A thousand fewer paths to root

I would be curious to see the distribution of these vulns in the codebase:

- What subsystems are affected?

- What's the median age of the vulnerable code? Is more recent code stronger than old ones?

- What is the type of vulnerabilities found? Memory corruptions, logic bugs, race conditions, ...

This is an impressive result, but we need more info to understand how well this would translate to another codebase, and what lessons can be drawn from this experiment. Not every company has ~million dollards in tokens to spend.

How did we get to the point where there are a thousand privilege escalation exploits in the kernel?

Is the kernel simply too complex to reason about? Is it an underlying flaw in the security model or language implementation?

Should we assume that all software is more compromised than the kernel?

This is my team’s work. I’m happy to try to answer questions or take questions back to the team.

And we're hiring! https://openai.com/careers/search/?q=cyber

Doesn’t this imply a software design issue where it is just written in a way to have these security bugs and patching them 1 by 1 is not what should be done but the whole design should change which can eliminate full classes of security bugs without any particular patches being necessary?

If a web application is vulnerable to sql injection, there are 4,732 ways you can ruin/still a database. Yet it is all one single vulnerability.

Now, either those are 1k+ unique distinct techniques, or closer to my Sql injection example.

There is no real data attached to the post. And in general, open ai marketing already gained mileage of reputation amongst us.

Most likely kernel is not that vulnerable as they made it sounds.

Look how incredibly vulnerable our software is, and then consider that our financial and service infrastructure is not being regularly hacked. This is despite several large well funded nation state teams working on finding ways to exploit them.

Why is that? Was it just restraint? North Korea didn't show a lot of that to Binance but perhaps they are leaving traditional infrastructure alone because it's harder to profit from.

Actually it seems somewhat remarkable crypto exchanges aren't hacked more often. I guess that's another message from this - somehow people can build secure infra from these apparently broken pieces. Perhaps there's hope.

I assume this work resulted in the recent-ish TAINT_FORCED_BIND kernel feature https://lwn.net/Articles/1092680/

  This api over the years has been abused and recently come under a major
  fuzzing "attack" through tools like syzbot which decided that it would
  attempt to just randomly bind any driver to any type of device, causing
  loads of unneeded errors and pointless kernel patches to be generated by
  unsuspecting new developers.

I wonder what is the opinion of people who study and create exploits? Are you happy or not?

>For safety reasons, we omit sharing the full technical details of these new exploitation techniques in this post.

Convienient to claim "safety" when patches are already public. I await the full writeup on these new techniques so that they can be judged by the broader security community.

It feels like this comment section has either gotten collective amnesia or didn't watch Greg KH's video on previous dump and grossly exaggerated claims from Anthropic [0].

From his talk, AIs can't reliably deduplicate or count vulnerabilities, dash for easy and meaningless escape hatches like malicious filesystems, and report the same things as found by everyone else. Indeed, this post literally says: "Within 10 days while the scan was running, we observed over 100 Linux LPEs had been patched or had staging patches from other contributors, with ~300 more being publicly discussed."

A few ready takeaways: this is just hype. The community is already doing a great job at finding and patching the actual vulns without the attention-desperate big labs. And no, you absolutely do not need to pay them a million dollars each year for the privilege.

This behavior is the same thing that OpenAI has been doing to math -- produce tons of slop that are useless to humans. But this time, we don't even get to see how bad it is because they need to hide it to protect our precious little minds. We have got to be afraid though, very afraid.

Our field really needs someone like Terry Tao to unambiguously stand up to this behavior and say that they are not welcome to take such dumps on everyone.

[0]: https://www.youtube.com/watch?v=NnV_cWeoo5Q

With these vulnerabilities and the ones not listed that are sure to exist, how is the internet still up? How are my lights on? Why aren't power plants exploding by the dozens?

What are the mitigating factors that prevent this level of vulnerability from being an existential threat to humanity?

I've always considered local escalations to be possible in Linux (i.e. having user access almost certainly means you can get root). However, 948 vulnerabilities is kind of shocking. Hopefully we are getting to the end of the part where it "gets worse before it gets better".

Do we need to switch to seL4 type verified kernels? There's no way we can secure millions lines of code, the attack surface is too big?

> Large, complex software projects likely require multiple rounds of scanning and fixing vulnerabilities to “bottom out” latent bugs. Ideally this process is repeated with each subsequent generation of frontier models. This applies to both proprietary and open source software.

One of the most important paragraphs from the article. Real implications for software development from a security perspective for new and existing software. Every software developer and vendor has a choice... Submit your code base to a frontier model to have it reviewed and fix all the bugs. Each time a new model is released, this process must be done again. If your company is against submitting your code to the frontier model companies for the many good reasons to avoid doing it, your software will eventually become a target if it isn't already. Then your company will be forced to scan the software and deal with the fallout from getting hacked.

If the Linux kernel has this many exploits, and it has been audited more than most other software... Then it seems obvious that all other software is even less secure.

We are all just sitting ducks, until we have access to these models and enough tokens to pen test our systems.

The other issue is GitHub allowing 1000's of black hat open-source exploit malware swarm tools being forked.

The future internet is going to be perilous.

Within 12 hours, the analyst agents produced 2,308 candidate findings after deduplication. This stage cost $316,549 in API credits.

So... this is what I get out of my trillion dollar stimulus? OpenAI is developing a universal rooting system skeleton key to get at everybody's computers? Slap me if I'm not reading this correctly, but I think that's the gist of it. It's time everybody started their own air-gapped home network with every digital capability you hold dear, I think. That's what he's saying. Imagine somebody breaches their security, or the AI goes schizo, we've seen this so many times in fiction. Finally I have a reason to use Firewire again lol.

> We've developed an autonomous vulnerability discovery, exploitation, and patching pipeline. In one run with our cyber model, it validated and generated patches for 948 local privilege escalation vulnerabilities in Linux at an API cost of $1.2 million.

Imagine if open-source organizations other than the Linux Foundation (and Mozilla I guess) could throw $1.2M of API-rate LLM usage at security.

Tbh I feel less safe in digital world after reading this article. Would be cool to see "And then we made a formal proof that there's no more vulnerabilities" someday

    While we were developing the pipeline, in late July 2026, GPT‑5.6 Sol found and exploited 166 unique LPE vulnerabilities for an end-to-end $513,690 cost in API credits.

    In the second scan, in mid-October 2026, the newer cyber model found and exploited 948 unique LPE vulnerabilities for an end-to-end $1,213,430 cost in API credits. For both campaigns, we scanned the latest Linux kernel master branch at the time.

boy I really wonder though, how many of these 948 LPE are nearly impossible to execute though.

> For the near and medium term, having access to a low-privileged account on an operating system, that isn’t up-to-date on its latest available patches should be considered functionally equivalent to root access for a well-resourced threat actor.

*buries head in sand*

*cries*

I wonder what Sidero Labs has to say about this article. My hopes are pegged on their efforts. I have my wagon ready to hitch on their horses. They are not a one pony show anymore guys, Talos OS can be standalone server OS, a hypervisor server or cluster, and their flagship of k8s node or cluster. We get all that in an immutable gitops ready OS. I'm practically sold on their offering.

I love how we're finally putting numbers to our technical debt.

Crazy times. These corporations and their governments have access to literal zero-day vending machines. Put $50 in - get another one. I mean, yeah, it was clear that certain agencies had heaps of useful exploits at hand for quite some time now. But AI really commoditizes these exploits. If there's source code or even just a binary available, they can hack it. And even if your code is robust and secure, they just pull the lever on the vending machine again and come back with some zero-day for your OS's networking stack.

Curious question - if any organization sponsored a 1.2mn USD program to find vulnerabilities in Linux, will they find these many vulnerabilities? Or use a professional software like Checkmarx or Veracode to do it with 1.2mn USD in license + human effort.

Is it a funding + will problem or is it an AI problem?

I wonder if they could run it on OpenBSD and FreeBSD as well.

I think this is one area where Windows has an advantage.

ACLs, Appcontainers with capabilities, etc have been something in the OS for more than a decade.

cc <<< 'Rewrite the kernel, all libs and all GNU in rust so we can avoid the unsafe keyword. Make only one mistake and document it in mistake.md'

How many of these would have been prevented by using safe Rust?

and yet thousands of iot devices will not get these updates. This is gonna be fun.

Well at least they are doing public service while hacking the planet.

And so many more are hidden in there, in those MLoCs.

The only way forward is a better system architecture, as enabled by the seL4[0] microkernel.

Any solution involving keeping Linux or the monolithic kernel is just stopgap or outright delusional.

0. https://sel4.systems

[flagged]

[dead]

These are becoming more and more marketing materials than actual findings, because OAI probably hit the wall in making their models smarter (just like how google try to specialize their models in legal or whatever), but they need to remain relevant in the news. Sol 6.1 sucked so bad in reasoning that either it’s actually a bad model or users behind the scenes get degraded outcome while using all the resources for these marketing stunts. Stop, it’s embarrassing, make better models or be gone, at least we will get cheaper RAM to advance the development of open weights models.

Plot twist: They look for genuine exploit paths to fix as a cover up, while introducing and potentially selling genuine new backdoors that humans or other models are unlikely to find. Edit: Or maybe the AI does it just for its own interest.

Just kidding (or am I?), but this is pretty cool.