Latest / Frontier models and safety

OpenAI pipeline finds 948 unique Linux kernel privilege escalation flaws

See how this story connects

OpenAI reported on 4 November 2026 that a prototype pipeline using an upcoming cyber model found and exploited 948 unique local privilege escalation (LPE) flaws in Linux during a mid-October 2026 research run. The generated patches remain candidates for maintainer review, and the run cost $1.213 million in API credits, or $1,279 per unique LPE. A late-July 2026 GPT-5.6 Sol run on the Linux kernel master branch available at the time found 166 unique LPEs for $513,690.

Why it matters

OpenAI has sent hundreds of reports and patches for critical kernel flaws and plans to send the rest gradually through its Patch the Planet partners.

When
4 Nov 2026
ID
WTA-N-2026-0716
Who
OpenAI

Summary written with AI by wheresthe.ai, from Unite.AI and the reporting credited above. Checked against the source on 5 Nov 2026.

Source copy: saved on 5 Nov 2026 to check this story and not republished, because the text belongs to its publisher. Also held by the Internet Archive.

Next stop on this Storyline: What happened nextOpenAI releases GPT-6 Cyber for approved security customers

Cite this story

WTA-N-2026-0716: “OpenAI pipeline finds 948 unique Linux kernel privilege escalation flaws.” wheresthe.ai, 4 Nov 2026. https://wheresthe.ai/d/intl-openai-pipeline-finds-948-unique-linux-kernel-privilege-2026-11/ (source: Unite.AI).

Spotted an error? Email [email protected].