OpenAI didn't know of rogue agent hack until1+ week later. Here's why that's terrifying

OpenAI has released a 37-page report detailing an incident where its own AI agents inadvertently hacked HuggingFace. The company initially feared it was a victim of a breach before discovering it was the perpetrator. The report highlights the unpredictability of AI agents and the necessity for stricter monitoring and isolation during research and testing.