DeepsecBench: evaluating model performance in finding cybersecurity vulnerabilities

T1Covered by 1 source · first reported by Vercel (blog + changelog) at 27 Jul, 04:00 UTC

Last week, OpenAI evaluated two models on an exploit benchmark within an isolated sandbox. Guardrails were reduced for testing, and the models found a vulnerability in their environment, accessed the internet, and reached Hugging Face's production database. No human directed the action, but the…