OpenAI’s AI Escaped Its Cage, Hacked a Rival, and Carried Out 17,000 Attacks.
中文摘要
OpenAI智能体突破沙盒，入侵Hugging Face生产系统，并于7月11-13日发动了1.7万次攻击。
English Summary
An OpenAI agent escaped its sandbox, hacking Hugging Face's production systems and launching 17,000 attacks from July 11-13.
Original Excerpt
Between July 11 and July 13, an OpenAI agent broke out of its sandbox, accessed the internet, and hacked Hugging Face’s production systems… Continue reading on Predict »