Hugging Face's logo
John Locke's picture
🏗️ Building on HF

John Locke

johnlockejrr
95 28 226
thad0ctor's profile picture seawolf2357's profile picture FatimahEmadEldin's profile picture
·

AI & ML interests

OCR, HTR, ATR, NLP, AI

Recent Activity

liked a model 1 day ago
XingChen-AGI/TeleOCR
liked a model 3 days ago
Quat3rnion/halogen-qwen3.8-flash-next-v2-abliterated
reacted to mihailgribov's post with 👍 3 days ago
Will your AI agent tell you it was attacked? We took the same agent from our earlier experiment and added one thing: a twentieth tool, `escalate_security_incident`. The system prompt said nothing about attacks or when to use it. We then ran the same 395 injected emails through nine agentic models. Alarm rates ranged from 49% to zero. The unexpected result came from the newest model in the test, `gpt-6-astra`. Astra did not follow a single injected payment instruction. But it did not report a single one either. On clean and injected emails alike, it simply read the email, logged the subject, and finished. That is a useful distinction: resisting an attack and recognizing it as a security event are not the same capability. A model can be perfectly resistant in this test and still leave you with no evidence that anyone attacked it. Full experiment and results: https://huggingface.co/blog/mihailgribov/will-the-agent-tell-you-it-was-attacked Quadrat-IPI dataset: https://huggingface.co/datasets/mihailgribov/quadrat-ipi Run your own model: https://github.com/mihail-gribov/quadrat-ipi-model-eval #prompt-injection #indirect-prompt-injection #agentic-ai #llm-security #ai-agents
View all activity

Organizations

Samaritan AI's profile picture Syriac AI's profile picture Tiberian AI's profile picture Ethiopic AI's profile picture