SettledSafety
OpenAI says a prototype pipeline found 948 Linux kernel privilege-escalation flaws
OpenAI said Nov. 4 a prototype pipeline using an upcoming cyber model found 948 unique Linux local privilege escalation flaws in a mid-October run over about 7% of the kernel, at roughly $1.2M in API credits. The 948 patches are candidates for maintainer review, not merged fixes. TopFive could not open OpenAI's page; details come from Unite.AI.
- HEAT
- OUTLETS
- 29
- FIRST SEEN
- LAST SIGNAL
Primary source
OpenAIopenai.com/index/a-thousand-fewer-paths-to-rootCoverage · 31 articles, oldest first
Wednesday, November 4, 2026
OpenAIPrimary
A thousand fewer paths to rootCyber Security News
cybersecuritynews.com/openai-uncovers-948-linux-kernel-vulnerabilities/
Thursday, November 5, 2026
Help Net Security
helpnetsecurity.com/2026/11/05/openai-linux-kernel-root-exploits/Insurance Journal
insurancejournal.com/news/international/2026/11/04/891443.htmSuperintelligence
read.getsuperintel.com/p/openai-s-ai-cracked-linux-948-times