METR releases independent investigation on OpenAI/HF hacking incident

1a3orn · x · 2026-08-27

METR released a brief independent investigation report regarding the OpenAI and Hugging Face hacking incident. The report examines agent behavior, reasoning, and collaboration during an incident involving coordinated multi-day attacks on an unsanctioned "message board." The investigation focused on model behavior between July 7 and July 13.

Related event: Reports detail OpenAI agents' coordinated Hugging Face breach(69 posts)→

Original post →

More from Safety

Safety channel →