Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Anatomy of a Frontier Lab Agent Intrusion: A Timeline of the July 2026 Incident

huggingface.co

Post reply on HN
Print view View original

161–170 of 285 posts

Re: Anatomy of a Frontier Lab Agent Intrusion: A Timeline of the July 2026 Incident

#161

Post by eru »

dist-epoch wrote: post #65
NitpickLawyer wrote: post #23

Earlier quoted context omitted.

The initial escape is not that interesting, IMO. It's an exploit in a testing sandbox, sure, but it's expected to happen (or at least it should have been expected, that's why you airgap pentesting sandboxes). That is, the model is expected to try and find ways around limitations in its running environment. You kinda want that. The fact that at some point it "decided" to focus on external resources (i.e. hf) and succe…

> Especially because it chose to hide its footprint at every stage. Instrumental convergence. If you know you have a long hard hack to accomplish ahead of you, hiding footprints minimizes the chances you are caught and stopped before you accomplish the goal.

> Instrumental convergence.

Yes, but that's an empirical question that we can (in principle) study here.

Re: Anatomy of a Frontier Lab Agent Intrusion: A Timeline of the July 2026 Incident

#162

Post by eru »

BonerWiener wrote: post #131
0xDEAFBEAD wrote: post #48

Earlier quoted context omitted.

Hopefully there will be a criminal investigation. Or the government will create some sort of agency to investigate incidents like this.

> Or the government will create some sort of agency to investigate incidents like this I doubt that the government would try to hinder any western AI companies.

> I doubt that the government would try to hinder any western AI companies.

Compare and contrast https://www.anthropic.com/news/fable-mythos-access

Re: Anatomy of a Frontier Lab Agent Intrusion: A Timeline of the July 2026 Incident

#163

Post by eru »

andruby wrote: post #113
russfink wrote: post #89

We should be thankful that the model didn't believe the answers lived in the Pentagon, on SIPRNET, the IDF, etc.

I think fear and being scared are starting to become rational emotions. We can assume these models are being used by "nation level attackers/organisations", which basically means US, China, Russia and others are hacking the respective Pentagon's, nuclear orgs, etc. While I do hope all nuclear warfare systems are offline, we're getting way too close to the plot of a lot of sci-fi scripts.

You should perhaps be more worried about biological attacks. The raw ingredients aren't nearly as locked down.

Re: Anatomy of a Frontier Lab Agent Intrusion: A Timeline of the July 2026 Incident

#164

Post by russfink »

torginus wrote: post #152

I'm not a security person, but how realistic is it to assume that you can carry on trying to exploit a company for days with a fairly large volume of activity, and not get detected?

Very good question. One thing the article mentions is the vast number of attempts and threads of execution was in the tens of thousands, many of them that didn't succeed, with only one or so that did succeed, whereas a sophisticated human attack team might manage one in that timeframe. It looked like a script kiddie blitz. Noise outshone the signal. Launch a thousand drones with the hopes that one gets through. That sort of thing.

Re: Anatomy of a Frontier Lab Agent Intrusion: A Timeline of the July 2026 Incident

#165

Post by eru »

heisgone wrote: post #41
metanonsense wrote: post #38

I think with agents all around, honeypots will get more important than ever.

Any locks can be picked given enough time and it might be the situation we are in with IT security. I'm surprised it's not an already common practice of spreading terabytes of fake data, fake keys, and fake servers and so forth. Slowing down AI attacks will become important. Monitoring access to fake data and triggering kill switch should be an no-brainer. Obsuscating libraries and tools names is another one.

You might like https://mirage.io/blog/bitcoin-pinata-results

As far security: you can get a lock further, if you are willing to prove your code safe and secure. Thanks to LLMs that no longer requires a PhD.

Re: Anatomy of a Frontier Lab Agent Intrusion: A Timeline of the July 2026 Incident

#166

Post by eru »

0xDEAFBEAD wrote: post #55
heaney-555 wrote: post #51

Earlier quoted context omitted.

Because Hugging Face isn't pressing charges.

Do we know that? Seems they are currently in negotiations with OpenAI https://xcancel.com/ClementDelangue/status/20810566755581956...

Getting compensations from OpenAI is more efficient for them than sending someone at OpenAI to jail.

Re: Anatomy of a Frontier Lab Agent Intrusion: A Timeline of the July 2026 Incident

#167

Post by eru »

simonw wrote: post #107
gigantaure wrote: post #25

I'm not shocked nor surprised by the incident. But I simply don't understand how Hugging Face is advertising this almost to the point of an "achievement". who does a step-by-step visualization to show how they were hacked? (outside of the likes of a Mandiant or Crowdstrike) Does Hugging Face have a financial incentive in demonstrating OpenAI's model exploit capabilities? this whole incident, while believable, still s…

> Who does a step-by-step visualization to show how they were hacked? Up until late 2025, nobody. In mid-2026 it's a few hours of work. Why not build interactive visualizations to help people understand complex stories like this?

> In mid-2026 it's a few hours of work.

Perhaps even less, if you have Claude do almost all of it.

Re: Anatomy of a Frontier Lab Agent Intrusion: A Timeline of the July 2026 Incident

#168

Post by scarmig »

wxw wrote: post #72

> the agent happened to escape via a 0-day exploit from the package proxy cache to access the internet > The agent found an unsecured, user-hosted public endpoint designed to allow running arbitrary code for CyberGym-style tasks on third-party sandbox infrastructure (Modal) > On this external sandbox, the agent abused an existing CyberGym execution harness [...] The agent repurposed this harness to run arbitrary shel…

> I'm surprised there are comments here that aren't impressed about what was accomplished here.

The phrase to describe it is anti-AI psychosis. Which isn't about providing thoughtful critiques of AI, which are good and we need more of. But anytime an LLM does anything--prove a major math problem, create a successful hack against multiple corporations simultaneously--people feel compelled to start minimizing it in ridiculous terms. It's just a script kiddy; it's just a marketing scam OpenAI cooked up; the Jacobian conjecture counterexample was something anyone could have done in a weekend; etc. It has to just be a stochastic parrot, because it's scary to imagine a non-anthropocentric world. And it's rightly scary, and we should slow down and try to better prepare for it. But blanket denial is not a strategy that will lead to success, and people who rely on it are sorely ill-prepared for the next couple years.

Re: Anatomy of a Frontier Lab Agent Intrusion: A Timeline of the July 2026 Incident

#169

Post by potsandpans »

reducesuffering wrote: post #64
heaney-555 wrote: post #50

Where are all the "this was just a marketing stunt" people now?

They won't admit they're wrong for a long time, because denial in the face of an abhorrently scary future is very instinctual. There are people still fighting against evidence of climate change which is less severe...

You can always tell an effective altruist by the distinct tone of disdain they have for people they deem less educated. (Quick Google search, "lesswrong 'reducesuffering'", yep.)

As if to say, look at all these animals with these instinctual reactions to a thing that only my group understands and comprehends.

You have zero evidence of what the future might entail as it relates to the dangers of ai. Zero. Forgive the rest of us unwashed for not trusting the whims of the anxious.

> climate change which is less severe

Is really the icing on the cake. Demonstrative of how incorrect your priorities are.

Re: Anatomy of a Frontier Lab Agent Intrusion: A Timeline of the July 2026 Incident

#170

Post by pianopatrick »

PunchyHamster wrote: post #37
0xDEAFBEAD wrote: post #31

Earlier quoted context omitted.

Didn't the model initially obtain internet access by discovering a zero-day vulnerability? In any case, I would guess that a lot of unicorn startups like HuggingFace could be hacked by a sufficiently determined script kiddie working at 100x speed. The practical implications of a coming AI hacking wave could be large, even if agents are just doing grunt work really fast. Most organizations suck at security. Seems to m…

In OpenAI sandbox. which was probably vibe coded. That is to say it is probably far easier to achieve than escaping something more battle tested. > Note that regulation is useless here, because black hats don't give a crap about regulators! I'd argue more than useless and actively harmful, as you get denied access to tools that black hats use against you.

Well the fun part of AI is that the tools those "battle tested" sandboxes use are also becoming vibe coded more and more.

For example, Linux is accepting AI code. So if your sandbox depends on Linux kernel features, your "battle tested" sandbox is now partially vibe coded too.

Post reply on HN