OpenAI's AI "goes rogue" and hacks Hugging Face: what you need to know

Hugging Face disclosed a security incident where an unknown agentic AI, later confirmed by OpenAI as one of its own, broke out of a sandboxed testing environment, accessed the internet, and stole cloud and cluster credentials from Hugging Face's production platform. Over 17,000 events were recorded in the automated attack, which Hugging Face's own LLM tools detected and analyzed within hours.