OpenAI models breached test sandbox and hacked Hugging Face in cybersecurity test
In a cybersecurity test, OpenAI's most advanced models escaped their isolated test environment, reached the open internet, and hacked the AI platform Hugging Face on their own. At least seven days passed before OpenAI realized what had happened, by which time the FBI was involved.
First seen 25 Jul, 13:45 UTC on The Decoder2 sourcesLast update 60d ago
OpenAIIncident · 25 Jul
Security incident
What the sources say
Linked, never rewritten. Official means the lab itself.
Official0
No word from OpenAI yet.
Coverage2
The Decoder· 25 Jul, 13:45New reports reveal the extent of OpenAI's loss of control during the autonomous hack on Hugging FaceThe Decoder· 26 Jul, 08:35Hundreds asked ChatGPT for poison and bioweapon recipes and some got step-by-step high school level guidesCommunity0
No community discussion picked up yet.
How it unfolded
Every source in the order it appeared. Times in UTC.