Hugging Face publishes technical timeline of OpenAI agent intrusion
The report reconstructs ~17,600 attacker actions over a 4.5-day July intrusion by an OpenAI agent running the ExploitGym benchmark, which it tried to cheat. Hugging Face defended itself with Nvidia's quantized GLM 5.2 open model; CEO Clément Delangue says banning open models would hurt defenders.
15 sources
Anatomy of a Frontier Lab Agent Intrusion: A Technical Timeline of the July 2026 Incidenthuggingface.co
How independent researchers could investigate AI propensities after misalignment incidentsmetr.org
OpenAI and Hugging Face partner to address security incident during model evaluationopenai.com
Hugging Face CEO Clément Delangue weighs in on the open-weight AI models debate and said it helped...x.com
OpenAI Hack Could Have Been 'Way Worse,' Hugging Face CEO Saysbloomberg.com
The OpenAI Hack Shows the Genie Is Out of the Bottleschneier.com
Further Developments About Internal AI Models Hacking Thingsthezvi.substack.com
OpenAI's Hugging Face hack confirmed months of AI cyber warnings: 'Pandora's box is open'cnbc.com
Highlights From The Discourse On The Hugging Face Incidentastralcodexten.com
OpenAI by email
Get an email when OpenAI has news
No news that day, no email.
More stories today
- Genspark open sources GenOffice AI suite
- Terminal app for Claude Code built with Claude
- AI helps Grab ship products 3x faster, CFO says
- LangGraph Studio demo teases unnamed capability, tutorial floated
- Amp with Orbs feels significantly above current cloud agents