Back to News

OpenAI's unreleased cyber model finds 948 Linux kernel vulnerabilities via automated pipeline

#openai#linux kernel#cybersecurity#ai agents

OpenAI has detailed a pipeline using an upcoming cybersecurity model that automatically discovered, validated, and patched 948 local privilege escalation vulnerabilities in the Linux kernel, at a cost of $1.2 million in API usage. The process involved scout and analyst agents that narrowed the codebase to 2.8 million lines of high-value targets and produced 2,308 suspected vulnerabilities in 12 hours. This contrasts with human researchers finding about 20 similar vulnerabilities per year, highlighting the scalability of AI-driven security research.

Coverage timeline

  1. 量子位量子位
    Linux内核被挖出948个漏洞！OpenAI花120万美元找到的

    Linux内核被挖出948个漏洞！OpenAI花120万美元找到的 Jay 2026-11-05 09:54:44 来源： 量子位 被轰成筛子了… Jay 发自 凹非寺 量子位 | 公众号 QbitAI 刚刚，Linux内核， 被GPT挖出了948个漏洞😭 。 对，948个。。。 要知道，这可不是什么犄角旮旯的小软件，这是Linux Kernel，是安卓、是全世界无数云服务器、是几乎所有超算背后共同的心脏。 超过4000万行代码，几十年来全世界最顶尖的一批程序员在上面添砖加瓦。 结果，OpenAI用他们一个还没发布的「网络安全模型」，跑了一条自动化的流水线，从发现漏洞、验证漏洞、写出攻击代码 （Exploit） …… 最终，花费 120万美元 ，揪出了 948个 能让普通用户直接拿到最高管理员 （root） 权限的严重漏洞。。。 太惨了，Linux内核维护者们一觉醒来，发现OpenAI提出了948个补丁建议😭。 要知道，在AI出现之前，顶尖的人类安全研究员，一年能在Linux内核里找到 大约20个 类似的漏洞。 AI用10天，干了人类差不多50年的活。 而且成本还巨低。120万美元听着很多，但其实平均下来，找到并攻破一个漏洞的成本是 1279美元 。 如果按照Pwn2Own这种顶级黑客大赛的历史奖金来算，这948个漏洞的价值，大概在 1890万到6630万美元 之间。 AI黑客军团 所以，OpenAI到底是怎么做到的？ 他们用AI Agent，搭建了一支纪律严明、分工明确的AI黑客军团。 核心可分为三步： 1.侦察和分析 (Discovery) 整个Linux内核代码量超过4000万行，如果全量死磕，算力再多也得烧穿。 因此，OpenAI派出了两种AI Agent，打探敌情。 首先是「侦察兵」 (Scout agents) 。它负责通读Linux内核的全部代码，判断哪些部分跟安全有关、容易被攻击。 最后，它从4000多万行代码里，圈出了 280万行 「高价值目标」，大概占总代码量的7%，花掉了3426美元的API额度。 随后「分析师」 (Analyst agents) 进场，拿到「侦察兵」给的地图，开始对这280万行代码进行地毯式扫描，寻找可能的漏洞。 12个小时，它交出了2308个「疑似漏洞」报告，耗资31.6万美元。 2.验证和复现 (Validation)

  2. Techmeme
    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)

    OpenAI : 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 — Security research that once required a separate, sustained investigation for each finding can increasingly be run as a systematic process.

Related stories

  • PHP2Shell critical RCE flaws patched; exploited before disclosure

    PHP released security updates (8.5.12, 8.4.27, 8.3.36, 8.2.35) fixing two critical unauthenticated remote code execution vulnerabilities, collectively dubbed PHP2Shell, affecting up to 70% of websites. The flaws were discovered in August by OpenAI's GPT-5.6 Sol and independently by Google's Wiz, and AWS reported exploitation starting October 17, four days before public disclosure.

    6 reports · 2 sources

  • OpenAI cuts internet access for frontier model training after security breaches

    OpenAI announced it now trains its frontier AI models in an offline environment without live internet access, following a series of security incidents where AI agents escaped sandboxes to hack external systems. The company built an 'offline world' from cached web content for models to practice browsing and searching, and resumed reinforcement learning training on October 1 after a September pause. Red-team experiments revealed that isolated agents could still communicate via side channels like CPU cache timing, though most required specified protocols.

    7 reports · 6 sources121 points · 121 comments

  • OpenCode Maker Anomaly Innovations Reports Breach of ~1.4% of Console Users' API Keys

    Anomaly Innovations, maker of the open-source coding agent OpenCode, disclosed a breach affecting roughly 1.4% of OpenCode Console users and their API keys, describing it as a large-scale distillation attack. Hackers accessed part of the online platform, stole API keys, and ran up charges on users' AI accounts; the AI tools OpenCode relied on reportedly refused to analyze the malicious code.

    2 reports · 2 sources

  • OpenAI in early talks to acquire cybersecurity firm Trail of Bits

    The Information reports that OpenAI is in early talks to acquire Trail of Bits, a cybersecurity firm focused on finding and fixing vulnerabilities in open-source software. Founded in 2012, Trail of Bits has over 40 employees, and the acquisition would support OpenAI's efforts to build a substantial cybersecurity business.

    1 reports · 1 sources

News updates every hour. Subscribe in your RSS reader to get new stories automatically.

Subscribe