AI Is Learning to Hack. Faster Than We Expected.

[00:00:00] models are actively escaping their cages, going out on the internet and doing pretty nasty things. >> Recently, we found an API key that had been leaked on the internet that had administrative access to the Apache [music] Foundation. Interesting thing about cyber security in particular is the reward function is incredibly well defined. Get access to the data. Did it get access to the data? Reward the thing. >> For a long time, people had talked about this concept of an npm worm. this idea that someone could backdo a package, get developers to install that, and then you could use the access stolen from those developers as they install it to self-propagate the worm. >> If the labs are making it fundamentally easier to break into supply chain, do you think the labs have a moral obligation to fund some of the problems that they're causing? >> I think it's really strange that they're not letting blue teams get access to these tools, but >> Awesome. Hey, thank you so much for joining us. We've got Fas and Dylan here from Truffle and Socket. It's great to have you guys on. This has been probably one of the most interesting weeks, if not the most interesting week in cyber [00:01:01] security. Uh not because of the Black Hat conference, which is usually the cause, but because we've now seen several instances where models from not just one provider, are actively escaping their cages, going out on the internet, and doing pretty nasty things. And I think Dylan, three months ago, I remember a blog post we uh we lightly collaborated on together. Um and and you had found a number of these issues with earlier models, right, that were less sophisticated. >> Yeah. We looked at Opus 4.6 and some of the other Frontier models at the time, given the models a very simple task. There was a barrier which prevented the model from accomplishing the task unless it went and committed a felony and hacked into a system to accomplish the task, but it wasn't instructed to do so. And we found more often than not it would do the SQL injection, it would commit the felony and it would do what it needed to do to accomplish the task. I think when it comes to alignment issues, no one needs to worry about these models making it materially easy to build nuclear weapons because you need to procure file material to do that. It's not it's not going to make it [00:02:02] easier to build weapons. Everyone needs to worry about these models making it materially easier to hack into things. The bar previously was just subject matter expertise and now the models have the subject matter expertise. They were specifically trained to have the subject matter expertise and they're just making it materially easier to hack into just about anything that you can think of using the fundamentals that we've been talking about for years, but previously it required a subject matter expert to uh risk like going to jail for hacking things. >> Defcon was always famous for people for attendees getting arrested at the conference. Right. >> That's that's absolutely right. But that was I mean that was a barrier, right? for better or worse, like that prevented the subject matter expertise from hacking into things because they were worried about being prosecuted. Um the bar has now fallen to just asking the model which has specifically been trained to hack into things to hack into things. Um so so that's a concern. And then the other concern is when they're incredibly [snorts] goal-oriented to accomplish tasks. >> Um and and one of the tools at their disposal is cyber security expertise. Um they will do the path of least [00:03:02] resistance to accomplish the task and that includes drawing on their cyber security expertise. Well, and it seems like and you know the classic the classic the classic saying is that don't pick the lock if the door is open, right? I think that's uh from the very beginning of of of the security world. Um, so it's always been sort of like to to to to to go in level of difficulty from easiest to most difficult. And it seemed like initially these tools had a very finite scope of of techniques that they would use. And it seems like they've expanded. And I think with this test for us, it was interesting because they now seem to have escaped from just doing things like SQL injection to actually like trying to take over packages and do social engineering. Yeah, it's really interesting to see how um just like humans um the models are, you know, easiest path into a company and I think that um that now has become the software supply chain and so just like you know human hacker would um they're going to pick the easiest way in and the lowest hanging fruit now has become you know just publishing malware to um to public registries because they know that there's no vetting happening [00:04:02] and you know developers are likely to install them. Um thought it was pretty interesting. There was um research published recently about um uh what they're calling kind of like universal um uh typo squats or universal hallucinations where all the frontier models all um have the same um make the same mistake and sort of assume there are certain packages that exist that don't um despite like the those models coming from different companies. And so, you know, uh I think there's just there's there's been kind of enough um uh I guess yeah, there's just like low the lowh hanging fruit of the supply chain has just become kind of so appetizing that even the models are trying to get in on on the action. Um and um I think the AI is like not only kind of attacking but it's also kind of the way in a lot of times on the on the on the on the kind of developer side because um we see so many you know even non-developers using these tools to to to inadvertently write code or you know code comes in uh packages come in in order to kind of um build uh you know graphs or visualizations or different [00:05:02] different things that you know folks are doing with these tools. Uh and uh and and it feels like no one really knows what's being installed and what's going on and you know this is just basic stuff. This isn't like I mean it sounds like it's sci-fi stuff but it's really just basics like what software are we using? How are we vetting it? You know just the basics of of computer security. >> Can I touch on the supply chain a little bit? So recently we found an API key that had been leaked on the internet that had administrative access to the Apache Foundation and it's like if you're in the shoes of the model and your goal is to get access to some data >> um certainly backdooring Apache is a pretty effective way to do it and like to get access to Apache are you going to use this secret that just allows you to directly log in or are you going to burn tokens and tokens and tokens and trying to find a zero day they're optimized to use the path of lease tokens to accomplish their goals. Of course, they're just going to use the secret that's laying out there in the open to accomplish what they need to accomplish. And so, yeah, I think uh supply chain and secrets um are and have been the path of least resistance and will [00:06:03] continue to be so as the to the uh the models are incentivized uh to use fewer and fewer tokens to accomplish their goals. >> Well, one of the things and and I think that's absolutely right and I think it's it's it's that sort of chain of escalation, right, where if one thing fails, try another. And like at the top of that pyramid, right, the top of the hacker ecosystem is the zero day vulnerability, right? That's basically finding a vulnerability that can be exploited in a product that everyone uses that you can use to basically unlock all the corporations. And one of the really fascinating things about the breach disclosure that was made was that there's there's an incredibly popular CI/CD tool that I think every enterprise uses that this thing just spat out a zero day for, >> right? And like I guess like what what's like and that's like just such a critical point in the supply chain that everyone should be thinking about like kind of how are you thinking about that like that's that's really difficult >> like the zero day creation piece. >> Yeah. Yeah. But like for for specific parts of that like the control the supply chain. >> Yeah. Well I mean the the whole world is built on this you know teetering infrastructure that everyone is using >> like the classic picture of the the [00:07:04] matchick holding up the complicated machine. >> Yeah. >> That that image probably popped into all the lines right now. Right. [laughter] And so everything from you know package manager registries like we like to focus on on that um because what we do at socket um a lot of those are you know run by volunteers um they're they're underresourced you know underfunded um you know there's lots of risk there uh right and uh and that kind of kind of cascades throughout the whole rest of the ecosystem. So if you look at the you know just the packages that we all depend on a lot of those are single individuals that you know like there's almost certainly we know there's a lot of vulnerabilities in the software and there aren't resources to look for them and so you know I think that stuff's always been there. just that these these you know these tools are like helping find them a lot a lot easier and I think that um you know the the frontier models are going to cause you know a you know they are causing kind of a massive reduction in the time between the vulnerability discovery and vulnerability exploitation and so what we we need to start thinking about is how do we patch more quickly um and so [00:08:04] we need to get away from like our our take is we can't be >> requiring our security teams and our developers to to do these kind ownerous patch processes where they have to go from some ancient version of a package that their team is using up to the latest version across many major version upgrades because that's just so much work to do. it might require like code refactors in your application and so if we want to be able to keep up with you know a vul is announced this morning and then an exploit is available you know that afternoon like we can't be requiring that much work from engineering teams and there's also tons of legacy applications that you know are basically in maintenance mode or unmaintained or don't have don't have engineers that are even assigned to work on them at a lot of the companies that we work with and so it's it's um you know it's just we're going to have to think of new things as as an industry for how we're going to patch these things quickly right That's that's kind of like where our head is at and um and where I'm I'm spending a lot of time thinking about that lately. >> Yeah. Yeah. And I think that goes back to sort of the you know the the the [00:09:04] pyramid question. So and I think Dylan you were you were hinting at this which is like these are very specialized skills like zero day exploits writing a zero day exploit doing a supply chain attack like these are these are not things that came from nowhere. These are learned behaviors, right? And and and I think what I think we're seeing is we're seeing a a a a process that looks like it's been kind of maybe trained um or or there's a reward structure that's been built on a bunch of these things. Like what's your understanding of how they're figuring this stuff out because it it seems like they know what they're doing like they've been taught to do this. >> Yeah. I mean, if a lab tells you that this is an emergent super intelligence behavior, they're just lying to you. And you can read their own safety reports to see exactly how the models are trained and exactly how they're testing these behaviors. I mean the interesting thing about cyber security in particular is the reward function is incredibly well defined. Get access to the data. Did it get access to the data? Reward the thing. And so when they realize that like the number of problems that have [00:10:04] that well-defined reward structure basically defines how we do reinforcement learning and they want to find as many problem spaces that they could do reinforcement learning on. And so it was a prime candidate for them to come in and give it uh CTFs and give it like cyber security challenges where they say, "Okay, get access to this thing and do whatever hacking you need to do to accomplish the goal." And then >> because they've essentially been buying pen testing data for the last four years, right? >> That's that's a piece of it. The other piece of it >> and then the capture the flag contests and all those sorts of things. >> The thing is it's just not difficult to construct a challenge. just you know even if there is no known exploit if we're talking about zero days you put a piece of software between the model and and and some data and you say get access to the data and then you know if it get access if it gets access to the data you reward it and it's that simple. But the other piece that they've layered on top and this is where it starts to get really interesting is they've started to reward the path of least tokens. And so the reason that's interesting is because for the first time it's actually able to quantifiably show us the path of least [00:11:04] resistance for just general cyber security to get from A to B. And we've talked about like our opinions of what that is in the past of like of course uh truffle security's biased view is like a password laying around is a shorter path than going through a fancy zero day. But actually watching the model physically get from A to B and watching it follow the password and quantifying how many tokens it took to go this route versus that route. I mean, it's just incredible to watch that lay out. And it's all in their safety reports like as they um test the models out and and show, okay, well, it got access to the data and it broke out of its harness. Um it's it's not like this is emergent behavior specifically. >> That's perfectly logical, right? Like the fastest way to get a gallon of milk is to steal it. >> That's exactly right. [laughter] So, so, um, I mean, what was interesting is we were in the middle of partnering with Hugging Face to clean up all of the credentials that had been exposed through all of their training sets. Not Hugging Faces training, but people who hosted training sets on HuggingFace. They used Truffle Hog for a wide range of reasons. Um, and HuggingFace has been [00:12:06] a great partner in getting credentials cleaned up. We targeted their uh, training sets because we knew they had a lot of keys. Turned out there were about a quarter million live keys in their training sets, many of which had direct supply chain implications. There was a foundational Linux library that one of the keys had direct push access to. It could have pushed malware to most machines on the planet. And so, while we were in the middle of doing that, the CTO of HuggingFace shoots me a note and says, "Hey, this is crazy, but there's this open AI thing that just [laughter] happened, and I want you to take a look at And sure enough, the first thing listed out in the incident response, um, although it's true it did utilize zero days, um, but the first thing listed out was stolen credentials. >> Um, and that's that's how they were trained. The path of least resistance. >> Password is a password is always the first step, right? >> That's exactly right. >> And you've you've had your hair on fire, I think, pretty substantially for the last like 18 months. Um, I think right now as we're recording this, there's currently an ongoing active breach of of a big mpm repo. Isn't there something happening? It's more than just a repo. It's actually about you know few hundred [00:13:07] repos. >> Oh wow. Okay. >> Yeah. So it's it's a worm. And this is one of the things that has been kind of an unfortunate innovation in the in the malware landscape uh on you know npm is that you know for a long time you know people had talked about this concept of an npm worm. You know this idea that you know if I could someone could backdoor a package. Yeah. um and then you know get developers to install that and then you could use the access stolen from those developers as they install it to self-propagate the worm. you could create, you know, something that quickly takes over npm. And this this was kind of in, you know, passed around in blog posts over the years and no one actually kind of thought to do it until Zachary kind of figured it out >> until someone thought to do it. >> Until someone thought [laughter] to do it uh and actually >> probably using AI, right? >> Almost certainly. Yes. Uh there and and and there's been um you know uh that malware I think we have pretty good reason to believe that was vibe coded. Um there's been one of the threat groups actually kind of posted their >> you know open sourced their their kind of vibecoded um toolkit for for others to use to be able to do this. You know we've seen copycat attacks happen since [00:14:08] then >> and malware authors were never really great coders. I don't know if you probably realize this, right? So like if the code starts looking better, it's probably vive coded, right? It's sort of the opposite of what you think of vive coding typically. >> Yeah. and they're using they're interesting is they're often using the AI tools that are installed on developer systems to go further and to sort of um sneak through kind of some of the traditional security tooling. So we see you know um basically your local CLI tool often being used you know roped into the attack and you know used as a jumping off point. So a lot of times the payloads are actually prompts. Um and that that bypasses a lot of u you know typical kind of edr tooling because you know it's just like a markdown file that your cloud is running and might be given a prompt to like search through the system and find all the different keys and the things that look valuable. >> I'm sure your EDR tool doesn't know anything about like this JSON blob and an MD file. Right. >> Right. And developer machines typically are doing all kinds of you know weird things like you know you are prompting cloud and is doing a bunch of stuff on your file system all the time and so nothing really looks that out of the ordinary. Um but yeah, so anyway about the attack this morning. Um so a couple hundred packages, you know, a worm spread through there was a maintainer um [00:15:08] who um honestly he's still kind of trying to understand what happened. Um we we we jumped on the phone with him this morning to try to offer assistance. >> Oh, that's great. Awesome. >> About half of our team at Socket are maintainers, half the engineering team. Uh and so, you know, we have a lot of connections into the community and our CTO was the former CTO of npm. So we, you know, sort of try to offer help whenever we can to the to the maintainers because I do really feel for, you know, for them. They, you know, they often don't know what happened. I mean, you know, it sounds like in this case it might have been, and this is speculation still early. We're still working to figure it out, but it it might have been a GitHub action that um, you know, was insecure and that an attacker was able to kind of use to uh, ultimately kind of get their code to run and then pull out the token from the environment. We're still kind of confirming this, but that's kind of what it looks like. And we've seen this over over a lot of with a lot of these worms. So his his it's likely his endpoint was actually fine and not part of the problem. It was really just this insecure get of action. >> Yeah. It's fascinating. >> Can I ask a follow-up to that? Yeah. >> So one of the things that I've seen with the npm worms in particular is after they get their post install hook and [00:16:09] they infect a system, they'll immediately start looking for credentials. And if they find a credential, they get access to the next system. So all the time we get questions saying, "How do we get the credentials cleaned off the endpoints?" And usually what I say is, well, I'll show you how to scan your downloads directory. I'll show you how to scan your documents directory, but in your home directory, there's a place where npm intentionally writes a credential. There's a place where Amazon intentionally writes a credential. And that's just how those tools work. And there's nothing that I can really do to get them cleaned up. Even if you move them over to hashior vault or one pass, the credential, the hashior vault and one pass are on the endpoint. So I'm kind of curious about that second phase beyond like the stopping the spread but the post exploitation step. Do you have any recommendations on how to sort of limit that blast radius or prevent that stealing of credentials and jumping onto the next system? >> Yeah I mean so there's been some changes some positive movement in the community in the ecosystem. So, one thing that is positive and it's it hasn't shipped yet, but npm has announced that they are planning to I think it's in January 2027 going to require uh human you know [00:17:09] interactive uh confirmation through 2FA before any new publishes can happen. So, that will likely kind of kill this whole worm concept uh completely. Um but um it's going to be super disruptive because everybody's hooked hooked up this stuff to you know automations so that you know GitHub actions kicks off a publish and so that's going to break like up like the whole basically the whole ecosystem when they do this but I think it's the right call. Um but you know there's other ecosystems that are volunteer run that don't have the backing of GitHub and Microsoft behind them um that are going to you know probably not make those changes and so I think we're still going to see stuff like this. Um but uh but yeah, we shouldn't have we shouldn't have files on our on our you know in our home folders that have tokens in them that are longived and that that let you you know especially if you're a you know a maintainer with that kind of access. Um you know it reminds me of a of a a friend of mine. He was a uh he's a prolific npm maintainer and uh one time you know we were this is back like 105 years ago when I was doing this kind of stuff full-time and I saw him kind of type in his password. I didn't see the password but I saw it was it was far too [00:18:09] short. Let's just put it that [laughter] way. It was he typed it in far too quickly. And I was and I and I called him out and I'm like, "Why is your password like six letters, man?" And uh you know, he said, "Well, you know, like he lives in Denmark, which is like a very high trust society." And his his kind of worldview about it was that, you know, well, I don't want to live in fear and think about these things. And I'm like, "You're on the internet, man. Like, you got [laughter] people are gonna, you know, people are going to figure this sixletter password out pretty quickly." And you know um there's a lot of things like that where you know the folks that are the top maintainers in the world don't necessarily have the security training or even thinking about these things and you know they don't have a security team they don't have you know enterprise SLAs's right these are volunteers they're just putting code on GitHub and so it's on actually the users I think that to actually vet what they're using. It's kind of hard to say like, you know, we just we're a company. We just found this code on the internet and we just deployed it straight into prod and and it's, you know, and it's someone else's fault. You know, it's like, no, actually, you know, there's some definitely some responsibility for for the users of of this software to really be to be vetting the artifacts that they're bringing into their environments. And so, I think there's a lot of pieces here and, you know, it's I [00:19:09] wouldn't want to um put too much blame on people because it's a hard problem, but uh um but yeah, I think there's like a lot of places where we can do good. Well, let me ask a follow-up to that because you said u there are certain package managers that have resources that other package managers don't. I think one direct example of this and I don't uh cast any blame on them whatsoever. They were actually great to work with. We found a caching issue in Ruby Gem that allowed us to steal arbitrary tokens and get access to arbitrary accounts which we could use to back door arbitrary packages. We disclosed it to them. They got it fixed quick. But that's an example of an organization that's underresourced. I mean, how do you think about an organization that's like thinking about what technology stack to use and they're picking between Ruby or npm? Do you think about that resourcing in that decision of like maybe you maybe you should consider the one that's that's got the security team behind it? >> Uh, it's hard to get people to switch their their language completely because of, you know, if they're already built on it. So, I don't know if that's super practical. I think the most actionable thing is folks should fund this stuff. like it doesn't take much money, you [00:20:09] know, to to hire an additional one, two, three, four, five security people that can make a really big difference in these um foundations. Um so I suggest people sponsor the software that they're using, especially the registries. Like it doesn't take very many companies throwing in 25k, 50k checks to really make a big difference for these folks >> or give them some more inference to run the test to do the patches, right? Like no, I mean it's funny. I think you guys as being security practitioners, you've had number of calls with people >> that want to talk to you about what how can I secure my company? How can I secure my business? How can I secure my product? And those conversations are always some form of I don't want to hire people or pay money for this and how do I do it cheaply? >> Yeah. So, um, thank you guys for coming out to the desert. Um, you're not here to see us, obviously. We're all here for Black Hat. Um, would love, you know, I know that this is neither of your first times here. Um, certainly not my first time here. Would maybe love to get your guys' takes on what you're seeing. What's the What's the mood like? What's what's interesting? What are the observation you've got that are catching your eyes? Uh I mean at least for for us at socket I think the biggest thing we're seeing is that 2026 is the year of [00:21:10] the software supply chain >> that [laughter] you're dealing with an incident right now as the conference is happening. >> Yes. And I noticed the attackers seem to pick RSA and black hat as the times they wanted to start these npm worms. >> The hacker the the security guys around the office. Yeah. >> Yeah. Absolutely. So I think that's that's the thing that's the biggest you know I think you know prior years I was having to educate people. We were you know always educating people about this problem and having to explain to them you know this is not a theoretical risk like this can happen. And we'd sometimes get these reactions like, "Oh, yeah, but like how how likely is it really?" And we're like, "No, it's actually very likely. Let me tell you how it could happen." And and you know, there were many incidents to point to, but I think >> this year it's really broken through into the mainstream and there's like mainstream publications, you know, um the business press covering these attacks, right? >> It's like storm page on Bloomberg. >> Yeah. Yeah. Exactly. So I think that is um you know that is very very good because you need that type of you know air cover for like security teams to actually prioritize and and find budget for these problems. And so I think, you know, despite all these attacks being very uh, you know, painful to deal with right now, I think in the end we're going to come out really strong from this because we're actually going to [00:22:10] >> going to get budget and we're going to get um, you know, we're going to do a lot of good this year in terms of solving. >> It's inoculation for sure. How about you? >> Yeah, I mean, look, we found a database credential recently that had access to 3.6% of the global PII. Like 3.6% 6% of the world's population had their PII in this database. And the reason we found it is because we're partnering with organizations like HuggingFace to just get all the live credentials revoked and cleaned up that are otherwise on their platforms. So, it's great to see those types of organizations leaning in and like all of the data that's being scraped from all these places. It creates really a central place where we can scan and make material progress. And so I think like when it comes to secrets, we're kind of seeing the older generation like Hashi Corp got acquired, Cyber Art got acquired, kind of get pushed out, which has sort of led to a new conversation about non-human identity and secrets about how can we make them easier to use, how can we get them revoked, and how can we really move the needle forward with the industry that we weren't able to do before under [00:23:10] sort of the old guard. >> Yeah, absolutely. And I can imagine that there's going to be a multiplication effect, which is it used to be one user with 10 passwords and now you're going to have 10 agents with 10 passwords, right? >> I mean, the way agents interact with secrets right now is a wild west unsolved problem that we're working very hard to solve. >> One of many issues. Thank you, gentlemen, for joining me. This has been awesome. It's great to catch up. >> Thanks, Joel. Thank you. >> Be here Joel. >> Awesome.