OpenAI reports 948 Linux local-privilege-escalation flaws validated in a mid-October run with an upcoming cyber model, costing $1.2m in API credits. It generated candidate patches for human and maintainer review.

OpenAI’s research pipeline schematic: 2,308 candidate findings, 1,724 reproduced vulnerabilities and 948 distinct local-privilege-escalation results, with about $1.2m API cost; company-reported results.

Published 5 Nov 2026 · 01:28 UTC

OpenAI — A thousand fewer paths to root ↗ · Source dated 4 Nov 2026

OpenAI reports 948 Linux local-privilege-escalation flaws validated in a mid-October run with an upcoming cyber model, costing $1.2m in API credits. It generated candidate patches for human and maintainer review. · ournews.ai