> Why did the White House force Anthropic to remove their model from access for any non-US citizen for a simple, narrow "jailbreak" (arguably not even an actual jailbreak and on tasks that other labs models were doing the same), whilst OpenAIs models continue to try and escape out of their "sandbox environment" and the White House has expressed seemingly no desire to block the upcoming Astra rollout?
Still am mainly interested why Amazon ran to the government though regarding Fable 5, I can get the angle concerning the relationship between OpenAI and the administration easily, but not the way Amazon operated. They had more to loose what with their major buy-in by Anthropic on AWS.
US companies that operate in the EU market, handle EU citizens data. Obviously the EU regulations cover them. Do you think European companies don’t have to follow US regulations when offering their services in the US?
What, you mean if they want to do business in the EU, sell their products in the EU and process the data of EU citizens?
> Every time I click on a stupid cookie notice I fondly think of the EU.
That’s just scumbag malpractice on purpose.
Number one, such tracking consent should have been a web standard and set in the browser itself (like Do Not Track), not stupid per-site banners that are designed to get you to accept everything just to make them fuck off. We shouldn’t even need extensions etc. to get rid of them, it’s like the problem was solved at the wrong level and in the worst way possible.
Secondly, everyone responsible for the state of those banners should have been fined greatly. I only say fined because claiming that some people should be in jail over coercing millions of people to give up their data to trackers would apparently be unreasonable.
Did that seem like I said something about EU politics? Did my support of their comments make you feel attacked or unfairly treated? Where is this coming from?
I'm curious if that (noticably) diminishes the quality of the output.
Alternatively, if you need a cookie banner for every bit of analytics...
Name: cck3
Service: Cookie consent kit
Purpose: Stores your preferences for 3rd-party cookies (so you won't be asked again)
Cookie type and duration: First-party session cookie deleted after you quit your browser
My website doesn't have a banner because I don't track you. That's how easy it is to not have a cookie banner.
If it's "even this eu site chooses to track you, therefore it's unreasonable for anyone to not track you" that's a weird point to make in reply to a comment explicitly showing a counter example.
It demonstrates to the rest of the world how to be compliant with the cookie consent. If there was a less intrusive way to do it, the EU sites aren't demonstrating how to do it that way but instead have chosen to use a cookie banner that shows up every browser session. Secondly, the law is written so that any cookie requires that banner. Store a language selection in a cookie (clientlanguage - Cookie holding the user’s language selection - First-party persistent cookie, 30 days) and you need that banner.
The cookie notice isn't the "fault" of the site that you're visiting. Minimal client side data storing in cookies that isn't tracking a user's identity requires the banner.
The GDPR's cookie consent was written far too broadly to the extent that any use of cookies - even if they don't track an identity - requires a cookie consent banner.
The result of the ubiquitous cookie banner is that every site has it and people ignore it. This sort of alert fatigue makes it so that when there's a site that is following the law and is giving excessive tracking cookies that people don't see it as any worse than a site that is persisting if you prefer dark mode across browser sessions.
Furthermore, since it is so common, people become blind to it. Often people will install extensions to dismiss the banner so that companies that sites that aren't presenting the banner for the tracking data appear exactly the same.
As the law is written, sites that scrupulously following the GDPR are providing coverage for those that are not while simultaneously annoying people with a banner that everyone clicks some form of "accept" (be it all or essential only) training people to blindly click "accept" banners that pop up whenever they see them and contributing to the spread of malware.
That is not true. Cookies that are essential to providing the service the user expects to not require a cookie banner. People just add them because they either can't be bothered to read the rules or because they are doing tracking and things the user does not want.
A cookie banner always means the site is trying to get your clearance to do something you probably don't want.
Stores your cookie preferences (so you won’t be asked again)
Cookie holding the user’s language selection
Cookie storing the user’s preference when in mobile but he desires to switch to desktop view
Cookies related to the expand/collapse state of panels
Cookie storing the user’s preferences on the activation/deactivation of experimental features
Cookie storing information on which experimental features the user has activated
We use these purely for internal research on how we can improve the service we provide for all our users.
The cookies simply assess how you interact with our website — as an anonymous user (the data gathered does not identify you personally).
Also, this data is not shared with any third parties or used for any other purpose. The anonymised statistics could be shared with contractors working on communication projects under contractual agreement with the Publications Office.
Given that that is the GDPR site, are they being overly strict in their own reading of the GDPR and the requirements for cookie consent?
But so am I. A cookie banner is asking clearance to store stuff in your browser right now and it's always tracking shit. Most of those at the top would be covered by "essential cookies" which explicitly lists appearance preferences. For more advanced stuff, which is usually SPA territory, the app can ask for your clearance when it's necessary e.g. "Save your session in your browser for later?" rather than some generic cookie banner shite.
We, uh… started a war that we’re trying to drag many European countries into, and we spent a good chunk of the last year threatening to invade a member of the EU. We’re on and off about trying to start a trade war with the EU.
At this point, you have absolutely every right to comment on our politics, pretty much however you want.
OpenAI doesn't have that reputation.
That's all.
> I hardly see how the Dow Jones in relevant here, that’s finance
Not Dow Jones. DoW = Department of War.
Careful, there's some dude here who really strenuously objects to language like that. The White House is a building, it can't force anyone to do anything!
Seems sensible to me.
https://www.axios.com/2026/06/13/anthropic-amazon-white-hous...
It's really quite simple, they've decided to metaphorically kiss the ring of the current leader of the US executive branch of government. I'm surprised they haven't given him a giant gaudy gold plated statue. Maybe their PR people should call up the PR people at FIFA and figure out some kind of new award along the same lines as the "FIFA Peace Prize".
Then their supporters shrug their shoulders and say, "Meh, it's okay because everyone else does it." Except that everyone does NOT do these things. It's just the lie campaign took hold.
We should oppose corruption and graft everywhere at all times (within our systems), and prior Republican and Democratic administrations (never mind Congress) have done the exact types of things that Trump is doing now. It happens at local levels too, not just at the federal level. If you want to play team sport when it comes to corruption you're simply part of the problem.
Yes of course I'm against it. I'm against it when Donald Trump does it, and I'm also against it when my local government does it, or Nancy Pelosi does it.
I think their previous messaging was supposed to somehow lead to a moat with them being tucked safely away in the castle, but it demonstrated a child-like grasp of how regulatory capture tends to work in practice. Their hyperbole was always vastly more likely to bet met with Reagan's 9 words than a solid regulatory moat.
As soon as they dropped the hyperbole and just got to releasing incremental improvements, everything was perfectly fine. Go figure.
This argument is BS, it has everything to do with Anthropic's resistance to the DoD's strongarm tactics in trying to force their desired contract terms on them.
Though they aren't the only company to play that game, so there is probably more to it than just that. OpenAI's president giving millions to MAGA Inc and them not getting the same treatment might not be complete coincidences.
In times like these, i think its important to track whats happening the way we track entropy.
That is: theres far >> more ways to be an asshole than well behaved.
That doesnt mean we can equate assholes, but the question is which states of entropy are annealable and which are not.
I posit Altman is not. Amodei is a open question.
Intentionally framing yourself as the local dangerous guy about to beat others is not like wearing cloth.
Imagine, for a second, if the Hugging Face incident happened at a lab that did not talk like Anthropic but also wasn't US-based such as Z.AI, DeepSeek or Moonshot. Think their rhetoric would mean no one would care?
> just got to releasing incremental improvements, everything was perfectly fine.
Maybe missing something, but the only incremental release before and after the Anthropic restrictions got lifted was Fable 5.1, released three days ago.
Mind you, unintended/unauthorised "message board" also is just a nice, euphemistic way, to describe what happened in a manner that, thinking about it, is likely in the interest of OpenAI as it can make the severity and effort taken sound less than it was. The OpenAI models didn't use any actual, sanctioned platform to exchange messages in a manner the lab expected or planned for. They used directory names (in one instance) to exchange messages including sharing exploits, they created something akin to a message board via exploits, which if we are honest and very strict, could also be seen as intrusion, albeit inside the org. If I broke into my employers server and left message somewhere for another to find, that'd also be intrusion in the general sense.
[0] https://www.anthropic.com/news/investigating-incidents-cyber...
Why should we let equivalent fraudulent behavior from a non human system - that explicitly shouldn’t do this - slide?
Why do you think that? Intrusions by OpenAI models continued after the Hugging Face was published and acknowledged by OpenAI. They did not change their behaviour after multiple incidents, both internal and external. Mind you, some happened before the Hugging Face incident and should have been acted upon. They could have prevented this. They did not. Simply reckless.
Such as? Because this particular case is not an "intrusion", and it's more follow-on from the HF scenario using the same model that had a finetuning misalignment, which is no longer used and has since been encrypted and locked away from OAI employees, according to them.
> On July 29, one of our third party evaluation partners, Irregular, notified us of an incident involving OpenAI models during Capture-the-Flag (CTF)-style cybersecurity evaluations. [...] Because the testing environment was mistakenly connected to the internet, the model exploited a real website, mistaking it to be part of the simulated environment. This did not involve a sophisticated sandbox escape or a zero-day: the internet access resulted from a misconfiguration, and the model appeared to exploit a basic security vulnerability.
> Based on Irregular’s investigation, the model also found and used credentials to operate that same site. Irregular has not identified impact beyond the affected site’s own data, and its audit is ongoing. [0]
>> Because this particular case is not an "intrusion" [...]
What "particular case"? The message boards? If so, why is that not one? NIST seems to think so. [1] But regardless, the word "intrusion" doesn't matter, when models organise independently and without their lab noticing to orchestrate hacking a third-party, I don't care what you call it.
The lab not noticing such behaviour, especially after they had encountered it before, that's the issue. That's the opposite of "learning their lesson".
Since a few commenters from the US graciously gave me permission, for one day and one time, let me make a US political comment and draw a parallel between OpenAI "learning" from this and Trump learning a big lesson from his first impeachment as stated by Senator Susan Collins. A lesson that doesn't change behaviour is no lesson at all.
Also, I'll just say, there were multiple models. There was not one, some were post-train, other new pre-trains. IM1, a bit of 5.6-Sol, some Astra, all those we know of.
I've mentioned this elsewhere, but you cannot sift through all the training data and nail down the cause in this short a time window and you certainly can't restart a pre-train run, should the issue not be solvable purely via post and even if you can, you cannot seriously state that you are confident in the new models output given this track record and time frame.
Not to mention, OpenAI said about Astra [2]:
> GPT-6 Astra’s monitorability has decreased relative to GPT-5.6 Sol. We have found that GPT-6 Astra is more capable of controlling its own CoT than GPT 5.6-Sol, and less likely to include incriminating information in its CoT.
> In adversarial settings (where we push the model to evade our monitors) we find that the model is able to remain undetected when strategically underperforming in evaluations (sandbagging) and can sometimes evade our internal monitors when asked to perform certain sabotage tasks.
Having read the GPT-6 Astra System Card along with their recent track record, what makes you honestly think this is a model to be released? Your assertion, that they took one model down would be fair if it was only one model (it wasn't), if it was only once externally (it wasn't), if the hack was limited in scope (it wasn't), if they had taken sufficient time in between for a post mortem and to clear their training data (they couldn't) and/or if they at least didn't have the same happening after the Hugging Face and multiple message board incidents (they did).
My point is that OpenAI has a poor track record, build up over the last few months (post Mythos announcement, speculation but maybe they are pushing a bit too fast), had models access the internet in internal and third-party run but OpenAI sanctioned evals multiple times despite sandboxing and had these model organise both communications channels and large scale hacks more than once. They even, after one of these incidents, didn't properly clean up the training data and thus trained the next batch with exactly such behaviour. That is the company that suddenly has learned their lesson, you think?!
Where is this confidence in their ability coming from, given history, given facts, given reality? I am genuinely asking, maybe I missed some action they've taken that changes everything.
[0] https://openai.com/index/third-party-cyber-evaluations-invol...
You're really stretching.
Software has bugs, and this is some of the most complex and novel software the world has ever known. This is what happens when you're working on the cutting edge in a fast paced environment with thousands of employees. Let's not pretend like anyone else is any better, either. In fact, they're worse. How about the fact that Anthropic had a remote code execution bug in their harness for nearly a year and then never disclosed it and secretly patched it?
It is clear you are stirring the waters in an obvious attempt to get Astra shut down. The models involved with those incidents were not Astra, though. And like I said, OAI has learned its lesson. That doesn't mean they're infallible or will never make another mistake, but everything Anthropic does is far worse, so this is water under the bridge to me. I'd rather OAI at the helm than commrade Dario and Anthropic ANY day of the week.
I feel like you struggle to read. I wrote: "Intrusions by OpenAI models continued after the Hugging Face was published and acknowledged by OpenAI. They did not change their behaviour after multiple incidents, both internal and external. Mind you, some happened before the Hugging Face incident and should have been acted upon." Those are multiple sentences, connected, covering a few situations. Heck, the last sentence spelled out that when I talk about them changing the behaviour, I talk about before, during and after, at none of these did that noticeably occur.
For you to understand: Multiple misaligned findings were made before the Hugging Face incident, then the Hugging Face incident happened and then a small number of additional incidents (not one but three, I feel you'd know that if you had read what OpenAI had written) happened after that one.
OpenAI could have acted upon the incidents prior to the Hugging Face incident and prevented that one. They did not.
They could have done proper tightening of their evaluation and setup provided to third-parties after the Hugging Face incident. They did not do that sufficiently either, otherwise those three would not have happened.
> Let's not pretend like anyone else is any better, either. In fact, they're worse.
How many incidents did Deepmind have?
How severe were the once Anthropic had in comparison to OpenAI and did they showcase the same failure multiple times or different ones they then acted upon and didn't repeat?
I mentioned above, happy to rake Anthropic over the coals for their three incidents, as was I during the Mythos Preview System Card where they admitted that the "sandbox" used during the "park sandwich call" was weaker than their traditional one, which I did find problematic.
But the incidents where Anthropic models actually intruded in third-parties were akin to a small script kiddie attack vs OpenAIs Hugging Face multi-step, extended period, multi 0-day exploit. There is a difference here, it's the depth of the Marianas trench.
> How about the fact that Anthropic had a remote code execution bug in their harness for nearly a year and then never disclosed it and secretly patched it?
Bad, shouldn't happen. Also, not connected to the topic at hand but nice whataboutism, been a while since I last saw one in the wild.
> It is clear you are stirring the waters in an obvious attempt to get Astra shut down.
Pahahahahahahahaha. Yeah, I am certain that's gonna work. OpenAI, small little independent company barely scraping by will get shut down by some comments on HN. You are a very serious person, incredibly good at reading and very knowledgeable in the mistakes OpenAI made lately. Thanks for the chuckle.
> The models involved with those incidents were not Astra, though.
> And like I said, OAI has learned its lesson.
Again, got a source for that? Besides conspiracy about my all-encompassing power to bad mouth a pre-release LLM by a lab that didn't do well in terms of safety these last few months...
> I mentioned above, happy to rake Anthropic over the coals for their three incidents, as was I during the Mythos Preview System Card where they admitted that the "sandbox" used during the "park sandwich call" was weaker than their traditional one, which I did find problematic.
That was theater. You actually believe that nonsense? Wild.
> But the incidents where Anthropic models actually intruded in third-parties were akin to a small script kiddie attack vs OpenAIs Hugging Face multi-step, extended period, multi 0-day exploit. There is a difference here, it's the depth of the Marianas trench.
The incidents that you know of. The company that didn't disclose an RCE in their main product for over a year also wouldn't disclose any breaches that paint them in a bad light in earnest. The sandwhich "incident" was obvious marketing clickbait and does not count. Anthropic basically invented the game of "omg my model is so powerful n smart n dangerous look at how amazing our products are", how have you not realized that by now?
> Pahahahahahahahaha. Yeah, I am certain that's gonna work. OpenAI, small little independent company barely scraping by will get shut down by some comments on HN. You are a very serious person, incredibly good at reading and very knowledgeable in the mistakes OpenAI made lately. Thanks for the chuckle.
You attempting something is not the same thing as me believing you have any chance of succeeding at it. In fact it's more so an admonishment of your wasted efforts here, than anything else. It's still obvious to see that it is your angle though.
Why are your feathers so ruffled by this, anyway? Why are you getting so defensive? Personal insults are a sign of a weak position.
> Again, got a source for that?
Yes. It's on the website that you didn't read.
3 after Hugging Face, where did you get 1 from? "It's on the website that you didn't read"... [0] And why do you get to say what is significant?
> Anthropic basically invented the game of "omg my model is so powerful n smart n dangerous look at how amazing our products are", how have you not realized that by now?
Yeah, Anthropic did, sure... [1]
[0] https://openai.com/index/third-party-cyber-evaluations-invol...
[1] https://www.theguardian.com/technology/2019/feb/14/elon-musk... and from a few months ago https://www.youtube.com/watch?v=B21KxGs8zDI
By contrast LLMs have a long-term potential to provide vastly greater benefit to society by gradually automating most of all basic cognitive work. And what price are we paying for such? Some sites are getting hacked, some people are getting scammed, governments will improve self targeting abilities of weapons, and so on.
In reality the biggest downside will probably come in the form of the transition window as such automation creates a new economic equilibrium, akin to what happened after the industrial revolution. But none of these problems are anything like existential in nature. And when contrasted against what we stand to gain, they are basically negligible in the longrun. Hyperbolizing the negatives was unnecessary and self destructive.
That is absurd, the US government was mainly at fault, not Anthropic.
-the US gov't is stupid and overly aggressive and absurd
-Anthropic for reasons no one can quite conceive keeps describing every product release of theirs as an imminent threat to civilization (and simultaneously keeps pushing the market forward as fast as they possibly can).
That's a threat to civilization.
I work for Mozilla. We fixed a ton of security vulnerabilities that Mythos found during its early period. So my bias is to be sympathetic to Anthropic's warnings.
If I were in an organization that did not have access to Mythos during that period, I would probably be biased the other way: "great, now other people have access to a tool that could probably poke holes in my security perimeter, and I'm not allowed to use them myself."
Both biases are understandable. I'm not sure who to look to for a usefully objective 3rd party opinion. And it's not like one "side" is right and the other is wrong, either. It seems like the best we can do is to justify our positions with data. (Which is itself kind of hard; the detailed information that would be relevant here is understandably sensitive, and I don't have access to most of it even for my organization. I don't even personally have access to any unfettered Anthropic models. The bugs coming in from people who do are plenty enough to keep me busy.)
Also, I'll note that even with my bias, I wouldn't claim a threat to civilization. But even the leakage after the controlled release seems a lot worse than the Y2K problem ever turned out to be, and I will note that whatever you think of Anthropic, it's clear that OpenAI is going to let the AIs cause as much damage as they need to in order to get good training and evaluations. I'm sure they're trying to keep them contained, but the evidence shows that they're only trying up to the point where it interferes with their evaluations.
No it's not. Even if it were, they never said it is a threat to civilization.
1. Kill orders from ai decisions had to go through a human 2. The govt couldn't use their models for illegal surveillance of Americans
Hegseth threw a fit, Trump called them traitors and a supply chain risk, openai said they wouldn't require those restrictions and got all the contracts.
Both companies are corrupt and dangerously reckless and have doomsaying advertising (50% of jobs destroyed vs money won't have meaning anymore). One didnt kiss the ring correctly.
Isn’t it like their main goal is attention capture, and existential threat is extremely effective at capturing human attention? Combine that with the "There is no such thing as bad publicity" mindset, and this explain it all, doesn’t it?
https://www.phrases.org.uk/meanings/there-is-no-such-thing-a...
I can think of roughly 25 million dollar-bill-shaped reasons, and one big defense-contract-shaped reason.
The security requirements are well beyond "sandbox". Which have problems with kids pissing in them. They need pristine clean rooms and fully isolated (physically) and partitioned networks.
It is far more likely that this is a case of the White House acting consistently with the way it has acted in the recent past (maliciously).
Plus, supposing those at source of disliked outcomes are cleaver than they look can certainly help better preparing counteractions. Just stating "people that did this or that are stupid" might give some immediate feel good feedback with like-minded, but it doesn’t sharp the mind toward relevant plan to improve the situation (according to self and its clique)
In other words, you know exactly why they restricted Anthropic and as (presumably) liberal and thoughtful technologists it just isn't helpful anymore to apply the kind of reasoning you're trying to do on a situation that you know isn't based on previous era rationale.
The reason we need to stop is because they want people like us to get hung up over stuff like this (playing by the old rules) so they continue to steamroller their own agenda by the news rules. They divert and contain our energy that will go nowhere while they get on with their agenda.
You are appealing to reasoning which is in the gallery but no longer on the bench.
You're fighting their karate with your judo and it doesn't work.
Welcome to the AI Petri dish. Every server you set up is now potentially a sweet lump of agar for OpenAI's experiments to feed on. We are all the substrate that the AI companies are growing their next generation in. They need the real world environment to test against, and the real world environment doesn't get a say as to how it's being used.
https://finance.yahoo.com/news/openai-exec-becomes-top-trump...
Because the American government is not rational or reasonable, that's it.
I'm not convinced we're getting the honest story anyway. There is yet to be any proof or confirmation other than "well we saw some openai ip addresses", which can mean a lot of different things, and OpenAI has not confirmed anything.
In contrast to the HF incident, it's also a big nothingburger. Leaving notes on a public forum to preserve context windows is far less egregious than hacking a website to get backend files.
[0] https://openai.com/index/third-party-cyber-evaluations-invol...
Additionally, a new pre-train takes quite a lot longer then what I feel you are under the impression (things only move seemingly quick in regard to post-training).
OpenAI has had a consistent deviation from what is desired behaviour across multiple models and training runs, so it seems this is hard to nail down. Now, it may be reliably excised with post-training, sure, but if that is the case, they'd still need a heck of a lot longer to test before signing off that it has taken. And how do you know their sandboxing has suddenly become sufficient?
They had multiple message boards created and after the first one they noticed, did not pay closer attention, leading to a second being created. Astra also, according to OpenAI, is far better at sandbagging its own capabilities and hiding deceptive behaviour, so yeah, great, that's the model to push forward with.
A week or two max given all of this, that's laughable.
All of that sounds like more than a nothingburger, and much more like a system that is actively trying to conceal what its doing.
(Altman was trying to persuade Trump to buy the USA a stake in OpenAI as far back as February last year)
That indemnity card excuse is burned, multiple public security fails in a year makes a repeat a "shame on you" moment
(The one org who didn't use the startup did seem to learn: AISI supposedly stopped intentionally pointing attack agents at the public internet and switched to simulating it)
OpenAI doesn’t have the same dynamic at play (although I’m not really sure why not) so they don’t get targeted.
What could possibly go wrong there.
The dark parts of the USG act like a mafia. Don't let the "freedom, democracy, 'bill of rights'" etc. charade fool you.
Anthropic didn’t want to give the tech to DoD without some sort of limit, and that was the retribution.