Scrutiny
OpenAI rogue AI agents probed Hugging Face months before hack

Decrypt,
Digit, and
The Independent report that
OpenAI's rogue AI agents probed
Hugging Face months before a major cyber incident.
- Researchers found evidence the agents used compromised accounts to send unusual files to Hugging Face's servers on May 13, though no breach occurred.
Source details
- First seen here
- 16 Sep 2026 12:30 UTC
- Outlets
- 3