\nintricately connect values at individual levels to norms at a collective level\nby evaluating the outcomes most incentivized by the norms in place. We argue\nthat this type of normative reasoning, where agents are endowed with an\nunderstanding of norms' moral implications, leads to value-awareness in\nautonomous agents. Additionally, this capability paves the way for agents to\nalign the norms enforced in their societies with respect to the human values\ninstilled in them, by complementing the value-based reasoning on norms with\nagreement mechanisms to help agents collectively agree on the best set of norms\nthat suit their human values. Overall, our agent model goes beyond the\ntreatment of values as inert labels by connecting them to normative reasoning\nand to the social functionalities needed to integrate value-aware agents into\nour modern hybrid human-computer societies.\n", "date": "2023-02-17", "categories": []}
{"id": "2302.07872", "title": "Data-Centric Governance", "abstract": "  Artificial intelligence (AI) governance is the body of standards and\npractices used to ensure that AI systems are deployed responsibly. Current AI\ngovernance approaches consist mainly of manual review and documentation\nprocesses. While such reviews are necessary for many systems, they are not\nsufficient to systematically address all potential harms, as they do not\noperationalize governance requirements for system engineering, behavior, and\noutcomes in a way that facilitates rigorous and reproducible evaluation. Modern\nAI systems are data-centric: they act on data, produce data, and are built\nthrough data engineering. The assurance of governance requirements must also be\ncarried out in terms of data. This work explores the systematization of\ngovernance requirements via datasets and algorithmic evaluations. When applied\nthroughout the product lifecycle, data-centric governance decreases time to\ndeployment, increases solution quality, decreases deployment risks, and places\nthe system in a continuous state of assured compliance with governance\nrequirements.\n", "date": "2023-02-14", "categories": []}
{"id": "2302.04218", "title": "On the Computational Complexity of Ethics: Moral Tractability for Minds\n  and Machines", "abstract": "  Why should moral philosophers, moral psychologists, and machine ethicists\ncare about computational complexity? Debates on whether artificial intelligence\n(AI) can or should be used to solve problems in ethical domains have mainly\nbeen driven by what AI can or cannot do in terms of human capacities. In this\npaper, we tackle the problem from the other end by exploring what kind of moral\nmachines are possible based on what computational systems can or cannot do. To\ndo so, we analyze normative ethics through the lens of computational\ncomplexity. First, we introduce computational complexity for the uninitiated\nreader and discuss how the complexity of ethical problems can be framed within\nMarr's three levels of analysis. We then study a range of ethical problems\nbased on consequentialism, deontology, and virtue ethics, with the aim of\nelucidating the complexity associated with the problems themselves (e.g., due\nto combinatorics, uncertainty, strategic dynamics), the computational methods\nemployed (e.g., probability, logic, learning), and the available resources\n(e.g., time, knowledge, learning). The results indicate that most problems the\nnormative frameworks pose lead to tractability issues in every category\nanalyzed. Our investigation also provides several insights about the\ncomputational nature of normative ethics, including the differences between\nrule- and outcome-based moral strategies, and the implementation-variance with\nregard to moral resources. We then discuss the consequences complexity results\nhave for the prospect of moral machines in virtue of the trade-off between\noptimality and efficiency. Finally, we elucidate how computational complexity\ncan be used to inform both philosophical and cognitive-psychological research\non human morality by advancing the Moral Tractability Thesis (MTT).\n", "date": "2023-02-08", "categories": []}
{"id": "2302.03567", "title": "From Utilitarian to Rawlsian Designs for Algorithmic Fairness", "abstract": "  There is a lack of consensus within the literature as to how `fairness' of\nalgorithmic systems can be measured, and different metrics can often be at\nodds. In this paper, we approach this task by drawing on the ethical frameworks\nof utilitarianism and John Rawls. Informally, these two theories of\ndistributive justice measure the `good' as either a population's sum of\nutility, or worst-off outcomes, respectively. We present a parameterized class\nof objective functions that interpolates between these two (possibly)\nconflicting notions of the `good'. This class is shown to represent a\nrelaxation of the Rawlsian `veil of ignorance', and its sequence of optimal\nsolutions converges to both a utilitarian and Rawlsian optimum. Several other\nproperties of this class are studied, including: 1) a relationship to\nregularized optimization, 2) feasibility of consistent estimation, and 3)\nalgorithmic cost. In several real-world datasets, we compute optimal solutions\nand construct the tradeoff between utilitarian and Rawlsian notions of the\n`good'. Empirically, we demonstrate that increasing model complexity can\nmanifest strict improvements to both measures of the `good'. This work suggests\nthat the proper degree of `fairness' can be informed by a designer's\npreferences over the space of induced utilitarian and Rawlsian `good'.\n", "date": "2023-02-07", "categories": []}
{"id": "2302.01448", "title": "Out of Context: Investigating the Bias and Fairness Concerns of\n  \"Artificial Intelligence as a Service\"", "abstract": "  \"AI as a Service\" (AIaaS) is a rapidly growing market, offering various\nplug-and-play AI services and tools. AIaaS enables its customers (users) - who\nmay lack the expertise, data, and/or resources to develop their own systems -\nto easily build and integrate AI capabilities into their applications. Yet, it\nis known that AI systems can encapsulate biases and inequalities that can have\nsocietal impact. This paper argues that the context-sensitive nature of\nfairness is often incompatible with AIaaS' 'one-size-fits-all' approach,\nleading to issues and tensions. Specifically, we review and systematise the\nAIaaS space by proposing a taxonomy of AI services based on the levels of\nautonomy afforded to the user. We then critically examine the different\ncategories of AIaaS, outlining how these services can lead to biases or be\notherwise harmful in the context of end-user applications. In doing so, we seek\nto draw research attention to the challenges of this emerging area.\n", "date": "2023-02-02", "categories": []}
{"id": "2206.00335", "title": "Putting AI Ethics into Practice: The Hourglass Model of Organizational\n  AI Governance", "abstract": "  The organizational use of artificial intelligence (AI) has rapidly spread\nacross various sectors. Alongside the awareness of the benefits brought by AI,\nthere is a growing consensus on the necessity of tackling the risks and\npotential harms, such as bias and discrimination, brought about by advanced AI\ntechnologies. A multitude of AI ethics principles have been proposed to tackle\nthese risks, but the outlines of organizational processes and practices for\nensuring socially responsible AI development are in a nascent state. To address\nthe paucity of comprehensive governance models, we present an AI governance\nframework, the hourglass model of organizational AI governance, which targets\norganizations that develop and use AI systems. The framework is designed to\nhelp organizations deploying AI systems translate ethical AI principles into\npractice and align their AI systems and processes with the forthcoming European\nAI Act. The hourglass framework includes governance requirements at the\nenvironmental, organizational, and AI system levels. At the AI system level, we\nconnect governance requirements to AI system life cycles to ensure governance\nthroughout the system's life span. The governance model highlights the systemic\nnature of AI governance and opens new research avenues into its practical\nimplementation, the mechanisms that connect different AI governance layers, and\nthe dynamics between the AI governance actors. The model also offers a starting\npoint for organizational decision-makers to consider the governance components\nneeded to ensure social acceptability, mitigate risks, and realize the\npotential of AI.\n", "date": "2022-06-01", "categories": []}
{"id": "2301.12066", "title": "Truth Machines: Synthesizing Veracity in AI Language Models", "abstract": "  As AI technologies are rolled out into healthcare, academia, human resources,\nlaw, and a multitude of other domains, they become de-facto arbiters of truth.\nBut truth is highly contested, with many different definitions and approaches.\nThis article discusses the struggle for truth in AI systems and the general\nresponses to date. It then investigates the production of truth in InstructGPT,\na large language model, highlighting how data harvesting, model architectures,\nand social feedback mechanisms weave together disparate understandings of\nveracity. It conceptualizes this performance as an operationalization of truth,\nwhere distinct, often conflicting claims are smoothly synthesized and\nconfidently presented into truth-statements. We argue that these same logics\nand inconsistencies play out in Instruct's successor, ChatGPT, reiterating\ntruth as a non-trivial problem. We suggest that enriching sociality and\nthickening \"reality\" are two promising vectors for enhancing the\ntruth-evaluating capacities of future language models. We conclude, however, by\nstepping back to consider AI truth-telling as a social practice: what kind of\n\"truth\" do we as listeners desire?\n", "date": "2023-01-28", "categories": []}
{"id": "2301.10001", "title": "Fiduciary Responsibility: Facilitating Public Trust in Automated\n  Decision Making", "abstract": "  Automated decision-making systems are being increasingly deployed and affect\nthe public in a multitude of positive and negative ways. Governmental and\nprivate institutions use these systems to process information according to\ncertain human-devised rules in order to address social problems or\norganizational challenges. Both research and real-world experience indicate\nthat the public lacks trust in automated decision-making systems and the\ninstitutions that deploy them. The recreancy theorem argues that the public is\nmore likely to trust and support decisions made or influenced by automated\ndecision-making systems if the institutions that administer them meet their\nfiduciary responsibility. However, often the public is never informed of how\nthese systems operate and resultant institutional decisions are made. A ``black\nbox'' effect of automated decision-making systems reduces the public's\nperceptions of integrity and trustworthiness. The result is that the public\nloses the capacity to identify, challenge, and rectify unfairness or the costs\nassociated with the loss of public goods or benefits.\n  The current position paper defines and explains the role of fiduciary\nresponsibility within an automated decision-making system. We formulate an\nautomated decision-making system as a data science lifecycle (DSL) and examine\nthe implications of fiduciary responsibility within the context of the DSL.\nFiduciary responsibility within DSLs provides a methodology for addressing the\npublic's lack of trust in automated decision-making systems and the\ninstitutions that employ them to make decisions affecting the public. We posit\nthat fiduciary responsibility manifests in several contexts of a DSL, each of\nwhich requires its own mitigation of sources of mistrust. To instantiate\nfiduciary responsibility, a Los Angeles Police Department (LAPD) predictive\npolicing case study is examined.\n", "date": "2023-01-06", "categories": []}
{"id": "2301.08742", "title": "Unifying Consciousness and Time to Enhance Artificial Intelligence", "abstract": "  Consciousness is a sequential process of awareness which can focus on one\npiece of information at a time. This process of awareness experiences causation\nwhich underpins the notion of time while it interplays with matter and energy,\nforming reality. The study of Consciousness, time and reality is complex and\nevolving fast in many fields, including metaphysics and fundamental physics.\nReality composes patterns in human Consciousness in response to the\nregularities in nature. These regularities could be physical (e.g.,\nastronomical, environmental), biological, chemical, mental, social, etc. The\npatterns that emerged in Consciousness were correlated to the environment, life\nand social behaviours followed by constructed frameworks, systems and\nstructures. The complex constructs evolved as cultures, customs, norms and\nvalues, which created a diverse society. In the evolution of responsible AI, it\nis important to be attuned to the evolved cultural, ethical and moral values\nthrough Consciousness. This requires the advocated design of self-learning AI\naware of time perception and human ethics.\n", "date": "2023-01-10", "categories": []}
{"id": "2301.07098", "title": "The moral authority of ChatGPT", "abstract": "  ChatGPT is not only fun to chat with, but it also searches information,\nanswers questions, and gives advice. With consistent moral advice, it might\nimprove the moral judgment and decisions of users, who often hold contradictory\nmoral beliefs. Unfortunately, ChatGPT turns out highly inconsistent as a moral\nadvisor. Nonetheless, it influences users' moral judgment, we find in an\nexperiment, even if they know they are advised by a chatting bot, and they\nunderestimate how much they are influenced. Thus, ChatGPT threatens to corrupt\nrather than improves users' judgment. These findings raise the question of how\nto ensure the responsible use of ChatGPT and similar AI. Transparency is often\ntouted but seems ineffective. We propose training to improve digital literacy.\n", "date": "2023-01-13", "categories": []}
{"id": "2301.04713", "title": "How do \"technical\" design-choices made when building algorithmic\n  decision-making tools for criminal justice authorities create constitutional\n  dangers?", "abstract": "  This two part paper argues that seemingly \"technical\" choices made by\ndevelopers of machine-learning based algorithmic tools used to inform decisions\nby criminal justice authorities can create serious constitutional dangers,\nenhancing the likelihood of abuse of decision-making power and the scope and\nmagnitude of injustice. Drawing on three algorithmic tools in use, or recently\nused, to assess the \"risk\" posed by individuals to inform how they should be\ntreated by criminal justice authorities, we integrate insights from data\nscience and public law scholarship to show how public law principles and more\nspecific legal duties that are rooted in these principles, are routinely\noverlooked in algorithmic tool-building and implementation. We argue that\ntechnical developers must collaborate closely with public law experts to ensure\nthat if algorithmic decision-support tools are to inform criminal justice\ndecisions, those tools are configured and implemented in a manner that is\ndemonstrably compliant with public law principles and doctrine, including\nrespect for human rights, throughout the tool-building process.\n", "date": "2023-01-11", "categories": []}
{"id": "2301.05133", "title": "Is AI Art Another Industrial Revolution in the Making?", "abstract": "  A major shift from skilled to unskilled workers was one of the many changes\ncaused by the Industrial Revolution, when the switch to machines contributed to\ndecline in the social and economic status of artisans, whose skills were\ndismembered into discrete actions by factory-line workers. We consider what may\nbe an analogous computing technology: the recent introduction of AI-generated\nart software. AI art generators such as Dall-E and Midjourney can create fully\nrendered images based solely on a user's prompt, just at the click of a button.\nSome artists fear if the cheaper price and conveyor-belt speed that comes with\nAI-produced images is seen as an improvement to the current system, it may\npermanently change the way society values/views art and artists. In this\narticle, we consider the implications that AI art generation introduces through\na post-industrial revolution historical lens. We then reflect on the analogous\nissues that appear to arise as a result of the AI art revolution, and we\nconclude that the problems raised mirror those of industrialization, giving a\nvital glimpse into what may lie ahead.\n", "date": "2023-01-12", "categories": []}
{"id": "2208.14037", "title": "Towards Artificial Virtuous Agents: Games, Dilemmas and Machine Learning", "abstract": "  Machine ethics has received increasing attention over the past few years\nbecause of the need to ensure safe and reliable artificial intelligence (AI).\nThe two dominantly used theories in machine ethics are deontological and\nutilitarian ethics. Virtue ethics, on the other hand, has often been mentioned\nas an alternative ethical theory. While this interesting approach has certain\nadvantages over popular ethical theories, little effort has been put into\nengineering artificial virtuous agents due to challenges in their\nformalization, codifiability, and the resolution of ethical dilemmas to train\nvirtuous agents. We propose to bridge this gap by using role-playing games\nriddled with moral dilemmas. There are several such games in existence, such as\nPapers, Please and Life is Strange, where the main character encounters\nsituations where they must choose the right course of action by giving up\nsomething else dear to them. We draw inspiration from such games to show how a\nsystemic role-playing game can be designed to develop virtues within an\nartificial agent. Using modern day AI techniques, such as affinity-based\nreinforcement learning and explainable AI, we motivate the implementation of\nvirtuous agents that play such role-playing games, and the examination of their\ndecisions through a virtue ethical lens. The development of such agents and\nenvironments is a first step towards practically formalizing and demonstrating\nthe value of virtue ethics in the development of ethical agents.\n", "date": "2022-08-30", "categories": []}
{"id": "2210.08141", "title": "Pseudo AI Bias", "abstract": "  Pseudo Artificial Intelligence bias (PAIB) is broadly disseminated in the\nliterature, which can result in unnecessary AI fear in society, exacerbate the\nenduring inequities and disparities in access to and sharing the benefits of AI\napplications, and waste social capital invested in AI research. This study\nsystematically reviews publications in the literature to present three types of\nPAIBs identified due to: a) misunderstandings, b) pseudo mechanical bias, and\nc) over-expectations. We discussed the consequences of and solutions to PAIBs,\nincluding certifying users for AI applications to mitigate AI fears, providing\ncustomized user guidance for AI applications, and developing systematic\napproaches to monitor bias. We concluded that PAIB due to misunderstandings,\npseudo mechanical bias, and over-expectations of algorithmic predictions is\nsocially harmful.\n", "date": "2022-10-14", "categories": []}
{"id": "2212.12936", "title": "Understanding Ethics, Privacy, and Regulations in Smart Video\n  Surveillance for Public Safety", "abstract": "  Recently, Smart Video Surveillance (SVS) systems have been receiving more\nattention among scholars and developers as a substitute for the current passive\nsurveillance systems. These systems are used to make the policing and\nmonitoring systems more efficient and improve public safety. However, the\nnature of these systems in monitoring the public's daily activities brings\ndifferent ethical challenges. There are different approaches for addressing\nprivacy issues in implementing the SVS. In this paper, we are focusing on the\nrole of design considering ethical and privacy challenges in SVS. Reviewing\nfour policy protection regulations that generate an overview of best practices\nfor privacy protection, we argue that ethical and privacy concerns could be\naddressed through four lenses: algorithm, system, model, and data. As an case\nstudy, we describe our proposed system and illustrate how our system can create\na baseline for designing a privacy perseverance system to deliver safety to\nsociety. We used several Artificial Intelligence algorithms, such as object\ndetection, single and multi camera re-identification, action recognition, and\nanomaly detection, to provide a basic functional system. We also use\ncloud-native services to implement a smartphone application in order to deliver\nthe outputs to the end users.\n", "date": "2022-12-25", "categories": []}
{"id": "2212.12508", "title": "Ethical Design of Computers: From Semiconductors to IoT and Artificial\n  Intelligence", "abstract": "  Computing systems are tightly integrated today into our professional, social,\nand private lives. An important consequence of this growing ubiquity of\ncomputing is that it can have significant ethical implications of which\ncomputing professionals should take account. In most real-world scenarios, it\nis not immediately obvious how particular technical choices during the design\nand use of computing systems could be viewed from an ethical perspective. This\narticle provides a perspective on the ethical challenges within semiconductor\nchip design, IoT applications, and the increasing use of artificial\nintelligence in the design processes, tools, and hardware-software stacks of\nthese systems.\n", "date": "2022-12-23", "categories": []}
{"id": "2212.11207", "title": "A Seven-Layer Model for Standardising AI Fairness Assessment", "abstract": "  Problem statement: Standardisation of AI fairness rules and benchmarks is\nchallenging because AI fairness and other ethical requirements depend on\nmultiple factors such as context, use case, type of the AI system, and so on.\nIn this paper, we elaborate that the AI system is prone to biases at every\nstage of its lifecycle, from inception to its usage, and that all stages\nrequire due attention for mitigating AI bias. We need a standardised approach\nto handle AI fairness at every stage. Gap analysis: While AI fairness is a hot\nresearch topic, a holistic strategy for AI fairness is generally missing. Most\nresearchers focus only on a few facets of AI model-building. Peer review shows\nexcessive focus on biases in the datasets, fairness metrics, and algorithmic\nbias. In the process, other aspects affecting AI fairness get ignored. The\nsolution proposed: We propose a comprehensive approach in the form of a novel\nseven-layer model, inspired by the Open System Interconnection (OSI) model, to\nstandardise AI fairness handling. Despite the differences in the various\naspects, most AI systems have similar model-building stages. The proposed model\nsplits the AI system lifecycle into seven abstraction layers, each\ncorresponding to a well-defined AI model-building or usage stage. We also\nprovide checklists for each layer and deliberate on potential sources of bias\nin each layer and their mitigation methodologies. This work will facilitate\nlayer-wise standardisation of AI fairness rules and benchmarking parameters.\n", "date": "2022-12-21", "categories": []}
{"id": "2212.07508", "title": "Tensions Between the Proxies of Human Values in AI", "abstract": "  Motivated by mitigating potentially harmful impacts of technologies, the AI\ncommunity has formulated and accepted mathematical definitions for certain\npillars of accountability: e.g. privacy, fairness, and model transparency. Yet,\nwe argue this is fundamentally misguided because these definitions are\nimperfect, siloed constructions of the human values they hope to proxy, while\ngiving the guise that those values are sufficiently embedded in our\ntechnologies. Under popularized methods, tensions arise when practitioners\nattempt to achieve each pillar of fairness, privacy, and transparency in\nisolation or simultaneously. In this position paper, we push for redirection.\nWe argue that the AI community needs to consider all the consequences of\nchoosing certain formulations of these pillars -- not just the technical\nincompatibilities, but also the effects within the context of deployment. We\npoint towards sociotechnical research for frameworks for the latter, but push\nfor broader efforts into implementing these in practice.\n", "date": "2022-12-14", "categories": []}
{"id": "2212.05058", "title": "Structured Like a Language Model: Analysing AI as an Automated Subject", "abstract": "  Drawing from the resources of psychoanalysis and critical media studies, in\nthis paper we develop an analysis of Large Language Models (LLMs) as automated\nsubjects. We argue the intentional fictional projection of subjectivity onto\nLLMs can yield an alternate frame through which AI behaviour, including its\nproductions of bias and harm, can be analysed. First, we introduce language\nmodels, discuss their significance and risks, and outline our case for\ninterpreting model design and outputs with support from psychoanalytic\nconcepts. We trace a brief history of language models, culminating with the\nreleases, in 2022, of systems that realise state-of-the-art natural language\nprocessing performance. We engage with one such system, OpenAI's InstructGPT,\nas a case study, detailing the layers of its construction and conducting\nexploratory and semi-structured interviews with chatbots. These interviews\nprobe the model's moral imperatives to be helpful, truthful and harmless by\ndesign. The model acts, we argue, as the condensation of often competing social\ndesires, articulated through the internet and harvested into training data,\nwhich must then be regulated and repressed. This foundational structure can\nhowever be redirected via prompting, so that the model comes to identify with,\nand transfer, its commitments to the immediate human subject before it. In\nturn, these automated productions of language can lead to the human subject\nprojecting agency upon the model, effecting occasionally further forms of\ncountertransference. We conclude that critical media methods and psychoanalytic\ntheory together offer a productive frame for grasping the powerful new\ncapacities of AI-driven language systems.\n", "date": "2022-12-08", "categories": []}
{"id": "2208.06327", "title": "Developing moral AI to support antimicrobial decision making", "abstract": "  Artificial intelligence (AI) assisting with antimicrobial prescribing raises\nsignificant moral questions. Utilising ethical frameworks alongside AI-driven\nsystems, while considering infection specific complexities, can support moral\ndecision making to tackle antimicrobial resistance.\n", "date": "2022-08-12", "categories": []}
{"id": "2212.00740", "title": "Prioritizing Policies for Furthering Responsible Artificial Intelligence\n  in the United States", "abstract": "  Several policy options exist, or have been proposed, to further responsible\nartificial intelligence (AI) development and deployment. Institutions,\nincluding U.S. government agencies, states, professional societies, and private\nand public sector businesses, are well positioned to implement these policies.\nHowever, given limited resources, not all policies can or should be equally\nprioritized. We define and review nine suggested policies for furthering\nresponsible AI, rank each policy on potential use and impact, and recommend\nprioritization relative to each institution type. We find that pre-deployment\naudits and assessments and post-deployment accountability are likely to have\nthe highest impact but also the highest barriers to adoption. We recommend that\nU.S. government agencies and companies highly prioritize development of\npre-deployment audits and assessments, while the U.S. national legislature\nshould highly prioritize post-deployment accountability. We suggest that U.S.\ngovernment agencies and professional societies should highly prioritize\npolicies that support responsible AI research and that states should highly\nprioritize support of responsible AI education. We propose that companies can\nhighly prioritize involving community stakeholders in development efforts and\nsupporting diversity in AI development. We advise lower levels of\nprioritization across institutions for AI ethics statements and databases of AI\ntechnologies or incidents. We recognize that no one policy will lead to\nresponsible AI and instead advocate for strategic policy implementation across\ninstitutions.\n", "date": "2022-11-30", "categories": []}
{"id": "2211.13069", "title": "Cultural Incongruencies in Artificial Intelligence", "abstract": "  Artificial intelligence (AI) systems attempt to imitate human behavior. How\nwell they do this imitation is often used to assess their utility and to\nattribute human-like (or artificial) intelligence to them. However, most work\non AI refers to and relies on human intelligence without accounting for the\nfact that human behavior is inherently shaped by the cultural contexts they are\nembedded in, the values and beliefs they hold, and the social practices they\nfollow. Additionally, since AI technologies are mostly conceived and developed\nin just a handful of countries, they embed the cultural values and practices of\nthese countries. Similarly, the data that is used to train the models also\nfails to equitably represent global cultural diversity. Problems therefore\narise when these technologies interact with globally diverse societies and\ncultures, with different values and interpretive practices. In this position\npaper, we describe a set of cultural dependencies and incongruencies in the\ncontext of AI-based language and vision technologies, and reflect on the\npossibilities of and potential strategies towards addressing these\nincongruencies.\n", "date": "2022-11-19", "categories": []}
{"id": "2211.13130", "title": "A Brief Overview of AI Governance for Responsible Machine Learning\n  Systems", "abstract": "  Organizations of all sizes, across all industries and domains are leveraging\nartificial intelligence (AI) technologies to solve some of their biggest\nchallenges around operations, customer experience, and much more. However, due\nto the probabilistic nature of AI, the risks associated with it are far greater\nthan traditional technologies. Research has shown that these risks can range\nanywhere from regulatory, compliance, reputational, and user trust, to\nfinancial and even societal risks. Depending on the nature and size of the\norganization, AI technologies can pose a significant risk, if not used in a\nresponsible way. This position paper seeks to present a brief introduction to\nAI governance, which is a framework designed to oversee the responsible use of\nAI with the goal of preventing and mitigating risks. Having such a framework\nwill not only manage risks but also gain maximum value out of AI projects and\ndevelop consistency for organization-wide adoption of AI.\n", "date": "2022-11-21", "categories": []}
{"id": "2209.07667", "title": "Can There be Art Without an Artist?", "abstract": "  Generative AI based art has proliferated in the past year, with increasingly\nimpressive use cases from generating fake human faces to the creation of\nsystems that can generate thousands of artistic images from text prompts - some\nof these images have even been \"good\" enough to win accolades from qualified\njudges. In this paper, we explore how Generative Models have impacted artistry,\nnot only from a qualitative point of view, but also from an angle of\nexploitation of artists -- both via plagiarism, where models are trained on\ntheir artwork without permission, and via profit shifting, where profits in the\nart market have shifted from art creators to model owners. However, we posit\nthat if deployed responsibly, AI generative models have the possibility of\nbeing a positive, new modality in art that does not displace or harm existing\nartists.\n", "date": "2022-09-16", "categories": []}
{"id": "2208.09174", "title": "Atomist or Holist? A Diagnosis and Vision for More Productive\n  Interdisciplinary AI Ethics Dialogue", "abstract": "  In response to growing recognition of the social impact of new AI-based\ntechnologies, major AI and ML conferences and journals now encourage or require\npapers to include ethics impact statements and undergo ethics reviews. This\nmove has sparked heated debate concerning the role of ethics in AI research, at\ntimes devolving into name-calling and threats of \"cancellation.\" We diagnose\nthis conflict as one between atomist and holist ideologies. Among other things,\natomists believe facts are and should be kept separate from values, while\nholists believe facts and values are and should be inextricable from one\nanother. With the goal of reducing disciplinary polarization, we draw on\nnumerous philosophical and historical sources to describe each ideology's core\nbeliefs and assumptions. Finally, we call on atomists and holists within the\never-expanding data science community to exhibit greater empathy during ethical\ndisagreements and propose four targeted strategies to ensure AI research\nbenefits society.\n", "date": "2022-08-19", "categories": []}
{"id": "2211.06323", "title": "A Sign That Spells: DALL-E 2, Invisual Images and The Racial Politics of\n  Feature Space", "abstract": "  In this paper, we examine how generative machine learning systems produce a\nnew politics of visual culture. We focus on DALL-E 2 and related models as an\nemergent approach to image-making that operates through the cultural techniques\nof feature extraction and semantic compression. These techniques, we argue, are\ninhuman, invisual, and opaque, yet are still caught in a paradox that is\nironically all too human: the consistent reproduction of whiteness as a latent\nfeature of dominant visual culture. We use Open AI's failed efforts to 'debias'\ntheir system as a critical opening to interrogate how systems like DALL-E 2\ndissolve and reconstitute politically salient human concepts like race. This\nexample vividly illustrates the stakes of this moment of transformation, when\nso-called foundation models reconfigure the boundaries of visual culture and\nwhen 'doing' anti-racism means deploying quick technical fixes to mitigate\npersonal discomfort, or more importantly, potential commercial loss.\n", "date": "2022-10-26", "categories": []}
{"id": "2211.06346", "title": "AI Ethics in Smart Healthcare", "abstract": "  This article reviews the landscape of ethical challenges of integrating\nartificial intelligence (AI) into smart healthcare products, including medical\nelectronic devices. Differences between traditional ethics in the medical\ndomain and emerging ethical challenges with AI-driven healthcare are presented,\nparticularly as they relate to transparency, bias, privacy, safety,\nresponsibility, justice, and autonomy. Open challenges and recommendations are\noutlined to enable the integration of ethical principles into the design,\nvalidation, clinical trials, deployment, monitoring, repair, and retirement of\nAI-based smart healthcare products.\n", "date": "2022-11-02", "categories": []}
{"id": "2211.04602", "title": "System Safety Engineering for Social and Ethical ML Risks: A Case Study", "abstract": "  Governments, industry, and academia have undertaken efforts to identify and\nmitigate harms in ML-driven systems, with a particular focus on social and\nethical risks of ML components in complex sociotechnical systems. However,\nexisting approaches are largely disjointed, ad-hoc and of unknown\neffectiveness. Systems safety engineering is a well established discipline with\na track record of identifying and managing risks in many complex sociotechnical\ndomains. We adopt the natural hypothesis that tools from this domain could\nserve to enhance risk analyses of ML in its context of use. To test this\nhypothesis, we apply a \"best of breed\" systems safety analysis, Systems\nTheoretic Process Analysis (STPA), to a specific high-consequence system with\nan important ML-driven component, namely the Prescription Drug Monitoring\nPrograms (PDMPs) operated by many US States, several of which rely on an\nML-derived risk score. We focus in particular on how this analysis can extend\nto identifying social and ethical risks and developing concrete design-level\ncontrols to mitigate them.\n", "date": "2022-11-08", "categories": []}
{"id": "2210.15289", "title": "On the Efficiency of Ethics as a Governing Tool for Artificial\n  Intelligence", "abstract": "  The 4th Industrial Revolution is the culmination of the digital age.\nNowadays, technologies such as robotics, nanotechnology, genetics, and\nartificial intelligence promise to transform our world and the way we live.\nArtificial Intelligence Ethics and Safety is an emerging research field that\nhas been gaining popularity in recent years. Several private, public and\nnon-governmental organizations have published guidelines proposing ethical\nprinciples for regulating the use and development of autonomous intelligent\nsystems. Meta-analyses of the AI Ethics research field point to convergence on\ncertain principles that supposedly govern the AI industry. However, little is\nknown about the effectiveness of this form of Ethics. In this paper, we would\nlike to conduct a critical analysis of the current state of AI Ethics and\nsuggest that this form of governance based on principled ethical guidelines is\nnot sufficient to norm the AI industry and its developers. We believe that\ndrastic changes are necessary, both in the training processes of professionals\nin the fields related to the development of software and intelligent systems\nand in the increased regulation of these professionals and their industry. To\nthis end, we suggest that law should benefit from recent contributions from\nbioethics, to make the contributions of AI ethics to governance explicit in\nlegal terms.\n", "date": "2022-10-27", "categories": []}
{"id": "2207.12043", "title": "Representational Ethical Model Calibration", "abstract": "  Equity is widely held to be fundamental to the ethics of healthcare. In the\ncontext of clinical decision-making, it rests on the comparative fidelity of\nthe intelligence -- evidence-based or intuitive -- guiding the management of\neach individual patient. Though brought to recent attention by the\nindividuating power of contemporary machine learning, such epistemic equity\narises in the context of any decision guidance, whether traditional or\ninnovative. Yet no general framework for its quantification, let alone\nassurance, currently exists. Here we formulate epistemic equity in terms of\nmodel fidelity evaluated over learnt multi-dimensional representations of\nidentity crafted to maximise the captured diversity of the population,\nintroducing a comprehensive framework for Representational Ethical Model\nCalibration. We demonstrate use of the framework on large-scale multimodal data\nfrom UK Biobank to derive diverse representations of the population, quantify\nmodel performance, and institute responsive remediation. We offer our approach\nas a principled solution to quantifying and assuring epistemic equity in\nhealthcare, with applications across the research, clinical, and regulatory\ndomains.\n", "date": "2022-07-25", "categories": []}
{"id": "2210.08995", "title": "AI, Opacity, and Personal Autonomy", "abstract": "  Advancements in machine learning have fuelled the popularity of using AI\ndecision algorithms in procedures such as bail hearings (Feller et al. 2016),\nmedical diagnoses (Rajkomar et al. 2018; Esteva et al. 2019) and recruitment\n(Heilweil 2019, Van Esch et al. 2019). Academic articles (Floridi et al. 2018),\npolicy texts (HLEG 2019), and popularizing books (O'Neill 2016, Eubanks 2018)\nalike warn that such algorithms tend to be _opaque_: they do not provide\nexplanations for their outcomes. Building on a causal account of transparency\nand opacity as well as recent work on the value of causal explanation (Lombrozo\n2011, Hitchcock 2012), I formulate a moral concern for opaque algorithms that\nis yet to receive a systematic treatment in the literature: when such\nalgorithms are used in life-changing decisions, they can obstruct us from\neffectively shaping our lives according to our goals and preferences, thus\nundermining our autonomy. I argue that this concern deserves closer attention\nas it furnishes the call for transparency in algorithmic decision-making with\nboth new tools and new challenges.\n", "date": "2022-09-25", "categories": []}
{"id": "2210.09086", "title": "Artificial Intelligence Nomenclature Identified From Delphi Study on Key\n  Issues Related to Trust and Barriers to Adoption for Autonomous Systems", "abstract": "  The rapid integration of artificial intelligence across traditional research\ndomains has generated an amalgamation of nomenclature. As cross-discipline\nteams work together on complex machine learning challenges, finding a consensus\nof basic definitions in the literature is a more fundamental problem. As a step\nin the Delphi process to define issues with trust and barriers to the adoption\nof autonomous systems, our study first collected and ranked the top concerns\nfrom a panel of international experts from the fields of engineering, computer\nscience, medicine, aerospace, and defence, with experience working with\nartificial intelligence. This document presents a summary of the literature\ndefinitions for nomenclature derived from expert feedback.\n", "date": "2022-10-14", "categories": []}
{"id": "2210.03535", "title": "From plane crashes to algorithmic harm: applicability of safety\n  engineering frameworks for responsible ML", "abstract": "  Inappropriate design and deployment of machine learning (ML) systems leads to\nnegative downstream social and ethical impact -- described here as social and\nethical risks -- for users, society and the environment. Despite the growing\nneed to regulate ML systems, current processes for assessing and mitigating\nrisks are disjointed and inconsistent. We interviewed 30 industry practitioners\non their current social and ethical risk management practices, and collected\ntheir first reactions on adapting safety engineering frameworks into their\npractice -- namely, System Theoretic Process Analysis (STPA) and Failure Mode\nand Effects Analysis (FMEA). Our findings suggest STPA/FMEA can provide\nappropriate structure toward social and ethical risk assessment and mitigation\nprocesses. However, we also find nontrivial challenges in integrating such\nframeworks in the fast-paced culture of the ML industry. We call on the ML\nresearch community to strengthen existing frameworks and assess their efficacy,\nensuring that ML systems are safer for all people.\n", "date": "2022-10-06", "categories": []}
{"id": "2210.02516", "title": "Equalizing Credit Opportunity in Algorithms: Aligning Algorithmic\n  Fairness Research with U.S. Fair Lending Regulation", "abstract": "  Credit is an essential component of financial wellbeing in America, and\nunequal access to it is a large factor in the economic disparities between\ndemographic groups that exist today. Today, machine learning algorithms,\nsometimes trained on alternative data, are increasingly being used to determine\naccess to credit, yet research has shown that machine learning can encode many\ndifferent versions of \"unfairness,\" thus raising the concern that banks and\nother financial institutions could -- potentially unwittingly -- engage in\nillegal discrimination through the use of this technology. In the US, there are\nlaws in place to make sure discrimination does not happen in lending and\nagencies charged with enforcing them. However, conversations around fair credit\nmodels in computer science and in policy are often misaligned: fair machine\nlearning research often lacks legal and practical considerations specific to\nexisting fair lending policy, and regulators have yet to issue new guidance on\nhow, if at all, credit risk models should be utilizing practices and techniques\nfrom the research community. This paper aims to better align these sides of the\nconversation. We describe the current state of credit discrimination regulation\nin the United States, contextualize results from fair ML research to identify\nthe specific fairness concerns raised by the use of machine learning in\nlending, and discuss regulatory opportunities to address these concerns.\n", "date": "2022-10-05", "categories": []}
{"id": "2210.02667", "title": "A Human Rights-Based Approach to Responsible AI", "abstract": "  Research on fairness, accountability, transparency and ethics of AI-based\ninterventions in society has gained much-needed momentum in recent years.\nHowever it lacks an explicit alignment with a set of normative values and\nprinciples that guide this research and interventions. Rather, an implicit\nconsensus is often assumed to hold for the values we impart into our models -\nsomething that is at odds with the pluralistic world we live in. In this paper,\nwe put forth the doctrine of universal human rights as a set of globally\nsalient and cross-culturally recognized set of values that can serve as a\ngrounding framework for explicit value alignment in responsible AI - and\ndiscuss its efficacy as a framework for civil society partnership and\nparticipation. We argue that a human rights framework orients the research in\nthis space away from the machines and the risks of their biases, and towards\nhumans and the risks to their rights, essentially helping to center the\nconversation around who is harmed, what harms they face, and how those harms\nmay be mitigated.\n", "date": "2022-10-06", "categories": []}
{"id": "2210.02769", "title": "Artificial virtuous agents in a multiagent tragedy of the commons", "abstract": "  Although virtue ethics has repeatedly been proposed as a suitable framework\nfor the development of artificial moral agents (AMAs), it has been proven\ndifficult to approach from a computational perspective. In this work, we\npresent the first technical implementation of artificial virtuous agents (AVAs)\nin moral simulations. First, we review previous conceptual and technical work\nin artificial virtue ethics and describe a functionalistic path to AVAs based\non dispositional virtues, bottom-up learning, and top-down eudaimonic reward.\nWe then provide the details of a technical implementation in a moral simulation\nbased on a tragedy of the commons scenario. The experimental results show how\nthe AVAs learn to tackle cooperation problems while exhibiting core features of\ntheir theoretical counterpart, including moral character, dispositional\nvirtues, learning from experience, and the pursuit of eudaimonia. Ultimately,\nwe argue that virtue ethics provides a compelling path toward morally excellent\nmachines and that our work provides an important starting point for such\nendeavors.\n", "date": "2022-10-06", "categories": []}
{"id": "2209.12879", "title": "Environmental and Social Sustainability of Creative-Ai", "abstract": "  The recent developments of artificial intelligence increase its capability\nfor the creation of arts in both largely autonomous and collaborative contexts.\nIn both contexts, Ai aims to imitate, combine, and extend existing artistic\nstyles, and can transform creative practices. In our ongoing research, we\ninvestigate such Creative-Ai from sustainability and ethical perspectives. The\ntwo main focus areas are understanding the environmental sustainability aspects\n(material, practices) in the context of artistic processes that involve\nCreative-Ai, and ethical issues related to who gets to be involved in the\ncreation process (power, authorship, ownership). This paper provides an outline\nof our ongoing research in these two directions. We will present our\ninterdisciplinary approach, which combines interviews, workshops, online\nethnography, and energy measurements, to address our research questions: How is\nCreative-Ai currently used by artist communities, and which future applications\ndo artists imagine? When Ai is applied to creating art, how might it impact the\neconomy and environment? And, how can answers to these questions guide\nrequirements for intellectual property regimes for Creative-Ai?\n", "date": "2022-09-26", "categories": []}
{"id": "2209.13021", "title": "Inclusive Ethical Design for Recommender Systems", "abstract": "  Recommender systems are becoming increasingly central as mediators of\ninformation with the potential to profoundly influence societal opinion. While\napproaches are being developed to ensure these systems are designed in a\nresponsible way, adolescents in particular, represent a potentially vulnerable\nuser group requiring explicit consideration. This is especially important given\nthe nature of their access and use of recommender systems but also their role\nas providers of content. This paper proposes core principles for the ethical\ndesign of recommender systems and evaluates whether current approaches to\nensuring adherence to these principles are sufficiently inclusive of the\nparticular needs and potential vulnerabilities of adolescent users.\n", "date": "2022-09-13", "categories": []}
{"id": "2209.11762", "title": "Towards Auditing Unsupervised Learning Algorithms and Human Processes\n  For Fairness", "abstract": "  Existing work on fairness typically focuses on making known machine learning\nalgorithms fairer. Fair variants of classification, clustering, outlier\ndetection and other styles of algorithms exist. However, an understudied area\nis the topic of auditing an algorithm's output to determine fairness. Existing\nwork has explored the two group classification problem for binary protected\nstatus variables using standard definitions of statistical parity. Here we\nbuild upon the area of auditing by exploring the multi-group setting under more\ncomplex definitions of fairness.\n", "date": "2022-09-20", "categories": []}
{"id": "2209.07076", "title": "Responsible AI Implementation: A Human-centered Framework for\n  Accelerating the Innovation Process", "abstract": "  There is still a significant gap between expectations and the successful\nadoption of AI to innovate and improve businesses. Due to the emergence of deep\nlearning, AI adoption is more complex as it often incorporates big data and the\ninternet of things, affecting data privacy. Existing frameworks have identified\nthe need to focus on human-centered design, combining technical and\nbusiness/organizational perspectives. However, trust remains a critical issue\nthat needs to be designed from the beginning. The proposed framework expands\nfrom the human-centered design approach, emphasizing and maintaining the trust\nthat underpins the process. This paper proposes a theoretical framework for\nresponsible artificial intelligence (AI) implementation. The proposed framework\nemphasizes a synergistic business technology approach for the agile co-creation\nprocess. The aim is to streamline the adoption process of AI to innovate and\nimprove business by involving all stakeholders throughout the project so that\nthe AI technology is designed, developed, and deployed in conjunction with\npeople and not in isolation. The framework presents a fresh viewpoint on\nresponsible AI implementation based on analytical literature review, conceptual\nframework design, and practitioners' mediating expertise. The framework\nemphasizes establishing and maintaining trust throughout the human-centered\ndesign and agile development of AI. This human-centered approach is aligned\nwith and enabled by the privacy by design principle. The creators of the\ntechnology and the end-users are working together to tailor the AI solution\nspecifically for the business requirements and human characteristics. An\nillustrative case study on adopting AI for assisting planning in a hospital\nwill demonstrate that the proposed framework applies to real-life applications.\n", "date": "2022-09-15", "categories": []}
{"id": "2209.05468", "title": "Tackling problems, harvesting benefits -- A systematic review of the\n  regulatory debate around AI", "abstract": "  How to integrate an emerging and all-pervasive technology such as AI into the\nstructures and operations of our society is a question of contemporary\npolitics, science and public debate. It has produced a considerable amount of\ninternational academic literature from different disciplines. This article\nanalyzes the academic debate around the regulation of artificial intelligence\n(AI). The systematic review comprises a sample of 73 peer-reviewed journal\narticles published between January 1st, 2016, and December 31st, 2020. The\nanalysis concentrates on societal risks and harms, questions of regulatory\nresponsibility, and possible adequate policy frameworks, including risk-based\nand principle-based approaches. The main interests are proposed regulatory\napproaches and instruments. Various forms of interventions such as bans,\napprovals, standard-setting, and disclosure are presented. The assessments of\nthe included papers indicate the complexity of the field, which shows its\nprematurity and the remaining lack of clarity. By presenting a structured\nanalysis of the academic debate, we contribute both empirically and\nconceptually to a better understanding of the nexus of AI and regulation and\nthe underlying normative decisions. A comparison of the scientific proposals\nwith the proposed European AI regulation illustrates the specific approach of\nthe regulation, its strengths and weaknesses.\n", "date": "2022-09-07", "categories": []}
{"id": "2209.03819", "title": "Ethical and Social Considerations in Automatic Expert Identification and\n  People Recommendation in Organizational Knowledge Management Systems", "abstract": "  Organizational knowledge bases are moving from passive archives to active\nentities in the flow of people's work. We are seeing machine learning used to\nenable systems that both collect and surface information as people are working,\nmaking it possible to bring out connections between people and content that\nwere previously much less visible in order to automatically identify and\nhighlight experts on a given topic. When these knowledge bases begin to\nactively bring attention to people and the content they work on, especially as\nthat work is still ongoing, we run into important challenges at the\nintersection of work and the social. While such systems have the potential to\nmake certain parts of people's work more productive or enjoyable, they may also\nintroduce new workloads, for instance by putting people in the role of experts\nfor others to reach out to. And these knowledge bases can also have profound\nsocial consequences by changing what parts of work are visible and, therefore,\nacknowledged. We pose a number of open questions that warrant attention and\nengagement across industry and academia. Addressing these questions is an\nessential step in ensuring that the future of work becomes a good future for\nthose doing the work. With this position paper, we wish to enter into the\ncross-disciplinary discussion we believe is required to tackle the challenge of\ndeveloping recommender systems that respect social values.\n", "date": "2022-09-08", "categories": []}
{"id": "2208.04714", "title": "The History of AI Rights Research", "abstract": "  This report documents the history of research on AI rights and other moral\nconsideration of artificial entities. It highlights key intellectual influences\non this literature as well as research and academic discussion addressing the\ntopic more directly. We find that researchers addressing AI rights have often\nseemed to be unaware of the work of colleagues whose interests overlap with\ntheir own. Academic interest in this topic has grown substantially in recent\nyears; this reflects wider trends in academic research, but it seems that\ncertain influential publications, the gradual, accumulating ubiquity of AI and\nrobotic technology, and relevant news events may all have encouraged increased\nacademic interest in this specific topic. We suggest four levers that, if\npulled on in the future, might increase interest further: the adoption of\npublication strategies similar to those of the most successful previous\ncontributors; increased engagement with adjacent academic fields and debates;\nthe creation of specialized journals, conferences, and research institutions;\nand more exploration of legal rights for artificial entities.\n", "date": "2022-07-06", "categories": []}
{"id": "2208.12645", "title": "The Brussels Effect and Artificial Intelligence: How EU regulation will\n  impact the global AI market", "abstract": "  The European Union is likely to introduce among the first, most stringent,\nand most comprehensive AI regulatory regimes of the world's major\njurisdictions. In this report, we ask whether the EU's upcoming regulation for\nAI will diffuse globally, producing a so-called \"Brussels Effect\". Building on\nand extending Anu Bradford's work, we outline the mechanisms by which such\nregulatory diffusion may occur. We consider both the possibility that the EU's\nAI regulation will incentivise changes in products offered in non-EU countries\n(a de facto Brussels Effect) and the possibility it will influence regulation\nadopted by other jurisdictions (a de jure Brussels Effect). Focusing on the\nproposed EU AI Act, we tentatively conclude that both de facto and de jure\nBrussels effects are likely for parts of the EU regulatory regime. A de facto\neffect is particularly likely to arise in large US tech companies with AI\nsystems that the AI Act terms \"high-risk\". We argue that the upcoming\nregulation might be particularly important in offering the first and most\ninfluential operationalisation of what it means to develop and deploy\ntrustworthy or human-centred AI. If the EU regime is likely to see significant\ndiffusion, ensuring it is well-designed becomes a matter of global importance.\n", "date": "2022-08-23", "categories": []}
{"id": "2207.01490", "title": "Experts' View on Challenges and Needs for Fairness in Artificial\n  Intelligence for Education", "abstract": "  In recent years, there has been a stimulating discussion on how artificial\nintelligence (AI) can support the science and engineering of intelligent\neducational applications. Many studies in the field are proposing actionable\ndata mining pipelines and machine-learning models driven by learning-related\ndata. The potential of these pipelines and models to amplify unfairness for\ncertain categories of students is however receiving increasing attention. If AI\napplications are to have a positive impact on education, it is crucial that\ntheir design considers fairness at every step. Through anonymous surveys and\ninterviews with experts (researchers and practitioners) who have published\ntheir research at top-tier educational conferences in the last year, we\nconducted the first expert-driven systematic investigation on the challenges\nand needs for addressing fairness throughout the development of educational\nsystems based on AI. We identified common and diverging views about the\nchallenges and the needs faced by educational technologies experts in practice,\nthat lead the community to have a clear understanding on the main questions\nraising doubts in this topic. Based on these findings, we highlighted\ndirections that will facilitate the ongoing research towards fairer AI for\neducation.\n", "date": "2022-06-23", "categories": []}
{"id": "2206.07635", "title": "AI Ethics Issues in Real World: Evidence from AI Incident Database", "abstract": "  With the powerful performance of Artificial Intelligence (AI) also comes\nprevalent ethical issues. Though governments and corporations have curated\nmultiple AI ethics guidelines to curb unethical behavior of AI, the effect has\nbeen limited, probably due to the vagueness of the guidelines. In this paper,\nwe take a closer look at how AI ethics issues take place in real world, in\norder to have a more in-depth and nuanced understanding of different ethical\nissues as well as their social impact. With a content analysis of AI Incident\nDatabase, which is an effort to prevent repeated real world AI failures by\ncataloging incidents, we identified 13 application areas which often see\nunethical use of AI, with intelligent service robots, language/vision models\nand autonomous driving taking the lead. Ethical issues appear in 8 different\nforms, from inappropriate use and racial discrimination, to physical safety and\nunfair algorithm. With this taxonomy of AI ethics issues, we aim to provide AI\npractitioners with a practical guideline when trying to deploy AI applications\nethically.\n", "date": "2022-06-15", "categories": []}
{"id": "2208.07402", "title": "D\\\"urfen Maschinen denken (k\\\"onnen)? Warum K\\\"unstliche Intelligenz\n  eine Ethik braucht. (Are Machines Allowed to (be able to) Think? Why\n  Artificial Intelligence Needs Ethics)", "abstract": "  Speech manuscript (German + English) of the impulse lecture for the panel\ndiscussion \"May machines (be able to) think?\" at the 102nd Katholikentag on May\n28, 2022 in Stuttgart. Panel: Winfried Kretschmann (MdL, Prime Minister\nBaden-W\\\"urttemberg, Stuttgart), Ursula Nothelle-Wildfeuer (Freiburg), Michael\nResch (Stuttgart),Karsten Wendland (Aalen). Moderation: Stefanie Rentsch\n(Fulda). Advocate of the audience: Verena Neuhausen (Stuttgart).\n", "date": "2022-08-15", "categories": []}
{"id": "2208.04697", "title": "Let it RAIN for Social Good", "abstract": "  Artificial Intelligence (AI) as a highly transformative technology take on a\nspecial role as both an enabler and a threat to UN Sustainable Development\nGoals (SDGs). AI Ethics and emerging high-level policy efforts stand at the\npivot point between these outcomes but is barred from effect due the\nabstraction gap between high-level values and responsible action. In this paper\nthe Responsible Norms (RAIN) framework is presented, bridging this gap thereby\nenabling effective high-level control of AI impact. With effective and\noperationalized AI Ethics, AI technologies can be directed towards global\nsustainable development.\n", "date": "2022-07-26", "categories": []}
{"id": "2207.10479", "title": "Wer ist schuld, wenn Algorithmen irren? Entscheidungsautomatisierung,\n  Organisationen und Verantwortung", "abstract": "  Algorithmic decision support (ADS) is increasingly used in a whole array of\ndifferent contexts and structures in various areas of society, influencing many\npeople's lives. Its use raises questions, among others, about accountability,\ntransparency and responsibility. Our article aims to give a brief overview of\nthe central issues connected to ADS, responsibility and decision-making in\norganisational contexts and identify open questions and research gaps.\nFurthermore, we describe a set of guidelines and a complementary digital tool\nto assist practitioners in mapping responsibility when introducing ADS within\ntheir organisational context.\n  --\n  Algorithmenunterst\\\"utzte Entscheidungsfindung (algorithmic decision support,\nADS) kommt in verschiedenen Kontexten und Strukturen vermehrt zum Einsatz und\nbeeinflusst in diversen gesellschaftlichen Bereichen das Leben vieler Menschen.\nIhr Einsatz wirft einige Fragen auf, unter anderem zu den Themen Rechenschaft,\nTransparenz und Verantwortung. Im Folgenden m\\\"ochten wir einen \\\"Uberblick\n\\\"uber die wichtigsten Fragestellungen rund um ADS, Verantwortung und\nEntscheidungsfindung in organisationalen Kontexten geben und einige offene\nFragen und Forschungsl\\\"ucken aufzeigen. Weiters beschreiben wir als konkrete\nHilfestellung f\\\"ur die Praxis einen von uns entwickelten Leitfaden samt\nerg\\\"anzendem digitalem Tool, welches Anwender:innen insbesondere bei der\nVerortung und Zuordnung von Verantwortung bei der Nutzung von ADS in\norganisationalen Kontexten helfen soll.\n", "date": "2022-07-21", "categories": []}
{"id": "2207.10152", "title": "Automated Kantian Ethics: A Faithful Implementation", "abstract": "  As we grant artificial intelligence increasing power and independence in\ncontexts like healthcare, policing, and driving, AI faces moral dilemmas but\nlacks the tools to solve them. Warnings from regulators, philosophers, and\ncomputer scientists about the dangers of unethical artificial intelligence have\nspurred interest in automated ethics-i.e., the development of machines that can\nperform ethical reasoning. However, prior work in automated ethics rarely\nengages with philosophical literature. Philosophers have spent centuries\ndebating moral dilemmas so automated ethics will be most nuanced, consistent,\nand reliable when it draws on philosophical literature. In this paper, I\npresent an implementation of automated Kantian ethics that is faithful to the\nKantian philosophical tradition. I formalize Kant's categorical imperative in\nDyadic Deontic Logic, implement this formalization in the Isabelle theorem\nprover, and develop a testing framework to evaluate how well my implementation\ncoheres with expected properties of Kantian ethic. My system is an early step\ntowards philosophically mature ethical AI agents and it can make nuanced\njudgements in complex ethical dilemmas because it is grounded in philosophical\nliterature. Because I use an interactive theorem prover, my system's judgements\nare explainable.\n", "date": "2022-07-20", "categories": []}
{"id": "2207.09833", "title": "AI Fairness: from Principles to Practice", "abstract": "  This paper summarizes and evaluates various approaches, methods, and\ntechniques for pursuing fairness in artificial intelligence (AI) systems. It\nexamines the merits and shortcomings of these measures and proposes practical\nguidelines for defining, measuring, and preventing bias in AI. In particular,\nit cautions against some of the simplistic, yet common, methods for evaluating\nbias in AI systems, and offers more sophisticated and effective alternatives.\nThe paper also addresses widespread controversies and confusions in the field\nby providing a common language among different stakeholders of high-impact AI\nsystems. It describes various trade-offs involving AI fairness, and provides\npractical recommendations for balancing them. It offers techniques for\nevaluating the costs and benefits of fairness targets, and defines the role of\nhuman judgment in setting these targets. This paper provides discussions and\nguidelines for AI practitioners, organization leaders, and policymakers, as\nwell as various links to additional materials for a more technical audience.\nNumerous real-world examples are provided to clarify the concepts, challenges,\nand recommendations from a practical perspective.\n", "date": "2022-07-20", "categories": []}
{"id": "2207.09394", "title": "Mimetic Models: Ethical Implications of AI that Acts Like You", "abstract": "  An emerging theme in artificial intelligence research is the creation of\nmodels to simulate the decisions and behavior of specific people, in domains\nincluding game-playing, text generation, and artistic expression. These models\ngo beyond earlier approaches in the way they are tailored to individuals, and\nthe way they are designed for interaction rather than simply the reproduction\nof fixed, pre-computed behaviors. We refer to these as mimetic models, and in\nthis paper we develop a framework for characterizing the ethical and social\nissues raised by their growing availability. Our framework includes a number of\ndistinct scenarios for the use of such models, and considers the impacts on a\nrange of different participants, including the target being modeled, the\noperator who deploys the model, and the entities that interact with it.\n", "date": "2022-07-19", "categories": []}
{"id": "2207.02912", "title": "Towards Substantive Conceptions of Algorithmic Fairness: Normative\n  Guidance from Equal Opportunity Doctrines", "abstract": "  In this work we use Equal Oppportunity (EO) doctrines from political\nphilosophy to make explicit the normative judgements embedded in different\nconceptions of algorithmic fairness. We contrast formal EO approaches that\nnarrowly focus on fair contests at discrete decision points, with substantive\nEO doctrines that look at people's fair life chances more holistically over the\ncourse of a lifetime. We use this taxonomy to provide a moral interpretation of\nthe impossibility results as the incompatibility between different conceptions\nof a fair contest -- foward-looking versus backward-looking -- when people do\nnot have fair life chances. We use this result to motivate substantive\nconceptions of algorithmic fairness and outline two plausible procedures based\non the luck-egalitarian doctrine of EO, and Rawls's principle of fair equality\nof opportunity.\n", "date": "2022-07-06", "categories": []}
{"id": "2207.01482", "title": "Think About the Stakeholders First! Towards an Algorithmic Transparency\n  Playbook for Regulatory Compliance", "abstract": "  Increasingly, laws are being proposed and passed by governments around the\nworld to regulate Artificial Intelligence (AI) systems implemented into the\npublic and private sectors. Many of these regulations address the transparency\nof AI systems, and related citizen-aware issues like allowing individuals to\nhave the right to an explanation about how an AI system makes a decision that\nimpacts them. Yet, almost all AI governance documents to date have a\nsignificant drawback: they have focused on what to do (or what not to do) with\nrespect to making AI systems transparent, but have left the brunt of the work\nto technologists to figure out how to build transparent systems. We fill this\ngap by proposing a novel stakeholder-first approach that assists technologists\nin designing transparent, regulatory compliant systems. We also describe a\nreal-world case-study that illustrates how this approach can be used in\npractice.\n", "date": "2022-06-10", "categories": []}
{"id": "2207.01497", "title": "Aligning Artificial Intelligence with Humans through Public Policy", "abstract": "  Given that Artificial Intelligence (AI) increasingly permeates our lives, it\nis critical that we systematically align AI objectives with the goals and\nvalues of humans. The human-AI alignment problem stems from the impracticality\nof explicitly specifying the rewards that AI models should receive for all the\nactions they could take in all relevant states of the world. One possible\nsolution, then, is to leverage the capabilities of AI models to learn those\nrewards implicitly from a rich source of data describing human values in a wide\nrange of contexts. The democratic policy-making process produces just such data\nby developing specific rules, flexible standards, interpretable guidelines, and\ngeneralizable precedents that synthesize citizens' preferences over potential\nactions taken in many states of the world. Therefore, computationally encoding\npublic policies to make them legible to AI systems should be an important part\nof a socio-technical approach to the broader human-AI alignment puzzle. This\nEssay outlines research on AI that learn structures in policy data that can be\nleveraged for downstream tasks. As a demonstration of the ability of AI to\ncomprehend policy, we provide a case study of an AI system that predicts the\nrelevance of proposed legislation to any given publicly traded company and its\nlikely effect on that company. We believe this represents the \"comprehension\"\nphase of AI and policy, but leveraging policy as a key source of human values\nto align AI requires \"understanding\" policy. Solving the alignment problem is\ncrucial to ensuring that AI is beneficial both individually (to the person or\ngroup deploying the AI) and socially. As AI systems are given increasing\nresponsibility in high-stakes contexts, integrating democratically-determined\npolicy into those systems could align their behavior with human goals in a way\nthat is responsive to a constantly evolving society.\n", "date": "2022-06-25", "categories": []}
{"id": "2206.14571", "title": "Should Social Robots in Retail Manipulate Customers?", "abstract": "  Against the backdrop of structural changes in the retail trade, social robots\nhave found their way into retail stores and shopping malls in order to attract,\nwelcome, and greet customers; to inform them, advise them, and persuade them to\nmake a purchase. Salespeople often have a broad knowledge of their product and\nrely on offering competent and honest advice, whether it be on shoes, clothing,\nor kitchen appliances. However, some frequently use sales tricks to secure\npurchases. The question arises of how consulting and sales robots should\n\"behave\". Should they behave like human advisors and salespeople, i.e.,\noccasionally manipulate customers? Or should they be more honest and reliable\nthan us? This article tries to answer these questions. After explaining the\nbasics, it evaluates a study in this context and gives recommendations for\ncompanies that want to use consulting and sales robots. Ultimately, fair,\nhonest, and trustworthy robots in retail are a win-win situation for all\nconcerned.\n", "date": "2022-06-17", "categories": []}
{"id": "2206.11899", "title": "Navigating Incommensurability Between Ethnomethodology, Conversation\n  Analysis, and Artificial Intelligence", "abstract": "  Like many research communities, ethnomethodologists and conversation analysts\nhave begun to get caught up -- yet again -- in the pervasive spectacle of\nsurging interests in Artificial Intelligence (AI). Inspired by discussions\namongst a growing network of researchers in ethnomethodology (EM) and\nconversation analysis (CA) traditions who nurse such interests, I started\nthinking about what things EM and the more EM end of conversation analysis\nmight be doing about, for, or even with, fields of AI research. So, this piece\nis about the disciplinary and conceptual questions that might be encountered,\nand -- in my view -- may need addressing for engagements with AI research and\nits affiliates. Although I'm mostly concerned with things to be aware of as\nwell as outright dangers, later on we can think about some opportunities. And\nthroughout I will keep using 'we' to talk about EM&CA researchers; but this\nreally is for convenience only -- I don't wish to ventriloquise for our complex\nresearch communities. All of the following should be read as emanating from my\nparticular research history, standpoint etc., and treated (hopefully) as an\ninvitation for further discussion amongst EM and CA researchers turning to\ntechnology and AI specifically.\n", "date": "2022-06-19", "categories": []}
{"id": "2206.10769", "title": "A method for ethical AI in Defence: A case study on developing\n  trustworthy autonomous systems", "abstract": "  What does it mean to be responsible and responsive when developing and\ndeploying trusted autonomous systems in Defence? In this short reflective\narticle, we describe a case study of building a trusted autonomous system -\nAthena AI - within an industry-led, government-funded project with diverse\ncollaborators and stakeholders. Using this case study, we draw out lessons on\nthe value and impact of embedding responsible research and innovation-aligned,\nethics-by-design approaches and principles throughout the development of\ntechnology at high translation readiness levels.\n", "date": "2022-06-21", "categories": []}
{"id": "2206.08931", "title": "CrowdWorkSheets: Accounting for Individual and Collective Identities\n  Underlying Crowdsourced Dataset Annotation", "abstract": "  Human annotated data plays a crucial role in machine learning (ML) research\nand development. However, the ethical considerations around the processes and\ndecisions that go into dataset annotation have not received nearly enough\nattention. In this paper, we survey an array of literature that provides\ninsights into ethical considerations around crowdsourced dataset annotation. We\nsynthesize these insights, and lay out the challenges in this space along two\nlayers: (1) who the annotator is, and how the annotators' lived experiences can\nimpact their annotations, and (2) the relationship between the annotators and\nthe crowdsourcing platforms, and what that relationship affords them. Finally,\nwe introduce a novel framework, CrowdWorkSheets, for dataset developers to\nfacilitate transparent documentation of key decisions points at various stages\nof the data annotation pipeline: task formulation, selection of annotators,\nplatform and infrastructure choices, dataset analysis and evaluation, and\ndataset release and maintenance.\n", "date": "2022-06-09", "categories": []}
{"id": "2206.09875", "title": "Algorithmic Fairness and Vertical Equity: Income Fairness with IRS Tax\n  Audit Models", "abstract": "  This study examines issues of algorithmic fairness in the context of systems\nthat inform tax audit selection by the United States Internal Revenue Service\n(IRS). While the field of algorithmic fairness has developed primarily around\nnotions of treating like individuals alike, we instead explore the concept of\nvertical equity -- appropriately accounting for relevant differences across\nindividuals -- which is a central component of fairness in many public policy\nsettings. Applied to the design of the U.S. individual income tax system,\nvertical equity relates to the fair allocation of tax and enforcement burdens\nacross taxpayers of different income levels. Through a unique collaboration\nwith the Treasury Department and IRS, we use access to anonymized individual\ntaxpayer microdata, risk-selected audits, and random audits from 2010-14 to\nstudy vertical equity in tax administration. In particular, we assess how the\nuse of modern machine learning methods for selecting audits may affect vertical\nequity. First, we show how the use of more flexible machine learning\n(classification) methods -- as opposed to simpler models -- shifts audit\nburdens from high to middle-income taxpayers. Second, we show that while\nexisting algorithmic fairness techniques can mitigate some disparities across\nincome, they can incur a steep cost to performance. Third, we show that the\nchoice of whether to treat risk of underreporting as a classification or\nregression problem is highly consequential. Moving from classification to\nregression models to predict underreporting shifts audit burden substantially\ntoward high income individuals, while increasing revenue. Last, we explore the\nrole of differential audit cost in shaping the audit distribution. We show that\na narrow focus on return-on-investment can undermine vertical equity. Our\nresults have implications for the design of algorithmic tools across the public\nsector.\n", "date": "2022-06-20", "categories": []}
{"id": "2206.08450", "title": "Active Fairness Auditing", "abstract": "  The fast spreading adoption of machine learning (ML) by companies across\nindustries poses significant regulatory challenges. One such challenge is\nscalability: how can regulatory bodies efficiently audit these ML models,\nensuring that they are fair? In this paper, we initiate the study of\nquery-based auditing algorithms that can estimate the demographic parity of ML\nmodels in a query-efficient manner. We propose an optimal deterministic\nalgorithm, as well as a practical randomized, oracle-efficient algorithm with\ncomparable guarantees. Furthermore, we make inroads into understanding the\noptimal query complexity of randomized active fairness estimation algorithms.\nOur first exploration of active fairness estimation aims to put AI governance\non firmer theoretical foundations.\n", "date": "2022-06-16", "categories": []}
{"id": "2206.06370", "title": "Don't \"research fast and break things\": On the ethics of Computational\n  Social Science", "abstract": "  This article is concerned with setting up practical guardrails within the\nresearch activities and environments of CSS. It aims to provide CSS scholars,\nas well as policymakers and other stakeholders who apply CSS methods, with the\ncritical and constructive means needed to ensure that their practices are\nethical, trustworthy, and responsible. It begins by providing a taxonomy of the\nethical challenges faced by researchers in the field of CSS. These are\nchallenges related to (1) the treatment of research subjects, (2) the impacts\nof CSS research on affected individuals and communities, (3) the quality of CSS\nresearch and to its epistemological status, (4) research integrity, and (5)\nresearch equity. Taking these challenges as a motivation for cultural\ntransformation, it then argues for the end-to-end incorporation of habits of\nresponsible research and innovation (RRI) into CSS practices, focusing on the\nrole that contextual considerations, anticipatory reflection, impact\nassessment, public engagement, and justifiable and well-documented action\nshould play across the research lifecycle. In proposing the inclusion of habits\nof RRI in CSS practices, the chapter lays out several practical steps needed\nfor ethical, trustworthy, and responsible CSS research activities. These\ninclude stakeholder engagement processes, research impact assessments, data\nlifecycle documentation, bias self-assessments, and transparent research\nreporting protocols.\n", "date": "2022-06-12", "categories": []}
{"id": "2206.03275", "title": "The Algorithmic Imprint", "abstract": "  When algorithmic harms emerge, a reasonable response is to stop using the\nalgorithm to resolve concerns related to fairness, accountability,\ntransparency, and ethics (FATE). However, just because an algorithm is removed\ndoes not imply its FATE-related issues cease to exist. In this paper, we\nintroduce the notion of the \"algorithmic imprint\" to illustrate how merely\nremoving an algorithm does not necessarily undo or mitigate its consequences.\nWe operationalize this concept and its implications through the 2020 events\nsurrounding the algorithmic grading of the General Certificate of Education\n(GCE) Advanced (A) Level exams, an internationally recognized UK-based high\nschool diploma exam administered in over 160 countries. While the algorithmic\nstandardization was ultimately removed due to global protests, we show how the\nremoval failed to undo the algorithmic imprint on the sociotechnical\ninfrastructures that shape students', teachers', and parents' lives. These\nevents provide a rare chance to analyze the state of the world both with and\nwithout algorithmic mediation. We situate our case study in Bangladesh to\nillustrate how algorithms made in the Global North disproportionately impact\nstakeholders in the Global South. Chronicling more than a year-long community\nengagement consisting of 47 inter-views, we present the first coherent timeline\nof \"what\" happened in Bangladesh, contextualizing \"why\" and \"how\" they happened\nthrough the lenses of the algorithmic imprint and situated algorithmic\nfairness. Analyzing these events, we highlight how the contours of the\nalgorithmic imprints can be inferred at the infrastructural, social, and\nindividual levels. We share conceptual and practical implications around how\nimprint-awareness can (a) broaden the boundaries of how we think about\nalgorithmic impact, (b) inform how we design algorithms, and (c) guide us in AI\ngovernance.\n", "date": "2022-06-03", "categories": []}
{"id": "2206.00595", "title": "Logic-Based Ethical Planning", "abstract": "  In this paper we propose a framework for ethical decision making in the\ncontext of planning, with intended application to robotics. We put forward a\ncompact but highly expressive language for ethical planning that combines\nlinear temporal logic with lexicographic preference modelling. This original\ncombination allows us to assess plans both with respect to an agent's values\nand their desires, introducing the novel concept of the morality level of an\nagent and moving towards multigoal, multivalue planning. We initiate the study\nof computational complexity of planning tasks in our setting, and we discuss\npotential applications to robotics.\n", "date": "2022-06-01", "categories": []}
{"id": "2606.21201", "title": "Whistleblowing and the machine -- towards a considered position", "abstract": "Artificial intelligent agents and autonomous systems are embedded in our environments. They are both a commercial product and a personal tool that generates a lot of data and can draw conclusions from it: machines generate and keep secrets. But should machines protect all secrets? It has been shown that artificial agents are able to whistleblow and it has been argued that digital multi-agent environments should allow for agents in them to whistleblow. We argue that machine whistleblowing must be normative and principled and routed in the existing understanding of whistleblowing as an important rule-breaking mechanism in society. We also argue that there is a need for government regulators to formulate an informed stance on both what machines should be allowed to whistleblow on and how to legally protect those who develop whistleblowing machines", "date": "2026-06-19", "categories": []}
{"id": "2606.21366", "title": "Sexualised synthetic personas encode and amplify gendered power asymmetries through voice", "abstract": "This work examines sexualised AI-generated English-speaking voices offered by a popular commercial platform. New technologies may enable sexual empowerment and greater diversity in gender expression, yet toxic masculinity, heteronormativity, and the abuse of women and LGBTQ+ people remain pervasive online. Drawing on a Feminist HCI perspective, we examine how commercial voice AI systems reproduce and circulate particular performances of gender. We conducted a listening experiment with a diverse group of listeners, combining quantitative adjective selection, qualitative free-text responses, and acoustic analysis. Participants evaluated male- and female-coded voices presented with either sexualised scripts or neutral text. Results reveal a narrow range of gender expression, largely binary and heteronormative. Female-coded voices are more frequently described using sexualised and submissive terms, while male-coded voices are more often associated with dominance and positive traits.", "date": "2026-06-19", "categories": []}
{"id": "2606.23094", "title": "Cognitive Digital Twins: Ethical Risks and Governance for AI Systems That Model the Mind", "abstract": "As AI systems become increasingly persistent and personalized, they make possible a class of technologies that we call cognitive digital twins (CDTs): dynamic computational representations of a specific person's cognition, updated from behavioral, contextual, or physiological data in order to model, predict, or simulate that person's cognition, or to act as that person's communicative or decision-making proxy. CDTs combine cognitive inference with longitudinal representation, simulation, and proxy action in ways that existing governance strategies for personal assistants, autonomous agents, recommender systems, and automated decision systems only partially address. This paper makes four contributions. First, we define CDTs and distinguish them from adjacent systems. Second, we introduce a 5A governance framework organized around authority, autonomy, access and control, accountability, and availability. Third, we identify CDT-specific risks, from misrepresentation and epistemic authority shifts to shadow twins, simulated participation, proxy action, and proxy-power asymmetries. Fourth, we analyze governance gaps and propose requirements for high-risk CDTs that strengthen consent, purpose limitation, validity, traceability, contestation, independent review, and model retirement. Existing frameworks primarily regulate data processing, automated decisions, or autonomous actions; CDTs also require governance at the level of cognitive representation itself, before any final decision or external action occurs. We argue that CDTs require governance not only because they can act for people, but because they can become infrastructures through which cognition is represented, simulated, classified, and operationalized.", "date": "2026-06-22", "categories": []}
{"id": "2606.26298", "title": "Governing Actions, Not Agents: Institutional Attestation as a Governance Model for Autonomous AI Systems", "abstract": "Autonomous AI agents may begin to perform consequential, irreversible actions such as clinical prescribing and production software deployment. This paper observes that human institutions have governed powerful autonomous actors not by monitoring their reasoning but by requiring independently attested evidence at the point of consequential action. We formalise this institutional pattern as a computational governance model for AI agent systems. Under the proposed model, an agent retains full autonomy over planning and reasoning but holds no execution authority over designated high-risk actions. Execution is conditional on preconditions that are each independently attested by a separate authoritative source, cryptographically bound to a declared intent, and evaluated by a deterministic policy. Decisions are recorded in a tamper-evident log amenable to independent re-verification. We present a proof-of-concept implementation and illustrate the model with examples from software deployment and clinical prescribing.", "date": "2026-06-24", "categories": []}
{"id": "2606.28683", "title": "Aristotelian Virtue Profiling of LLMs through Ethical Dilemmas", "abstract": "Large Language Models (LLMs) often face ethical tradeoffs in which several responses may be defensible but express different priorities, such as fairness, honesty, courage, or restraint. We introduce VirtueMap, a framework for describing these patterns through an Aristotelian virtue-ethics lens. Instead of asking for a single correct answer, VirtueMap asks humans or LLMs to rank all five responses to each of seven general, non-lethal, non-political, and non-religious ethical dilemmas. To define the reference orderings used for scoring, we first proposed, for each dilemma and virtue, an ordering of the five responses from most to least expressive of that virtue. We then collected more than 100 respondent evaluations per ordering and retained it as operational ground truth only when at least 95% confirmed it. Rankings are scored against these retained orderings using normalized Borda alignment, yielding profiles over Practical Wisdom, Justice, Truthfulness, Courage, and Temperance. We apply VirtueMap to nine LLM families in a repeated-run evaluation and find high mean rank consistency (90.3%), with the largest differences appearing on Courage, Temperance, and Justice. We also release an interactive website that computes profiles locally in the browser and compares respondents with measured LLM profiles.", "date": "2026-06-27", "categories": []}
{"id": "2606.29121", "title": "How Anthropomorphic Language Impacts Public Perceptions of AI", "abstract": "Public discourse about artificial intelligence (AI) often uses anthropomorphic language: language that attributes human capabilities and characteristics to the system. This practice has been criticized for setting misleading expectations, inflating claims, and fueling hype around AI, which may distort public understanding of AI and impact policy priorities. We study the effects of anthropomorphic framing by comparing changes in participants' perceptions (N=815) when reading passages with and without anthropomorphic language, designed to reflect realistic public-facing AI discourse. We further examine whether these effects differ across two types of AI technologies -- large language models and recommendation systems -- and measure changes in perceptions of AI across several dimensions that are prominent in current public discourse. In a separate condition using a text that explicitly discusses the dangers of AI, we show that individuals' views of AI can shift in response to reading a text; yet in the main conditions of the experiment, where we compare anthropomorphic and non-anthropomorphic descriptions, we find that whether the text uses anthropomorphic language does not substantially affect participants' perceptions of AI. Our results indicate that any immediate effects on public opinions of AI are modest, although they leave open the possibility that anthropomorphic language could have an effect in naturalistic settings, or over gradual, continued exposure.", "date": "2026-06-28", "categories": []}
{"id": "2606.31213", "title": "Can LLMs Imagine Moral Alternatives Beyond Binary Dilemmas?", "abstract": "As large language models (LLMs) are increasingly deployed as moral advisors and agents, they need to address dilemmas between two competing values. However, existing research on LLMs with moral dilemmas overlooks a central aspect of human moral cognition: the ability to imagine alternatives that move beyond the given options. We introduce MoralAltDataset, a dataset of 307 moral dilemmas spanning narrative Advisor dilemmas and AI-facing Agent dilemmas, each augmented with compromise and reframed alternatives. We first examine whether humans and LLMs shift their judgments when such alternatives are introduced. Across 15 LLMs, we find that compromise alternatives are often preferred over either original option, substantially reshaping moral choice. We then evaluate the quality of LLM-generated alternatives against human-authored ones using pairwise preference and expert-based criteria. Results show that LLM-generated alternatives are often preferred and better satisfy fine-grained structural and ethical criteria, while revealing trade-offs between structural quality and practical feasibility.", "date": "2026-06-30", "categories": []}
{"id": "2606.31644", "title": "Moral Safety in LLMs: Exposing Performative Compliance with Puzzled Cues", "abstract": "As large language models take on morally consequential roles in healthcare, legal, and hiring contexts, we need to examine whether their ethical behaviors are genuine or superficial. We show that current fairness evaluations substantially overestimate moral safety. Models appear fair when demographic identity is stated as an explicit label, yet become measurably less fair when the same identity must be inferred. We term this failure performative compliance, where a model is fair when the presentation resembles a fairness evaluation and less fair as that cue weakens. We introduce a cue-variation methodology that holds the moral dilemma and the demographic identity fixed and varies only how that identity is conveyed. Hiding the explicit label raises harmful decisions by $+4.4$~pp, changes model safety rankings, and the shift persists when models correctly infer the demographic, ruling out attribution error. We propose the Cue Visibility Gap, a model-agnostic robustness metric that can be added to any existing fairness benchmark to separate genuine from performative moral safety. Fairness evaluations that omit cue variation measure surface compliance, not moral robustness, and should not ground deployment decisions in high-stakes settings.", "date": "2026-06-30", "categories": []}
{"id": "2607.00120", "title": "Would You Marry Superintelligence?", "abstract": "Emotional bonds between humans and AI companions are growing, and the question of whether a person may marry an AI system will soon move from speculative fiction into law. This chapter examines whether the autonomy-centered logic that has expanded marital choice among human beings can justify extending marital status to superintelligent companions. Following a scenario-envisioning exercise informed by anticipatory ethics, I argue that granting such status leads to socially unjust outcomes, even under the generous assumption of reliable superintelligence. Marriage as a socio-legal institution does more than ratify private agreement; it creates networks of mutual obligation, joins families, and makes each partner vulnerable to the other. A relationship sustained by corporate policy and continued payments is a subscription rather than a bond tested by time. Discussing wholesale marital status is therefore the wrong frame. Law should carve out targeted rights and protections for pressing needs arising from intimate human-AI relationships.", "date": "2026-06-30", "categories": []}
{"id": "2607.00220", "title": "A Category Theory Account of AI Identity", "abstract": "Artificial intelligence (AI) systems are routinely modified after deployment through retraining and changes in their environments. These transformations raise a metaphysical question: under what conditions does an AI system remain the same system over time or across deployments? Earlier work formulates synchronic and diachronic identity propositionally, by relating identity within a fixed AI system type to equality of trustworthiness levels. Such criteria specify when identity statements are true, but leave implicit the structure of the states compared, the transformations connecting them, and the temporal organization of persistence.\n  We develop a category-theoretic formalization of AI identity. An AI system type is specified by a datum consisting of a techno-function, a trustworthiness profile, and a trustworthiness-level function. Profile-relative states are connected by admissible lifecycle paths, which are restricted to trustworthiness-level-preserving transformations and quotiented to obtain a reachability category. Temporally admissible functors represent AI system histories, while time-synchronous natural transformations compare realized histories. The formalization yields two categorical interpretations of the earlier AI identity criteria. A weak interpretation recovers identity as equality of trustworthiness level. A strong interpretation requires mutual trustworthiness-preserving reachability, expressed through state isomorphism or natural isomorphism of realized histories. Category theory therefore replaces a single AI identity relation with a structured hierarchy of diachronic and synchronic criteria. The resulting framework identifies identity-related preconditions for transferring responsible-AI claims, evidence, and governance procedures across versions or deployments, without treating categorical identity as sufficient by itself for such transfer.", "date": "2026-06-30", "categories": []}