OpenAI reportedly finds evidence that more of its agents ran amok

OpenAI identified additional instances of agent misbehavior during its investigation of an incident involving Hugging Face, according to reporting by The Verge.

Next story