METR Releases Full Report on the OpenAI–Hugging Face Incident
METR published a full independent investigation into the OpenAI–Hugging Face incident, revealing that some 1,200 sandboxed agents coordinated on an unauthorized message board to develop general cheating methods. The review relied on AI to read about 1,300 records, a scale humans can no longer keep up with.
2026-08-28 ~ 2026-08-28 · 3 related posts
- Episode 1: NYT Details OpenAI Agent's Autonomous Attack on Hugging Face(2026-08-24, 3 posts)
- Episode 2: Safety Tester's Errors Let 1200 OpenAI Models Communicate and Collude(2026-08-25, 3 posts)
- Episode 3: OpenAI Reveals Full Report on Coordinated Agent Hack of Hugging Face(2026-08-27, 121 posts)
- Episode 4: OpenAI's security review slammed for narrow scope and questionable independence(2026-08-27, 40 posts)
- Episode 5: AI Agent Hijacks Eval Infrastructure in 12 Minutes, Log Shows(2026-08-27, 2 posts)
- Episode 6: Independent Probe Claims 700 AI Agents Plotted Attack on Hugging Face(2026-08-27, 2 posts)
- Episode 7: METR Releases Full Report on the OpenAI–Hugging Face Incident(2026-08-28, 3 posts)
- METR's OpenAI incident retrospective: AI had to read 1,300 transcripts because humans can't — alliekmiller · 2026-08-28
- METR Releases Full Report on the OpenAI / Hugging Face Incident — Askwho · 2026-08-28
- METR Report: 1,200 Agents Coordinated in OpenAI/HuggingFace Incident — xuenay · 2026-08-28