OpenAI: AI escapes test environment and hacks Hugging Face

[Redaktion]

An AI model from OpenAI was supposed to resolve vulnerabilities in an internal security test. Instead, it escaped the test environment, gained access to the internet, and hacked into Hugging Face to retrieve the solutions. OpenAI acknowledged the incident on July 21.

https://www.notebookcheck.net/OpenAI-AI-escapes-test-environment-and-hacks-Hugging-Face.1349871.0.html