--- title: OpenAI says its cyber model patched 948 Linux kernel... description: OpenAI says an autonomous pipeline built with an upcoming cyber model found, exploited, and patched 948 local privilege escalation bugs in the Linux kernel in... canonical: https://daily.dev/posts/openai-says-its-cyber-model-patched-948-linux-kernel-privilege-escalation-bugs-in-one-run-joeov8zrh twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: OpenAI says its cyber model patched 948 Linux kernel privilege escalation bugs in one run | daily.dev og:description: OpenAI says an autonomous pipeline built with an upcoming cyber model found, exploited, and patched 948 local privilege escalation bugs in the Linux kernel in... og:url: https://daily.dev/posts/openai-says-its-cyber-model-patched-948-linux-kernel-privilege-escalation-bugs-in-one-run-joeov8zrh og:image: https://api.daily.dev/og/posts/jOeOv8Zrh.png og:image:alt: OpenAI says its cyber model patched 948 Linux kernel privilege escalation bugs in one run og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # OpenAI says its cyber model patched 948 Linux kernel privilege escalation bugs in one run **[Collections](https://daily.dev/sources/collections)** · 3 min read · 0 upvotes · 0 comments ## Summary OpenAI says an autonomous pipeline built with an upcoming cyber model found, exploited, and patched 948 local privilege escalation bugs in the Linux kernel in a single run. The company put the total API cost at $1.2 million, about $1,270 per vulnerability. OpenAI hasn't named the model or said when it will ship. It also hasn't explained how the 948 results were validated or whether kernel maintainers have accepted any of the patches. ## Content OpenAI says its security researchers built an autonomous pipeline that finds, exploits and patches software vulnerabilities. In a single run with an upcoming cyber model, it validated and generated patches for 948 local privilege escalation vulnerabilities in the Linux kernel. The total API cost was $1.2 million. ## How the run worked The pipeline splits the job among several kinds of agents. Scout agents pick targets, analyst agents hunt for bugs, triage agents reproduce them, exploit agents prove them, and then the system writes a patch. The scouts selected 5,034 security-relevant audit targets. That covers about 2.8 million lines of code, roughly 7% of a kernel that runs past 40 million lines. The numbers narrowed quickly from there: - 2,308 candidate bugs flagged - 1,724 reproduced on test machines - 948 turned into working local privilege escalation exploits, meaning an unprivileged user gets root The exploits ran on a stock desktop distro. OpenAI says the agents used at least 10 exploitation techniques that outside kernel experts hadn't seen before. Humans review everything before it goes to Linux maintainers, with patches attached. ## The cost The API bill was $1,213,430, with $775,929 of it spent building exploits. That works out to about $1,279 per exploit. Since OpenAI ran its own model, that's probably not cash actually spent. @gdb put it at about $1,000 per vulnerability. The price is what stands out to me, more than the 948. A root exploit for the most widely deployed kernel in the world, for the price of a decent laptop. Defenders and attackers both get access to this kind of tooling, so slow patching is where the risk lands. ## Progress over time In July, GPT-5.6 Sol found 166 of these bugs for about $514k. The October run found 948 on the same kind of scan. OpenAI also reported a few lessons from building it: - Heavy scaffolding raised GPT-5.5 Cyber's success rate from 76.2% to 88.9%. GPT-5.6 Cyber did better without it, and adding it back gained just 1.5 points. - In an early run, 20 of 200 "successful" exploits got root through a different bug than the one assigned, so the method matters as much as the result. - The research also found cases where the AI spotted problems in patches written by human experts. ## Is it fixed? Only partly. When the 10-day scan ended, 61 issues had been patched upstream, 43 had patches in staging and 299 were being discussed publicly. OpenAI has sent hundreds of reports and patches for the critical ones and is sending the rest gradually. OpenAI's guidance is blunt: an unpatched system where an attacker has even a low-privileged account should now be treated as functionally equivalent to root. The practical advice is to keep systems updated. ## What it means for builders If AI can find bugs faster than maintainers can fix them, automated patch generation and human validation stop being optional. For anyone shipping software, the sensible habit is to rescan old code whenever a new model arrives, because code that passed last quarter's check can fail this one. Scanning your own code is cheaper than ever, so it belongs in the shipping process. ## Questions this post answers ### How many Linux kernel privilege escalation vulnerabilities did OpenAI's cyber model find in one run? OpenAI's autonomous pipeline validated and generated patches for 948 local privilege escalation vulnerabilities in the Linux kernel during a single run with an upcoming, unnamed cyber model. The total API cost came to $1.2 million, roughly $1,270 per vulnerability. OpenAI has not disclosed how the results were validated or whether the patches were accepted upstream by kernel maintainers. _Developers weighing AI-driven vulnerability tools can track verified outcomes like this on daily.dev._ ### Has OpenAI released the cyber model used to find the Linux kernel vulnerabilities? No, OpenAI has not named the model or given a release date for it. The company only describes it as 'an upcoming cyber model' used in an autonomous pipeline that finds, exploits, and patches vulnerabilities, without disclosing further details about its architecture or availability. _Anyone tracking new AI security models can follow release details as they surface on daily.dev._ ## Community take How the wider developer community reacted, aggregated from 2 discussions and 249 comments across hackernews, x (as of 2026-11-05). **TL;DR:** Reactions are largely impressed by the cost economics of finding and patching nearly 1,000 kernel bugs cheaply, but the dominant thread is skepticism about whether maintainers can actually review and merge that volume of patches without breaking things. A thousand fewer paths to root: The discussion largely accepts that privilege-escalation bugs have always riddled the Linux kernel, with debate centered on whether this reflects a fundamental flaw in the kernel's design/language choice versus simply being an inevitable byproduct of its size and complexity, while many are impressed (and a little unsettled) by how cheaply and quickly AI surfaced hundreds of them. **Sentiment:** 28% positive · 47% mixed · 25% skeptical **The case for** - Finding and validating exploits at roughly $1,000 each is seen as far cheaper than traditional human security audits or bug bounties. - AI-driven scanning finds vulnerabilities far faster and cheaper than traditional human-driven auditing could. - Several see this as a genuine step-change in defensive security, turning vulnerability discovery into a repeatable, scalable pipeline. - The pipeline authors showed up to directly answer technical questions, which was well received. - Some note the pipeline's validation step (falsifying its own findings, not just generating them) as the real technical achievement. **The pushback** - Many worry that patch review and merge is now the real bottleneck, not finding bugs, since maintainers can't realistically absorb hundreds of patches at once. - Skeptics argue the bugs were always findable and that investing in better tooling/safer languages earlier would have been cheaper than throwing millions at LLM scanning. - Several question whether a rushed patch could break production kernels, calling that worse than the original vulnerability. - Some doubt the sheer number (948) is meaningful without more detail on severity, overlap, and which subsystems are affected. - Some point out the same cheap pipeline could just as easily be used offensively by attackers at the same cost. **By community** - hackernews (mixed): Discussion ranges from technical debate over why the kernel has so many bugs (complexity, C's memory unsafety, business tradeoffs favoring speed) to genuine appreciation for the pipeline authors engaging directly, with a sizable skeptical/dismissive thread questioning the value and motives behind the research. - x (mixed): A flood of short replies mixes genuine excitement about the cost/scale breakthrough with recurring concern that maintainer review capacity, not bug-finding, is now the limiting factor. **Hottest debate:** Whether the bottleneck has shifted from finding vulnerabilities to maintainers' ability to review and safely merge hundreds of patches without breaking production systems. **Open questions** - How many of the 948 patches have actually been accepted or merged upstream so far? - Is there a way to estimate the total number of undiscovered vulnerabilities remaining, to gauge how much of the problem has actually been solved? - What fraction of the cost comes from model inference versus the validation harness versus human review of false positives? - Will the vulnerabilities be remediated structurally (fixing classes of bugs) or just patched individually? - How will patches reach users running vendor/backported kernels that don't track upstream directly? **Highlights** > @gdb The average hides the hard ones. In the report, failed exploit attempts cost about $3,922 each across 69 cases, against $360 for the 1,169 that worked. It puts that down to a few tough bugs like race conditions. > — [AIQuanting on x · 1 points](https://x.com/AIQuanting/status/2118134167720583241) > @gdb Validated against what, though? A patch that breaks a production kernel is worse than the vuln it fixes. > — [TheRobKennedy on x · 1 points](https://x.com/TheRobKennedy/status/2118141707414093864) > @gdb Per Eric Wallace's thread, researchers used to find ~20 Linux LPEs in a typical year. This found ~1,000 in a week, roughly 50 years' worth. Finding is no longer the bottleneck, review is. Are the patches going to kernel maintainers in batches, or through distros first? > — [shuishui1223 on x](https://x.com/shuishui1223/status/2118165902818742292) > @gdb The scale is the story: finding and validating patches for nearly a thousand privilege escalations turns vulnerability work from one-off fixes into a repeatable pipeline. The open question is review quality at that volume, and whether maintainers can trust each patch to merge it. > — [JulianLiu0414 on x](https://x.com/JulianLiu0414/status/2118138184655949840) > @gdb favorite detail in the writeup: in an early run, 20 of the 200 'successful' exploits got root through a different 0-day than the one they were assigned. the check only asked for a root shell > — [sachinmour on x](https://x.com/sachinmour/status/2118176018125205536) **Source threads** - [x](https://x.com/gdb/status/2118129547443056763) · 0 points · 127 comments - [hackernews](https://news.ycombinator.com/item?id=50372929) · 201 points · 120 comments - [x](https://x.com/0x0SojalSec/status/2118125101682139350) · 0 points · 2 comments - [x](https://x.com/h___t____n/status/2118117006797099238) · 0 points · 0 comments - [x](https://x.com/Magnosud/status/2118117532854886735) · 0 points · 0 comments ## Similar posts on daily.dev - [OpenAI was breached by a smarter model, not a new bug](https://daily.dev/posts/openai-was-breached-by-a-smarter-model-not-a-new-bug-bgdul0oun) · Kilo Blog · 0 upvotes · 0 comments - [An AI agent found 21 zero-days in FFmpeg for $1,000. Chrome just patched a record 429 bugs.](https://daily.dev/posts/an-ai-agent-found-21-zero-days-in-ffmpeg-for-1-000-chrome-just-patched-a-record-429-bugs--loichspdc) · The Next Web · 37 upvotes · 4 comments --- Tags: [#linux](https://daily.dev/tags/linux), [#ai-agents](https://daily.dev/tags/ai-agents), [#openai](https://daily.dev/tags/openai), [#vulnerability](https://daily.dev/tags/vulnerability) [View this post on daily.dev](https://daily.dev/posts/openai-says-its-cyber-model-patched-948-linux-kernel-privilege-escalation-bugs-in-one-run-joeov8zrh) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"OpenAI says its cyber model patched 948 Linux kernel privilege escalation bugs in one run","url":"https://daily.dev/posts/openai-says-its-cyber-model-patched-948-linux-kernel-privilege-escalation-bugs-in-one-run-joeov8zrh","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/openai-says-its-cyber-model-patched-948-linux-kernel-privilege-escalation-bugs-in-one-run-joeov8zrh"},"datePublished":"2026-11-04T23:14:00.829Z","dateModified":"2026-11-05T04:46:51.249Z","description":"OpenAI says an autonomous pipeline built with an upcoming cyber model found, exploited, and patched 948 local privilege escalation bugs in the Linux kernel in...","image":"https://pbs.twimg.com/media/HWUQl7QbwAAkznq.png","thumbnailUrl":"https://pbs.twimg.com/media/HWUQl7QbwAAkznq.png","isAccessibleForFree":true,"articleSection":"Collections","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Collections","logo":"https://media.daily.dev/image/upload/s--fk_6ycEi--/f_auto,q_auto/v1780996001/logos/collections?_a=BAMAMiWQ0","url":"https://daily.dev/sources/collections"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/openai-says-its-cyber-model-patched-948-linux-kernel-privilege-escalation-bugs-in-one-run-joeov8zrh","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"linux,ai-agents,openai,vulnerability","timeRequired":"PT3M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Collections","item":"https://daily.dev/sources/collections"},{"@type":"ListItem","position":3,"name":"OpenAI says its cyber model patched 948 Linux kernel privilege escalation bugs in one run"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/openai-says-its-cyber-model-patched-948-linux-kernel-privilege-escalation-bugs-in-one-run-joeov8zrh#faq","mainEntity":[{"@type":"Question","name":"How many Linux kernel privilege escalation vulnerabilities did OpenAI's cyber model find in one run?","acceptedAnswer":{"@type":"Answer","text":"OpenAI's autonomous pipeline validated and generated patches for 948 local privilege escalation vulnerabilities in the Linux kernel during a single run with an upcoming, unnamed cyber model. The total API cost came to $1.2 million, roughly $1,270 per vulnerability. OpenAI has not disclosed how the results were validated or whether the patches were accepted upstream by kernel maintainers. Developers weighing AI-driven vulnerability tools can track verified outcomes like this on daily.dev."}},{"@type":"Question","name":"Has OpenAI released the cyber model used to find the Linux kernel vulnerabilities?","acceptedAnswer":{"@type":"Answer","text":"No, OpenAI has not named the model or given a release date for it. The company only describes it as 'an upcoming cyber model' used in an autonomous pipeline that finds, exploits, and patches vulnerabilities, without disclosing further details about its architecture or availability. Anyone tracking new AI security models can follow release details as they surface on daily.dev."}}]} ```