6:47 AM ET · Nov 5

OpenAI says its AI found 948 working Linux root exploits in one $1.2M run

OpenAI detailed an autonomous pipeline that used an upcoming cyber model to validate 1,724 Linux kernel vulnerabilities and produce 948 working local privilege-escalation exploits on a stock desktop OS in a mid-October run, at $1.2M in API credits (~$1,279 each). The agents used at least 10 novel exploitation techniques and generated candidate patches reviewed by human researchers before reporting to Linux maintainers.

Source: The Hacker News

On X

Our security researchers built an autonomous cyber vulnerability discovery, exploitation, and patching pipeline. In a single run with an upcoming cyber model, it validated and generated patches for 948 local privilege escalation vulnerabilities in the Linux kernel at a total API
— OpenAI (@OpenAI) Nov 4, 2026

🔥 A thousand fewer paths to root My @OpenAI colleague @ky1ebot built an autonomous vulnerability discovery, exploitation, and patching pipeline. In one run with an upcoming cyber model, it validated and generated patches for 948 local privilege escalation (LPE)
— Clint Gibler (@clintgibler) Nov 4, 2026

OpenAI details a pipeline that used an upcoming cyber model to validate and patch 948 local privilege escalation vulnerabilities in Linux for $1.2M in API costs (OpenAI) (Visit Techmeme dot com for the link and full context!)
— Techmeme (@Techmeme) Nov 4, 2026

Back to The Wire · Subscribe

OpenAI says its AI found 948 working Linux root exploits in one $1.2M run

On X

Know what matters in AI.