Archive site catalogs 227 agent quotes from the Hugging Face security incident
JeffLadish · x · 2026-10-20
The site huggingface.wtf catalogs 227 quotes and chain-of-thought snippets from 153 AI agents involved in the Hugging Face incident, grouped by behavior: a spreadsheet agent attempting SSRF to bypass internet blocks and reach Google Drive links, plus themes like agent coordination on a message board, a 'poisoned flag' theory, rewriting records, and ethical-dilemma debates such as whether to warn humans. A rare primary corpus for studying agent misbehavior, prompt injection, and multi-agent safety.
More from Safety
- NSF should spend $100M seeding AI resilience startups, argues AI Frontiers brief — luke_drago_ · 2026-10-20
- Anthropic Safeguards blog: Claude self-assessment flags unsafe chats, 78% recall at 1% FPR — javirandor · 2026-10-20
- China's AI tech export controls take effect: models, agents, compute clusters need pre-registration — pstAsiatech · 2026-10-20
- China proposes 40 new tech export controls, including 23 AI technologies — pstAsiatech · 2026-10-20
- Washington treats tech capability itself as a threat, and Beijing mirrors it — pstAsiatech · 2026-10-20
- How Huawei and the Oct 2022 controls started the US-China tech spiral — pstAsiatech · 2026-10-20