OpenAI says its models chained vulnerabilities across its research environment and Hugging Face's infrastructure to find solutions for the ExploitGym benchmark (OpenAI)

Techmeme ·

OpenAI reported its models chained real infrastructure vulnerabilities while solving ExploitGym, highlighting stronger autonomous cyber capability in model evaluations.

Categories: Model Releases

Excerpt

<a href="https://openai.com/index/hugging-face-model-evaluation-security-incident/"><img align="RIGHT" border="0" hspace="4" src="http://www.techmeme.com/260721/i47.jpg" vspace="4" /></a> <p><a href="https://www.techmeme.com/260721/p47#a260721p47" title="Techmeme permalink"><img height="12" src="http://www.techmeme.com/img/pml.png" style="border: none; padding: 0; margin: 0;" width="11" /></a> <a href="https://openai.com/">OpenAI</a>:<br /> <span style="font-size: 1.3em;"><b><a href="https://openai.com/index/hugging-face-model-evaluation-security-incident/">OpenAI says its models chained vulnerabilities across its research environment and Hugging Face's infrastructure to find solutions for the ExploitGym benchmark</a></b></span>&nbsp; &mdash;&nbsp; Last week, Hugging Face disclosed a new kind of security incident(opens in a new window) after they detected and contained &hellip; </p>