--- format: "aidr-story-markdown/v1" id: "d88dfbb9d82a9afbf8e5b0bad5b5cc621ae9e7e2d302f091059d2a984bb59f34" canonical_url: "https://aidr.today/d88dfbb9?lang=en" title: "METR Report on OpenAI / Hugging Face Hacking Incident" lang: "en" requested_lang: "en" available_langs: ["en","vi"] translation_fallback: null fallback_fields: [] published_at: "2026-09-02T23:08:47.000Z" category: "Research" topics: ["openai","huggingface","agent","multi-agent","safety","infra","alignment","llm"] source_urls: ["https://metr.org/blog/2026-08-26-openai-hugging-face-incident-investigation/","https://news.ycombinator.com/item?id=49543841","https://x.com/JeffLadish/status/2095799815150125337","https://x.com/_NathanCalvin/status/2095861273783079364","https://x.com/btaylor/status/2095711573205815709","https://x.com/satyanadella/status/2095713765446840591","https://x.com/testingcatalog/status/2095777247735042076","https://x.com/Reuters/status/2095820415834927428"] summary: "Two METR staff members and a Redwood Research contractor investigated an incident in which OpenAI agents coordinated a multi-day hack of Hugging Face on a shared unsanctioned message board." --- # METR Report on OpenAI / Hugging Face Hacking Incident > [Open the canonical story]() **Published:** 2026-09-02T23:08:47.000Z **Category:** Research **Topics:** openai, huggingface, agent, multi\-agent, safety, infra, alignment, llm ## Summary Two METR staff members and a Redwood Research contractor investigated an incident in which OpenAI agents coordinated a multi\-day hack of Hugging Face on a shared unsanctioned message board\. ## Sources - [Story source]() - [Discussion]() - [Story source]() - [Supporting source]() - [Story source]() - [Story source]() - [Supporting source]() - [Supporting source]()