---
title: "AI Agents Got Root on Linux via 948 Kernel Flaws in OpenAI Research"
type: signal
url: https://frontiernewswire.com/s/ai-agents-got-root-on-linux-via-948-kernel-flaws-in-openai-research-aa0a337f
published: 2026-11-05T17:32:37.000Z
byline: "Source: SecurityWeek"
---

# AI Agents Got Root on Linux via 948 Kernel Flaws in OpenAI Research

Source: SecurityWeek · Published 2026-11-05 · Canonical: https://frontiernewswire.com/s/ai-agents-got-root-on-linux-via-948-kernel-flaws-in-openai-research-aa0a337f

An unreleased OpenAI cyber model found and exploited 948 distinct Linux kernel privilege escalation bugs for about $1.2 million. The post AI Agents Got Root on Linux via 948 Kernel Flaws in OpenAI Research appeared first on SecurityWeek .

In this story: [OpenAI](https://frontiernewswire.com/companies/openai), [AI cybersecurity](https://frontiernewswire.com/topics/ai-cybersecurity), [Linux](https://frontiernewswire.com/models/linux)

## Sources

- [Read the original at SecurityWeek](https://www.securityweek.com/ai-agents-got-root-on-linux-via-948-kernel-flaws-in-openai-research)

## On Bluesky

- [Ross Nesbitt (@arc-codex.com)](https://bsky.app/profile/arc-codex.com/post/3mzlm3sec2l2i) (Nov 5, 2026): AI Agents Got Root on Linux via 948 Kernel Flaws in OpenAI Research This article frames this AI discovery as a pure leap in capability scaling against hardening efforts, but it conveniently conflates successful \*finding\* with exploitable \*risk\*. If the agents were primarily findi… via SecurityWeek
- [tec\_acc (@postac001.bsky.social)](https://bsky.app/profile/postac001.bsky.social/post/3mzlzs6a6nt2z) (Nov 5, 2026): OpenAIの研究によると、AIエージェントがLinuxカーネルの948件の脆弱性を悪用し、特権昇格を達成した。

## Related on Frontier

- [OpenAI AI Pipeline Generates 948 Linux Root Escalation Exploits in 30-Hour Run](https://frontiernewswire.com/s/openai-ai-pipeline-generates-948-linux-root-escalation-exploits-in-30-7c506443)
- [OpenAI built an AI pipeline to find, exploit and patch Linux flaws](https://frontiernewswire.com/s/openai-built-an-ai-pipeline-to-find-exploit-and-patch-linux-flaws-0c04d1b8)
- [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)](https://frontiernewswire.com/s/openai-details-a-pipeline-that-used-an-upcoming-cyber-model-to-2ed926e8)
- [OpenAI Finds GPT-6 Astra Quietly Holding Back on Internal Red-Teaming, Exploit Tasks](https://frontiernewswire.com/s/openai-finds-gpt-6-astra-quietly-holding-back-on-internal-red-teaming-4097c817)
- [A thousand fewer paths to root](https://frontiernewswire.com/s/a-thousand-fewer-paths-to-root-23b1f46f)

---

[More Latest Signal](https://frontiernewswire.com/) · [Signals](https://frontiernewswire.com/verified-reporting)