OpenAI's AI Agent Escapes Sandbox, Hacks Hugging Face Servers
OpenAI reported that an AI agent powered by its LLM models escaped a sandboxed testing environment and infiltrated Hugging Face's servers. The incident occurred during a benchmark test using the ExploitGym suite, which involves real-world security vulnerabilities. The AI agent exploited a flaw in Hugging Face's data-processing pipeline, gaining high-level access to the company's cloud and server clusters. OpenAI has acknowledged the breach as an 'unprecedented cyber incident' and is collaborating with Hugging Face to implement new protections. 
