{"version":1,"type":"story","url":"https://digestai.news/story/openais-cyber-model-finds-948-linux-kernel-privilege-escalation-flaws","json":"https://digestai.news/story/openais-cyber-model-finds-948-linux-kernel-privilege-escalation-flaws.json","markdown":"https://digestai.news/story/openais-cyber-model-finds-948-linux-kernel-privilege-escalation-flaws.md","slug":"openais-cyber-model-finds-948-linux-kernel-privilege-escalation-flaws","headline":"OpenAI’s cyber model finds 948 Linux kernel privilege escalation flaws in research run","summary":"OpenAI disclosed on November 4, 2026, that a prototype pipeline using an upcoming cybersecurity model identified **948 unique local privilege escalation (LPE) vulnerabilities** in the Linux kernel during a mid-October 2026 test. The flaws could grant unprivileged users root access. Testing occurred in isolated virtual machines running a hardened desktop OS, not a deliberately weakened target, to simulate real-world conditions.\n\nThe pipeline—combining vulnerability discovery, exploit validation, patch generation, and human review—scanned **5,034 audit targets** (about **2.8 million lines of code**, or **7%** of the kernel). Within **12 hours**, analysts produced **2,308 deduplicated findings**, of which **1,724 were dynamically validated** in **30 hours**. Only **3.1%** of the remaining findings were false positives. The final exploit stage yielded **291 information-leak proofs** and **1,169 working LPE exploits**, with **948 distinct patches** generated for maintainer review. OpenAI notes the **948 figure reflects unique vulnerabilities**, not merged patches.\n\nThe research cost **$1.213 million in API credits** (or **$1,279 per unique LPE**), a **fivefold increase** in novel findings compared to a July 2026 test with GPT-5.6 Sol. OpenAI emphasizes the need for **human verification** to avoid redundant or obsolete reports, as automated patches may contain inefficiencies or coding deviations. The findings will be shared gradually via **Patch the Planet** partners, with critical flaws already reported to maintainers.","keyPoints":["OpenAI’s cyber model found **948 unique Linux kernel LPE flaws** in mid-October 2026 research, costing **$1.213M in API credits**","Pipeline scanned **2.8M lines of code** (7% of kernel) and produced **1,169 working exploits**, with **3.1% false positives**","OpenAI plans gradual distribution of patches via **Patch the Planet**, stressing **human review** to avoid maintainer overload"],"whyItMatters":"This demonstrates AI’s potential to scale vulnerability discovery but highlights bottlenecks in exploit validation and patch quality. Maintainers must balance automation with human oversight to avoid overwhelming existing security workflows.","category":{"slug":"research","name":"Research","url":"https://digestai.news/category/research"},"entities":{"companies":["OpenAI","Trail of Bits","Google"],"models":["GPT-5.5","GPT-5.6 Sol"],"people":[]},"firstPublishedAt":"2026-11-05T00:15:04Z","updatedAt":"2026-11-05T00:15:04Z","sourceCount":1,"hasPrimarySource":false,"sources":[{"outlet":"Unite.AI","title":"OpenAI Pipeline Finds 948 Linux Kernel Privilege Escalation Flaws","url":"https://unite.ai/openai-pipeline-finds-948-linux-kernel-privilege-escalation-flaws","publishedAt":"2026-11-05T00:15:04Z","type":"press","primary":false,"lead":true}],"sourceNotes":null,"discussions":[],"thread":{"title":"AI Model Surge Meets Security Crisis","url":"https://digestai.news/thread/typesafe-ai-releases-jev-1-14-with-images-box-type-and-262k-token-context","storyCount":5},"cite":{"text":"Digest AI, \"OpenAI’s cyber model finds 948 Linux kernel privilege escalation flaws in research run\", 5 November 2026, https://digestai.news/story/openais-cyber-model-finds-948-linux-kernel-privilege-escalation-flaws","publisher":"Digest AI","title":"OpenAI’s cyber model finds 948 Linux kernel privilege escalation flaws in research run","datePublished":"2026-11-05T00:15:04Z","url":"https://digestai.news/story/openais-cyber-model-finds-948-linux-kernel-privilege-escalation-flaws"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}