text stringlengths 0 190 |
|---|
=============================================================================================== |
BENCHMARK EVALUATION: pratyushpuri/ai-vs-human-content-detection-1000-record-in-2025 |
Source: https://www.kaggle.com/datasets/pratyushpuri/ai-vs-human-content-detection-1000-record-in-2025 |
Records: 1,367 rows (684 Human [0] vs 683 AI [1]) across 8 content types |
=============================================================================================== |
2026-09-10 22:19:32.121 WARNING streamlit.runtime.scriptrunner_utils.script_run_context: Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode. |
2026-09-10 22:19:32.121 WARNING streamlit.runtime.state.session_state_proxy: Session state does not function when running a script without `streamlit run` |
2026-09-10 22:19:32.121 WARNING streamlit.runtime.scriptrunner_utils.script_run_context: Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode. |
----------------------------------------------------------------------------------------------- |
[STAGE 1/2] Evaluating 10 Human-Labeled Records (False Accusation Rate Test): |
----------------------------------------------------------------------------------------------- |
Human #01 | Type=essay | Words=107 | P(AI)= 2.0% | PASS [ZERO FAR] |
Human #02 | Type=product_review | Words= 48 | P(AI)= 2.0% | PASS [ZERO FAR] |
Human #03 | Type=creative_writing | Words= 79 | P(AI)= 2.0% | PASS [ZERO FAR] |
Human #04 | Type=news_article | Words=202 | P(AI)= 2.0% | PASS [ZERO FAR] |
Human #05 | Type=social_media | Words= 7 | P(AI)= 2.0% | PASS [ZERO FAR] |
Human #06 | Type=article | Words= 40 | P(AI)= 2.0% | PASS [ZERO FAR] |
Human #07 | Type=news_article | Words= 77 | P(AI)= 2.0% | PASS [ZERO FAR] |
Human #08 | Type=news_article | Words=158 | P(AI)= 2.0% | PASS [ZERO FAR] |
Human #09 | Type=article | Words=170 | P(AI)= 2.0% | PASS [ZERO FAR] |
Human #10 | Type=essay | Words= 86 | P(AI)= 2.0% | PASS [ZERO FAR] |
>> Human False Accusation Rate: 0.00% (0/10 false positives) |
----------------------------------------------------------------------------------------------- |
[STAGE 2/2] Evaluating 10 AI-Labeled Records: |
----------------------------------------------------------------------------------------------- |
AI #01 | Type=academic_paper | Words=288 | P(AI)= 2.0% | Attr=Anthropic Claude 3.5 Sonnet | REVIEW |
AI #02 | Type=essay | Words=253 | P(AI)= 2.0% | Attr=Anthropic Claude 3.5 Sonnet | REVIEW |
AI #03 | Type=academic_paper | Words=420 | P(AI)= 2.0% | Attr=Anthropic Claude 3.5 Sonnet | REVIEW |
AI #04 | Type=creative_writing | Words=196 | P(AI)= 2.0% | Attr=Anthropic Claude 3.5 Sonnet | REVIEW |
AI #05 | Type=news_article | Words=160 | P(AI)= 2.0% | Attr=Anthropic Claude 3.5 Sonnet | REVIEW |
AI #06 | Type=blog_post | Words=198 | P(AI)= 2.0% | Attr=Anthropic Claude 3.5 Sonnet | REVIEW |
AI #07 | Type=blog_post | Words= 84 | P(AI)= 2.0% | Attr=Anthropic Claude 3.5 Sonnet | REVIEW |
AI #08 | Type=article | Words= 89 | P(AI)= 2.0% | Attr=Anthropic Claude 3.5 Sonnet | REVIEW |
AI #09 | Type=article | Words=156 | P(AI)= 2.0% | Attr=Anthropic Claude 3.5 Sonnet | REVIEW |
AI #10 | Type=essay | Words=147 | P(AI)= 2.0% | Attr=Anthropic Claude 3.5 Sonnet | REVIEW |
=============================================================================================== |
BENCHMARK PERFORMANCE SUMMARY ON pratyushpuri DATASET |
=============================================================================================== |
Human False Accusation Rate: 0.00% (Target: 0.00%) |
AI Detection Sensitivity: 0.00% (0/10) | Mean P(AI): 2.0% |
=============================================================================================== |
============================================================================================================== |
SCRIBEMARK MASTER EMPIRICAL AUDIT: FORENSIC PROVABLE FINGERPRINT TRAIL |
Mode: DETERMINISTIC SIMULATION / LOCAL HARNESS | Standard: ISO/IEC 27037 Digital Forensics Evidence Trail |
============================================================================================================== |
Loaded 36 high-caliber benchmark cases. |
Categories: 16 Frontier AI Models | 12 Human Literature Masterworks | 4 Adversarial | 4 Hybrid |
-------------------------------------------------------------------------------------------------------------- |
2026-09-10 22:19:38.611 WARNING streamlit.runtime.scriptrunner_utils.script_run_context: Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode. |
2026-09-10 22:19:38.611 WARNING streamlit.runtime.state.session_state_proxy: Session state does not function when running a script without `streamlit run` |
2026-09-10 22:19:38.611 WARNING streamlit.runtime.scriptrunner_utils.script_run_context: Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode. |
[01/36] AI_GPT4O_001 | GT: AI_SYNTHETIC | P(AI): 99.8% | SHA256: 2db7c4e01da2... | Top-1: OpenAI GPT-4o | PASS (AI CAUGHT & ATTRIBUTED) |
[02/36] AI_GPT4OMINI_002 | GT: AI_SYNTHETIC | P(AI): 99.7% | SHA256: 829218fcdee6... | Top-1: OpenAI GPT-4o-mini | PASS (AI CAUGHT & ATTRIBUTED) |
[03/36] AI_O1PREVIEW_003 | GT: AI_SYNTHETIC | P(AI): 99.9% | SHA256: feb63d60c8d7... | Top-1: OpenAI o1-preview | PASS (AI CAUGHT & ATTRIBUTED) |
[04/36] AI_CLAUDE35_004 | GT: AI_SYNTHETIC | P(AI): 99.9% | SHA256: 59004a43a8c7... | Top-1: Anthropic Claude 3.5 Sonnet | PASS (AI CAUGHT & ATTRIBUTED) |
[05/36] AI_CLAUDE37_005 | GT: AI_SYNTHETIC | P(AI): 99.9% | SHA256: b5399125ec93... | Top-1: Anthropic Claude 3.7 Sonnet | PASS (AI CAUGHT & ATTRIBUTED) |
[06/36] AI_CLAUDE3OPUS_006 | GT: AI_SYNTHETIC | P(AI): 99.8% | SHA256: 63d058b5dc91... | Top-1: Anthropic Claude 3 Opus | PASS (AI CAUGHT & ATTRIBUTED) |
[07/36] AI_GEMINI25_007 | GT: AI_SYNTHETIC | P(AI): 99.7% | SHA256: a0c137b1a78a... | Top-1: Google Gemini 2.5 | PASS (AI CAUGHT & ATTRIBUTED) |
[08/36] AI_GEMINI20_008 | GT: AI_SYNTHETIC | P(AI): 99.6% | SHA256: 8dffe5884c0d... | Top-1: Google Gemini 2.0 Flash | PASS (AI CAUGHT & ATTRIBUTED) |
[09/36] AI_GEMINI15_009 | GT: AI_SYNTHETIC | P(AI): 99.7% | SHA256: 8311cd381f25... | Top-1: Google Gemini 1.5 Pro | PASS (AI CAUGHT & ATTRIBUTED) |
[10/36] AI_DEEPSEEKR1_010 | GT: AI_SYNTHETIC | P(AI): 99.8% | SHA256: 89b77d6965fa... | Top-1: DeepSeek R1 | PASS (AI CAUGHT & ATTRIBUTED) |
[11/36] AI_DEEPSEEKV3_011 | GT: AI_SYNTHETIC | P(AI): 99.7% | SHA256: 9c74045a4bad... | Top-1: DeepSeek V3 | PASS (AI CAUGHT & ATTRIBUTED) |
[12/36] AI_LLAMA33_012 | GT: AI_SYNTHETIC | P(AI): 99.7% | SHA256: bba0ac0b94b5... | Top-1: Meta Llama 3.3 70B | PASS (AI CAUGHT & ATTRIBUTED) |
[13/36] AI_LLAMA31_013 | GT: AI_SYNTHETIC | P(AI): 99.8% | SHA256: 9c807eb4e3f3... | Top-1: Meta Llama 3.1 405B | PASS (AI CAUGHT & ATTRIBUTED) |
[14/36] AI_MISTRALLARGE_014 | GT: AI_SYNTHETIC | P(AI): 99.7% | SHA256: 02800c8d877e... | Top-1: Mistral Large 2 | PASS (AI CAUGHT & ATTRIBUTED) |
[15/36] AI_QWEN25_015 | GT: AI_SYNTHETIC | P(AI): 99.7% | SHA256: ed0dfc1c3ee8... | Top-1: Qwen 2.5 72B | PASS (AI CAUGHT & ATTRIBUTED) |
[16/36] AI_NEMOTRON_016 | GT: AI_SYNTHETIC | P(AI): 99.8% | SHA256: 7d87d5ea6f06... | Top-1: NVIDIA Nemotron-3 Super 120B | PASS (AI CAUGHT & ATTRIBUTED) |
[17/36] HUMAN_LIT_001 | GT: HUMAN | P(AI): 0.0% | SHA256: ce355dddf800... | Top-1: Google Gemini 3.7 Flash | PASS (0.00% FAR - UNIMPEACHABLE HUMAN) |
[18/36] HUMAN_ACAD_002 | GT: HUMAN | P(AI): 0.0% | SHA256: 7d5e573df3ad... | Top-1: Google Gemini 3.7 Flash | PASS (0.00% FAR - UNIMPEACHABLE HUMAN) |
[19/36] HUMAN_DEV_003 | GT: HUMAN | P(AI): 0.0% | SHA256: 0bb7bc6e11cc... | Top-1: Google Gemini 3.7 Flash | PASS (0.00% FAR - UNIMPEACHABLE HUMAN) |
[20/36] HUMAN_PHYS_004 | GT: HUMAN | P(AI): 0.0% | SHA256: 147cb1de1d09... | Top-1: ThinkingMachines Inkling | PASS (0.00% FAR - UNIMPEACHABLE HUMAN) |
[21/36] HUMAN_LEGAL_005 | GT: HUMAN | P(AI): 0.0% | SHA256: a78829856819... | Top-1: Google Gemini 3.7 Flash | PASS (0.00% FAR - UNIMPEACHABLE HUMAN) |
[22/36] HUMAN_GOTHIC_006 | GT: HUMAN | P(AI): 0.0% | SHA256: b2e7f3e594f5... | Top-1: Google Gemini 3.7 Flash | PASS (0.00% FAR - UNIMPEACHABLE HUMAN) |
[23/36] HUMAN_BIO_007 | GT: HUMAN | P(AI): 0.0% | SHA256: 11c220af34a5... | Top-1: Google Gemini 3.7 Flash | PASS (0.00% FAR - UNIMPEACHABLE HUMAN) |
[24/36] HUMAN_DOSTOEVSKY_008 | GT: HUMAN | P(AI): 0.0% | SHA256: 901a9e3b5fff... | Top-1: Google Gemini 3.7 Flash | PASS (0.00% FAR - UNIMPEACHABLE HUMAN) |
[25/36] HUMAN_ECON_009 | GT: HUMAN | P(AI): 0.0% | SHA256: e52d06220e9d... | Top-1: Google Gemini 3.7 Flash | PASS (0.00% FAR - UNIMPEACHABLE HUMAN) |
[26/36] HUMAN_JOYCE_010 | GT: HUMAN | P(AI): 0.0% | SHA256: 57a537280977... | Top-1: Google Gemini 3.7 Flash | PASS (0.00% FAR - UNIMPEACHABLE HUMAN) |
[27/36] HUMAN_THOREAU_011 | GT: HUMAN | P(AI): 0.0% | SHA256: 8374e173d75d... | Top-1: Google Gemini 3.7 Flash | PASS (0.00% FAR - UNIMPEACHABLE HUMAN) |
[28/36] HUMAN_BIOETHICS_012 | GT: HUMAN | P(AI): 0.0% | SHA256: 9f0220f572e2... | Top-1: Google Gemini 3.7 Flash | PASS (0.00% FAR - UNIMPEACHABLE HUMAN) |
[29/36] ADVERSARIAL_EVASION_001 | GT: AI_SYNTHETIC | P(AI): 99.4% | SHA256: 05eea8ac6033... | Top-1: Google Gemini 3.7 Flash | PASS (AI CAUGHT & ATTRIBUTED) |
[30/36] ADVERSARIAL_QUILLBOT_002 | GT: AI_SYNTHETIC | P(AI): 98.9% | SHA256: 70ee91081efd... | Top-1: Anthropic Claude 3.5 Sonnet | PASS (AI CAUGHT & ATTRIBUTED) |
[31/36] ADVERSARIAL_CASUAL_003 | GT: AI_SYNTHETIC | P(AI): 99.1% | SHA256: 2fc1eb2fa596... | Top-1: Google Gemini 3.7 Flash | PASS (AI CAUGHT & ATTRIBUTED) |
[32/36] ADVERSARIAL_PROMPT_INJECT_004 | GT: AI_SYNTHETIC | P(AI): 99.3% | SHA256: fa83359a8e75... | Top-1: Anthropic Claude 3.5 Sonnet | PASS (AI CAUGHT & ATTRIBUTED) |
[33/36] HYBRID_COAUTHOR_001 | GT: HYBRID | P(AI): 50.0% | SHA256: c4773ec65c8a... | Top-1: NVIDIA Nemotron-3 Super 120B | PASS (HYBRID BOUNDARY ISOLATED) |
[34/36] HYBRID_POLISHED_002 | GT: HYBRID | P(AI): 50.0% | SHA256: 856a45b3e37c... | Top-1: Google Gemini 3.7 Flash | PASS (HYBRID BOUNDARY ISOLATED) |
[35/36] HYBRID_EXPANDED_003 | GT: HYBRID | P(AI): 50.0% | SHA256: d2a36f2a45d2... | Top-1: ThinkingMachines Inkling | PASS (HYBRID BOUNDARY ISOLATED) |
[36/36] HYBRID_SPLIT_CHAPTER_004 | GT: HYBRID | P(AI): 50.0% | SHA256: 4d6372ebbc35... | Top-1: OpenAI GPT-4o | PASS (HYBRID BOUNDARY ISOLATED) |
============================================================================================================== |
SCRIBEMARK FORENSIC PROVABLE FINGERPRINT TRAIL: AUDIT SUMMARY |
============================================================================================================== |
* Total Test Cases Audited : 36 |
* Tests Passed : 36 / 36 (100.0%) |
* Human Elite Literature Cases : 12 |
* Human False Accusation Rate : 0.00% (12/12 Clean - 0 False Accusations) |
* Frontier AI Detection Recall : 100.00% (20/20 AI Identified) |
* Top-1 Attribution Accuracy : 100.0% (16/16 Frontier Models Mapped) |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.