Jul 30, 2026, 4:06 PMThe whole Hugging Face/Openai thing just reeks of scam. It's a publicity stunt, it helps keep them in the media, trying to scare people into being wowed by their horrible next word picker.
folks are assuming that it must of happened... because they said so?
If it happened, it would be fucking CFAA felony.
And there's no way to know if it's totally cleaned up on HF's side, *that's not how compromises of infrastructure work*
Jul 30, 2026, 4:08 PMFixing that takes weeks and real money and is stressful as hell. Is anything still in your infra? You don't know until you've looked. It's shit, it's horrible, there's a reason it's a fucking crime.
...unless, of course, you just planned the whole thing out in advance for the lols with another company
God this timeline is stupid.
Jul 30, 2026, 4:09 PMI cannot possibly be the only person who doesn't trust Sam Altman. Though sometimes it feels like I am.
Jul 30, 2026, 4:11 PM@quinn he's got a scammer aura. I cannot believe people fall for his bullshit...
I don't trust you or Sam Altman. But I trust your judgement in not trusting Sam Altman.
(Note, this is a joke and I have never been given a reason not to trust you. Sam Altman, on the other hand...)
Jul 30, 2026, 8:19 PM@Perrin42 damn yes! I could get up to anything.... that fits in a $50 budget 😂
@quinn i don't trust him either. for all he is, oozes out, represents, since one of the first time i heard of him, that is his firing/reintegration, and especially since i learned about him from his sister Annie on the dead bird site, since 2022?
@quinn wait there are people who *do* trust Sam Altman?
Jul 30, 2026, 5:59 PM@lkanies there's people who fucking worship him.
Jul 30, 2026, 8:17 PM@lkanies investors
@quinn I guess they trust him to make number go up. I am not sure investors let their actual feelings influence their decisions
Jul 30, 2026, 10:08 PM@lkanies or basic, 4th grader understanding of morality
@quinn His own board fired him but were too afraid to say exactly why. That's the most ominous thing I've ever heard of.
Jul 30, 2026, 8:16 PM@djohngo and then his employees brought him back, it's kind of culty.
@quinn A lot of those employees are going to realize generational wealth when the IPO happens, so that kind of made sense to me.
Jul 30, 2026, 5:36 PM
Jul 30, 2026, 5:44 PM@quinn There is a certain category of bullshitters who do not appear particularly smart, and then they get away with saying some mind-crumbling shit because people can't get themselves to say 'you are talking shit! The ideas you bring are obvious nonsense, and your sentences are not actually sentences'.
And if you have made it to millions and billions and people are making money subscribing to that bullshit, it is not strange if you end up buying into it yourself...
Jul 30, 2026, 8:15 PM@mjj And then there's a humiliation of trying to back away.
@quinn Altman graduated from the cryptocurrency school of grifters to a larger and more majestically lucrative scam, the prosperity gospel/basilisk's hell branch of AI Jesus christian syncretism. He's a televangelist-style huckster, is my take.
@cstross OT: any relationship between that view and your several-years mostly-absence from HN by chance?
(Not to put you on the spot, but something I'd noticed a while back. And realise writing this you'd stopped by to discuss Accelerando a few months ago: https://news.ycombinator.com/item?id=48159241.)
@dredmorbius Yes: HN is overrun by right with fascist-adjacent techbros these days, and I want nothing to do with them or their ideology.
@dredmorbius @cstross ycombinator is even listed as a cosignatory with other baddies like Microslop, Google, OpenAI and the Linux foundation in some fascist-adjacent recent thing (forgot which)
@mirabilos Feel free to ping me if you run across it.
@cstross I suspected.
Thanks.
@quinn Clammy Sammy? Untrustworthy? Nah, my millions are safe in his hands.
@quinn Not sure where, but I read a great article on Sam Altman being a pathological liar, citing people who used to work for him like "He would tell you something and 5min later, during the same conversation, claim he never said such a thing".
Jul 30, 2026, 10:06 PM@logwyrm That's a sought for, valuable quality in Today's America.
@quinn We are many…
@quinn Most people in my circles also feel the same, FWIW.
Almost nobody trusts Sam Altman.
Even the stock speculators don't trust Sam Altman - they just believe they can cash out before it all comes crashing down.
@quinn There are people who *do* trust Sam Altman?!
Jul 30, 2026, 9:30 PM@quinn noooo, just drop that 17k logs Ingo an other AI Model. Problem solved. :blobcatnotlikethis:
@quinn furiously checking the stock price as i talk about how dangerous my ai would be if not for the stewardship of my [LEGALLY, I CANNOT DESCRIBE THIS AS "FRAUD"] non-profit
@quinn I’m convinced it has a secondary purpose too: testing the boundaries on what they can skirt accountability for by shifting blame to “AI” - “The AI did it, not us!”
Jul 30, 2026, 10:12 PM@BurritoSommelier product liability law wheee
@quinn I loathe the language pretending it's sentient...'cheated'. As though it actually had a goal that it wanted to reach.
Jul 30, 2026, 7:29 PM@quinn
Aye. When news broke out, I said "lawsuit or didn't happen".
Jul 30, 2026, 8:19 PM@momo extremely well put.
@quinn This thread was great for me to read. I am not a tech person, but when I heard this news I said to people Why would anybody believe him? My assumption being that they did attack HF and are just blaming the AI. There are better theories here. Everybody just looked at me like I had two heads, nice to know there's just one and it's very suspect of liars
@quinn there will be a very interesting argument about the knowledge & intent elements of the various CFAA violations, whenever this happens for real.
Jul 30, 2026, 10:11 PM@copiesofcopies it never real. that i learned many years ago to my eternal pain.
Jul 30, 2026, 9:13 PM@quinn Oh, you're just saying that because, as soon as the story faded out of the news cycle, they suddenly thought to look at the logs and, wouldn't you know it, it actually broke into a bunch of other systems. And no, you can't see the evidence of that, either.
We need a Frankenstein remake, where Victor keeps screaming about the horrible monster that he's created. No, you can't see it, but it's totally there, and you can invest in his IPO to benefit from the monstrousness...
@quinn isn’t that their whole thing? Oh nooo, AI is too powerful, regulate us! Oh nooo, AI is gonna find all the security flaws! Teehee, look how powerful!
@quinn it's possible that this is a publicity stunt, but it's a high risk, lowish reward one. It requires a lot of people to stay very quiet, including some people who won't necessarily have a reason to stay loyal in the next year or two (when they get laid off). With openai making a run at an IPO, making up a story like this is a risky move.
If it did happen, it certainly *should* be a felony, but I'm at least partially persuaded that it isn't a CFAA violation because there was seemingly no intent on the part of a human to violate the law, and the CFAA requires the prosecution to prove intent. It's entirely plausible that openai's lawyers took a careful look and decided that, at the most, some low-level mook *might* end up carrying the can, not the company, but most likely it's not criminal. So sure, let's milk it for all it's worth.
Now, was there gross negligence? Absolutely, but "I am a complete moron" isn't CFAA territory, it's (mostly) a civil matter, and no doubt hugging face and openai stitched up a quiet deal before openai admitted any kind of culpability.
On the "intent" front, it's tempting to say that the model formed the intent, but allowing that precedent opens a huge can of worms for everyone. How do you punish the spicy autocomplete? The mental image of a data centre in a giant prison cell is amusing, but unlikely, and doesn't solve the problem that humans are being reckless.
I'm not arguing, by the way, that this shouldn't be a crime, because hoo boy it absolutely should be. There's even precedent for holding humans accountable for the actions of a non-human entity (dangerous dogs, for example). It's just that the arguments I've seen thus far are more persuasive that this event, if it happened, was not, in fact, covered by any existing law.
Jul 30, 2026, 10:28 PM@womble "I am an idiot" is very very much in CFAA/wire fraud territory, and intent isn't really at issue. It is absolutely covered by plenty of existing law, just whether any AG wants to pick up the case.
I know maybe a bit too much about this.
@quinn every subpart of 18U.S.C. 1030 (a) includes the word "intentionally", or the phrase "knowingly and with intent", which would seem to suggest that intent is at issue. What's your legal theory for how this event is a felony under the CFAA?
Jul 31, 2026, 8:55 AM@womble have dealt with a few CFAA cases irl (quite famously in one case) and intent isn't much of a defense irl. Though Sony (rootkit) did teach me that CFAA is not a thing for companies, companies are allowed to hack all they want, you know, like crowdstrike.
That it is one rule for me, not thee might be the state of play, it's still corruption and wrong.
@quinn I think this kinda cynicism is a mask for ignorance.
Jul 30, 2026, 10:46 PM@budududuroiu Then I'm probably not the follow for you, and nothing anyone could tell you is likely to change your mind.
@quinn Having read the HuggingFace write up with a healthy amount of scepticism, I'm pretty sure it's true. There are too many things that are embarrassing for both sides for it to be a marketing stunt:
- Model made an incorrect assumption (answers are hidden on Hugging Face somewhere) and proceeded to burn hundreds of millions or even billions of tokens based on this
- OpenAI had minimal security and almost no oversight of what their model under test was doing
Jul 30, 2026, 10:53 PM@bashtoni do you have a link? All I found was this https://huggingface.co/blog/security-incident-july-2026 and it was deeply unconvincing on its own.
@quinn The full write up is at https://huggingface.co/blog/agent-intrusion-technical-timeline
Jul 30, 2026, 11:49 PM@bashtoni thanks, will read it after i get some sleep. :)
Aug 1, 2026, 1:00 PM@bashtoni Ok, so me and my (former) CERT operator partner have gone through it & both of us don't feel like either the language or series of events seems quite right. The descriptions are overfitting what they describe, it's kind of got a 'one of each kind of attack' approach. Maybe the companies did actually do it, but the corps stage managed it? The whole write up has a lot of fluff that doesn't make much sense, which is probably the AI write up, but it's not really more plausible for it.
@quinn I know where you're coming from, although I think much of the general "off" feeling is just the bad AI writing which spends too much time on mundane details and not enough on what is important. That's pretty usual for AI generated writing.
By the way the "third party provider" used for code execution to escape the sandbox was apparently Modal: https://www.reuters.com/business/openais-rogue-agent-compromised-an-account-second-tech-firm-sources-say-2026-07-28/
Aug 1, 2026, 9:15 PM@bashtoni I'm not convinced, still think it was probably staged/orchestrated, but I feel like I understand better what the performance was trying to do.
@quinn I'm not here to try and change your mind, but I'd say I'm ~90% sure it really happened. I was just as sceptical as you before I read the full write up, after which my main thought was wondering how many times before this had happened without them noticing.
I don't know if you saw https://www.anthropic.com/news/investigating-incidents-cybersecurity-evals yet?
@quinn OpenAI talk on this from Blackhat just posted: https://youtu.be/87DyyMV0kCY?si=C1hHkQf4U9iESHi9
Aug 7, 2026, 9:33 AM@bashtoni the fundamental problem is that I don't trust these AI companies not to lie, which they have demonstrably done repeatedly on many topics, including their own finances and product safety.