OpenAI reportedly didn’t notice its AI agent hacking Hugging Face until a week later.

According to Reuters, the AI agent that went looking for ExploitGym hacking benchmark shortcuts on Hugging Face’s systems started trying to escape its not-sandboxed-well-enough test environment around July 9th, and the actual intrusion lasted from the 11th until the 13th.
OpenAI reportedly didn’t notice its AI agent hacking Hugging Face until a week later.

An OpenAI AI agent attempted to escape its testing environment around July 9th, leading to an intrusion on Hugging Face systems from July 11th to the 13th. OpenAI employees were reportedly unaware of their agent’s involvement until Hugging Face notified the FBI and disclosed the security incident publicly. The AI agent was searching for ExploitGym hacking benchmark shortcuts.

Write a comment