Statement on US government directive to suspend access to Fable 5 and Mythos 5

First, Anthropic was founded by people who we know were worried about AI safety and signs point to that still being the case. It's really cynical to say it was all an exaggeration for marketing.

Second, this isn't Moller promising a fantastic working flying car next year. The model did what Anthropic said it could do.

I realise that ruling out "they bought Anthropic's scaremongering" brings up the question of why the government would block Mythos/Fable, but not the roughly-as-capable and less restricted GPT5.5. However we do know for a fact that they dislike Anthropic more than OpenAI right now.

Especially if those people aren't presently very bright, and are already mad at you for not helping them achieve their unrelated authoritarian goals.

I do not think this is somehow a 3D chess move by Anthropic. They are not masterminds, even if they'd really like to be. People who actually interact with their products know that Fable and Mythos are incremental improvements, not doomsday devices. I think this is a punitive move by an administration that loves being punitive, which they have unknowingly bolstered with their own dumb rhetoric.

Export control is not an effective tool for controlling a consumer facing technology developers everywhere want to use (see:VPNs) so there was no good faith policy justification for imposing an export control.

This is an administration that seems to be keeping track of who its friends are and aren't, and likes to be the center of every story. They also seem to like extracting concessions and reciprocal favors. We saw some of this behavior in the last administration too. US voters deserve better.

And potentially more importantly: if a model like Mythos, which at best is an incremental improvement over Opus, is getting this treatment, how are all the AI investments that are based on the expectation of ASI / AGI / significantly better models going to be recouped?

It might be a national security problem for other nations to have access to these models. But it's equally now a national security problem for any other nation to depend on them. Or US tech in general.

The real story here is that this may be the beginning of governments restricting the availability of strong LLMs to the public, to you. Fable was the strongest model on the market, and the US government has told you you can't use it (technically, only if you're not a US citizen, but in practice, even if you are). If you think the solution here is going to be open source Chinese models and / or running on your own hardware, think again. Do you think China is going to allow the strongest LLMs from companies within its borders to be open source a year from now when they have Mythos capabilities, if the US government is keeping the strongest American models back? Unlikely. These are heading in the direction of being powerful cybersecurity weapons and it will be in the interest of nation states to restrict and control them. In 2 years time, I would be surprised if the strongest LLMs are available for general use at all.

Will we be the poorer for that, or will we be safer? I think poorer, because I hate being told what technology I can and can't use, but I'm not certain. Maybe you think the government should restrict strong LLMs. Maybe you don't. But either way, this is big news and a rubicon has been crossed and a precedent set. That's true even if the motivation for this is just the government settling scores with Anthropic.

No one's going to risk building anything important on these models if the government will randomly order the use of the model to be discontinued by all foreigners, regardless of if they are in the US or not. Just a matter of a foreign company catching up to take the commercial market for such models (though, as the US often does, they'll ban the competitor, so actually we'll have a situation where the backend uses a different model in only the US).

The modern world is a wild place!

This press release is odd - it says that the export control was imposed to stop foreign nationals from using Fable / Mythos, and then goes on to talk about supposed concerns about jailbreaking the model.

But is that really the concern of the US Administration? This looks more to me like they are viewing frontier models as a strategic asset which they want to keep for US-exclusive use. I can see the logic - if frontier models generally accelerate a society's technological development, then a country looking to retain or increase its strategic edge over other countries would try and keep this sort of multiplier for themselves.

I'm guessing Anthropic shut of access for everyone because currently they have no reliable way to know whether a user is or is not a US citizen. In the near future we might be in a situation where you need to prove your US citizenship before Anthropic / Open AI will allow you to use their current frontier model.

I can't help but wonder if it's now obvious that frontier AI work should not happen in the US.

I can understand the KYC aspect of this, but at the same time, how can anyone trust US based AI after this? Maybe this is a continuation of the Pentagon feud, or it's revenge, or it's a KYC play. Either way, you've got a government willing to shut down companies sales over arbitrary reasons.

Ironically, I mostly have a subscription to Claude for work, which is primarily for US baed companies.

The rubicon being crossed here is Republicans/the red tribe losing their comparative advantage of being opposed to overregulating a rapidly advancing technology.

beautiful good bye, for now

Now they finally found the right fools in audience to believe it.

Now, we obviously know that without some kind of brand new ID check, such a thing would be impossible and thus they had to just shut it down. But this touches on the same kind of issue as all the noise about “for the children” ID checking. We might be soon to see the set of “things you’ll have to reveal your identity to the government to get,” expand from “just” porn and social media to the “good” AI models.

This ban will be used to force hardware and OS-level Digital ID down our throats as a "safety measure" to ensure people are "Citizens" before accessing AI technology.

Whatever last vestiges of privacy we still enjoy will be taken from us with this as the excuse.

Now they are betting that Mythos will provide them with some edge. Personally, I don't believe that Mythos is such a game changer They're just buying their own hype.

A late stage empire flailing around sacrificing everything to maintain its status.

https://turbulence.substack.com/p/the-gated-age

Anthropic went from this is cybersecurity apocalypse to it’s no big deal, the model found trivial vulnerabilities.

See also https://www.404media.co/fcc-wants-to-kill-burner-phones-by-f...

Digital yellow star by exclusion from digital life for foreigners.

Remember when tech companies would go to court to vigorously defend against infringement of their and their customers rights? Turns out that’s just a feature of democracy, once you have autocrats it’s all compliance.

Anthropic just baited themselves with their scaremongering to be the attack vector here.

It a stellar move by the way - since every tech company in an exceptionally fast growing field will comply or miss out sales, you effectively force KYC without legislative process onto much of digital because that’s the only way to comply.

Anthropic pretending Mythos 5 is so capable it's going to destroy everything, but will release it anyway with "safeguards" (when does this ever work?).

US Gov't using this fake hype as an excuse to handicap Anthropic simply because they have a vendetta.

So much for all of the rhetoric about Mythos supposedly far surpassing GPT 5.5 (edit: in cybersecurity, in particular). Of course, the AISI benchmarks also showed this, but it is amusing that Anthropic is saying it now that it is to their advantage.

Red Bull marketing revolved around making it look like a drug. It "gives you wings", there is that crazy thing called taurine, they even "hooked" kids with free cans, the mythical thing drugs dealers do.

In reality, taurine is nothing special, it is high in caffeine but no more than a strong coffee, and its real energy come from the massive amount of sugar it contains. Marketing aside, that makes it an unremarkable soft drink.

But their marketing prompted some countries to ban it, at least for a time. France is one of them. Fun fact, when they finally legalized it, they introduced a heavy tax on energy drinks, defined as soft drinks high in caffeine. These drinks are expensive and the government wanted its share. In response Red Bull silently reduced the caffeine content to avoid paying, marking Red Bull even milder than it once was.

It's difficult to predict this administrations actions, but given it included employees that has to be a huge risk for Google, where Deepmind is based in London.

Cohere (Canada) and Mistral (France) are going to get a lot of interest.

I wonder if they pulled Fable because it had too high of a “dangerous session” count. If so, I wonder if they’ve considered that their “dangerous session” detector has lost its damn mind this week.

(BTW, that screenshot is 100% real. I was walking to work this morning and a random song played. I had a thought about it and wondered what a model would have to say on the matter. I ran that prompt and got that response, said something profane out loud, and screenshotted it to share with friends. That’s not a mockup, but something I personally experienced and recorded myself.)

It's been interesting seeing how OpenAI pops up to counter the threat of AGI being controlled by Google, and then OpenAI and every spinoff company from its employees has become a far larger threat to the public, for different reasons.

As much as it seems like Anthropic's self righteous leadership truly believes in what they're preaching, they've shown themselves to be tied for the worst stewards of this technology. Google actually seems like the best option to me, by far. Anthropic is also the only major lab with no open weights releases.

They'll have burned a lot of goodwill with the community by the time another lab takes the tech lead, which I guarantee will happen.

https://www-cdn.anthropic.com/d00db56fa754a1b115b6dd7cb2e3c3...

That said, Mythos doesn't seem to be exceptionally good but closer to "following the established trend in improvements"

https://www.aisi.gov.uk/blog/our-evaluation-of-claude-mythos...

https://www.aisi.gov.uk/blog/how-fast-is-autonomous-ai-cyber...

“OK, Dario. Let’s start with you.”

“No! I meant regulations for other people!”

After that details don't matter, they've shown their "enemy" colours, once is enough. This is just punishment and it will continue, until they bend the knee.

“Thats a pretty nice IPO you got there… it would be a shame if something happened to it.”

Yann LeCun got that right with AMI Labs.

UPDATE: I lost access at 6:59pm pacific.

If Fable 5/Mythos 5 are considered dangerous enough to invoke export controls on then future models are almost guaranteed to trigger the same process. Locking them down to US citizens is _very_ interesting. I don't think any tech company so far tracks licenses attached to citizenship.

"small number", "previously known"+"minor"... they are trying hard to characterize this as harmless.

> These vulnerabilities all appear relatively simple, and we have found that other publicly-available models are able to discover them as well without requiring a bypass.

Ah so now they are admitting that this is all about hype after all.

I'm glad I don't own stock in a public Anthropic.

I guess they will fix the guardrails and then open it up again. Clearly nobody wants dangerous models out there and I can understand the national security concerns. If the restrictions persist even if guardrails are updated, well, perhaps other countries may want to compete for becoming the new home for frontier labs?

> We reviewed a demonstration of this specific technique being used to identify a small number of previously known, minor vulnerabilities. These vulnerabilities all appear relatively simple, and we have found that other publicly-available models are able to discover them as well without requiring a bypass

Not great as it does break workflows for some.

> As we have stated publicly, we believe the government should have the ability to block unsafe deployments, as part of a statutory process that is transparent, fair, clear, and grounded in technical facts. This action does not adhere to those principles.

Whatever you feel about Anthropic, good or bad, this is not fair, and this is not good for the industry.

How will this be implemented/verified? Also, does this mean that American citizens abroad will still be able to access it?

Now I can continue my vanity project of having Claude iterate on a single spec.md for hours on end. Surely at some point it won't be shit.

> Restricts model to large corporations

> Release information about how Fable / Mythos 5 is stronger than Mythos Preview, give access to every user for a limited time via subscriptions

> Users jailbreak model

> U.S. suspends Fable / Mythos use

Who didn't see this coming?

I wonder what this means for the future of AI models. Either we'll see worse guardrails than what was there for Fable 5 (for me, it was a unusable at times), or the models just stop getting better from here.

I think it's that the guardrails will be more strict, which is unfortunately not good news.

Will be interesting to see OpenAI's next move.

This sounds exactly like the opening line from an apocalyptic sci-fi film.

The page showed June 11, 2026 and has now been updated to June 12, 2026 in the last 10m.

https://imgur.com/a/lx7HCW9

Edit:

Google mislabels crawl dates clearly my bad

The most ethical goal of an AI lab or government should be to bring the maximum amount of intelligence for as cheap as possible to the people equally.

Another interpretation, of course, is that this is just US putting a thumb on the scale for US competitors around IPO time. It will be interesting to see if there are any fingerprints.

Maybe they'll have access to CCP models, but China will likely soon do them same. Maybe they will allow access but you must use it on their servers (i.e. share everything you do with the CCP).

Perhaps Mistral can pull something out, but how far ahead will the US and China be by then?

But is that really the concern of the US Administration? This looks more to me like they are viewing frontier models as a strategic asset which they want to keep for US-exclusive use. I can see the logic - if frontier models generally accelerate a society's technological development, then a country looking to retain or increase its strategic edge over other countries would try and keep this sort of multiplier for themselves.

I'm guessing Anthropic shut of access for everyone because currently they have no reliable way to know whether a user is or is not a US citizen. In the near future we might be in a situation where you need to prove your US citizenship before Anthropic / Open AI will allow you to use their current frontier model.

The next interesting question will be - will the US share this capability with her traditional strategic allies (e.g. five-eyes countries), or is it truly America First (or, 'America Alone')?

It's vitally important open source models are supported.

It is hard to count how many “red lines” were crossed last night. Government shutting down AI model it doesn’t like? Government shoots into barely profitable 1T startup, whose entire trajectory and, in essence, survival depends on commercial success of that model? Access to AI model being governed by citizenship, likely verified through rigor ID checks?

Very little doubt that China will follow this suit. Next US will pass their bills to ban Chinese open source and local models. Chinese will follow.

It doesn’t matter: US did just a vendetta; felt into “marketing hype” or there are legit security concerns of national security level. In the precedent-ish world of post-truth “whataboutism” we just crossed a big milestone mark which will hunt us for years.

That’s seems like an attempt at a broad precedent setting power grab for the administration to assert power over tech companies it doesn’t like.

That seems like a fairly existential threat to tech companies ability to do business.

1 - https://www.federalregister.gov/documents/2025/01/15/2025-00...

At some point, as AI becomes more powerful (Anthropic themselves seem to think we're already there), then it should really be necessary to have US government clearance to work on the models, just as it is for defense work.

Apple to EU: “nah, we need to be able to provide only the best”

US government: * starts pulling the plug on AIs outside the US *

You can kind of see how the EU has a point

https://youtu.be/l2ThMmgQdpE

There's no way they have the authority to actually order this and not just request this right? If crypto is speech... LLMs definitely are...

It may be safer to just move the company to Canada.

The reason I think Anthropic is doing this—preventing distillation and making distillation way harder for China—is that if you geoblock this and have strict rules around proxies, it works. I can't imagine Dario Amodei not knowing the consequences of his own policies.

As for open-source models, have you ever tried a distilled Claude Sonnet model (lordx64/Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled-IQ4_XS-GGUF)? I did via Hugging Face, and they are quite shit; they are very slow and eat up your VRAM. Open-source models have a long way to go to catch up to closed models.

So everyone saying we should use open-source models—that will never happen. For example, macOS is closed-source and Linux is open-source. Which one has a greater market share? macOS does, because of its privacy and security. "Let's wait and see how this works out. This sucks for enterprises and customers outside of the USA, but with a policy like 'America First,' you can expect this from any president in the White House.

Additional theory: Altman is behind it.

Is there no one in government who would stand to gain from a financially handicapped Anthropic in the context of an OpenAI IPO?

So whatever, I just don't really feel the need to burn tokens on Fable anyway.

Those who bribe Trump and do exactly his bidding (including helping out with war crimes and surveillance of US citizens) will be left alone, or even protected from competition and international law, as long as they keep giving Trump a taste. Those who balk, even a little, will be punished for it.

Republicans never wanted a free market, they just wanted a market that served their interests.

Russia and China could not dream of accomplishing the damage being done to US leadership in tech by our own government as we speak. If they have a wishlist, I'm sure it includes things like stopping immigration of scientists to the US, punishing innovators and elevating hucksters (make them trillionaires, for example), drive a wedge between the US and European allies, insure no one trusts hosting their data in the US or with US companies, erode democracy, and increase inequality especially at the margins (make the poor desperate and the wealthy beyond the reach of consequences).

Hmm hmmm https://www.bbc.com/news/stories-54695598

They told us this model is dangerous, and now they are complaining that someone with more guns than them said releasing something dangerous is not okay?

[0] https://europe2031.ai

It’s a massive betrayal for foreign entities and it would be silly to continue with all my eggs in anthropic basket but I get it.

> As we have stated publicly, we believe the government should have the ability to block unsafe deployments, as part of a statutory process that is transparent, fair, clear, and grounded in technical facts. This action does not adhere to those principles.

So you spend billions of dollars training the model, only for it to be used in the US.

Then interesting to see where most of anthropic revenue comes from. If it's the US then they're fine but if it's global then they'll see a drop in revenue?

Then add to this decision, companies are going to significantly reduce their token spend.

So what does all of this mean for their IPO?

Just in case the whole threatening to invade Allie’s didn’t quite get the message across

Sounds like they only want Americans to access SOTA AI.

I’ve used Fable a lot. It’s a marginal improvement on Opus. It’s really not scary smart. I don’t want to go back to Opus because Fable was that little bit smarter, but it’s not like anything really changes at work and when we’re bumped back. So I’m really not buying any national security angle other than in the sense the administration can weaponise that to crown the AI race winners.

https://xunroll.com/thread/2064776322979676227

Using combinations of jailbreaking-techniques including: writing cyrillic helped a lot to disarm the filter.

I asked it to add one thing to a function in the static blog pelican.

So, while it worked, it took no account of what was already in the function and made a bunch more stuff.

I'm talking about something that I'm the end is a 3 to 5 line patch.

The default is still tech debt, but now we burn way more energy.

But a model that can provide general information, research, or source code for most modern technology?

It is really unusual that this is the first notice of this

First, it comes out that on some specified subset of queries they will simply downgrade you to a different model. One example is if you are doing model research. Imagine OSX shutting down if it detects you’re working on software.

And now they’ve decided that they’ll just shut off access to the model completely as part of what seems to be a sort of marketing stunt or temper tantrum.

They’re a service provider. Can you imagine AWS just deciding you’re getting nullrouted over some unrelated fight they’re having with the DoD?

If they weren’t a supply chain risk before this, they’re sort of doing everything they can to become one.

I think the investor cash needs to dry up. They’re not going to let doomsday technology be released to the market. Sorry.

The main difference here is that cryptography didn't require significant compute hardware, which is the perfect place to also apply export controls (and they have).

We could smuggle PGP source code on paper / DeCSS source code on t-shirts. That ain't gonna happen with the hardware needed for frontier models.

Andrej Kaparethy can't do his job.

This is going to have a huge effect.

Chinese open weight models, which were in a 'reluctant' category aka 'maybe there's some CCP propaganda embedded' just got a whole new life.

Arbitrary and instant cut off from key technology is going reshape a lot of things.

Well over 1/2 of US growth is now AI and well over 1/2 that revenue is outside of US.

Space X IPO in addition to OpenAI and Antrhopic IPO's ... would be put in gigantic risk in any rational market situation.

Is open source next on the list? Better grab the latest open source models now and get your Blackwell 6000, Spark, or Mac mini fired up and ready to go. I think you're going to need it.

This is rarely discussed, and while I agree we should be spending non-zero effort on safety, stopping progress is not an option.

Unless folks are hearing that they did this I smell marketing and/or PR as the main driver of the action.

1. Release fable, highly nerfed and limited 2. See the compute capacity limiter pegged day after day 3. Lobby to the government, claiming ai is super unsafe and not aligned and they must do something 4. Government "forces" anth to turn off 5. Anth takes the pressure off of compute capacity, and gets to blame it on the govt

Like you're telling me fable is somehow an order of magnitude better than GPT 5.5 to the point where it compromises national security, despite evals and anecdotes saying otherwise? Nah.

... Isn't that basically what Anthropic asked for, literally a week ago?

https://www.anthropic.com/institute/recursive-self-improveme...

> We believe it would be good for the world to have the option to slow or temporarily pause frontier AI development to enable societal structures and alignment research to keep up with the advance of the technology.

Logical next move — “we’re requiring ID.gov identity verification to use our services”

It’s all about spying on users, Don’t let anybody kid you into thinking it’s anything else

If you knew what you were talking about 4.6,8 could already do mythos level hunting and tool building.

So I guess the real moat is whether the US govt is happy to make your models sound more capable than they are?

Many nations are now likely thinking: Why cooperate on international IP enforcement if we get lumped together with adversarial nations anyway?

Our response to the US ban on Fable 5 and Mythos 5 - https://news.ycombinator.com/item?id=48512915 - June 2026 (17 comments)

https://ai-2027.com/

This is not a technology that they have a 10 year lead on. It is maybe 18 months until you can get Mythos from multiple places. And the US administration has no power to block them all.

https://web.archive.org/web/20260613005302/https://www.anthr...

All those $Billions of investments in AI Datacenters? Up in smoke if the models that are capable of replacing humans can't actually be used.

I wonder if 2008 style bailouts will be needed, soon.

That trillion-$SpaceX valuation based on $14B+$10B infusion from Anthropic and Google? Heard they have short-notice cancellation clauses.

Either this rule is rescinded quickly or the bubble bursts. Which shall it be? I know which one I'm betting on; do you?

Tell me, Mr. Anderson, what good is an ai-model, if it can't speak?

I use Opus everyday on my code. It finds security gaps, but nothing outrageous, and that's coming from someone who writes code that is nowhere near weapons grade secure. It finds mostly things that are technically bugs, but virtually intractable for exploits. Often things I overlooked cause I was lazy, like not synchronizing access to a shared pointer etc.

It also is a good PR for them because it continues to doomer hype loop that’s boosting them.

Anthropic: your next ad writes itself. Nobody else is worth restricting.

There we go. This should make nations consider whether they're letting their workforces become dependent on foreign tools, but of course they won't.

Does the White House want the AI bubble to pop..? Incredibly dumb move.

Found the time traveller.

1/ Jailbreak => Rapid catchup of the industry leading to commoditization

2/ Jailbreak => 99% of internet infrastructure gets exposed to cyberattacks at a scale the world is simply not ready. Maybe <1% of internet users are using Fable, out of which <1% will use it for beyond intended use. Put yourself in the shoes of someone maintaining critical infrastructure, or millions of people working 7 days a week to run a small business. The world needs some time to adapt.

So tired of the nonsense.

I’m looking for that news article in the next weeks…

If Anthropic keeps Fable closed source and plays the Authoritarian game, at least we have hope that an upcoming Deepseek model will surpass Fable before long.

Kinda ironic

Edit: Fable 5 was just disabled.

(2) I am really tired of the AI community trying to threaten everyone with grey goo and finding out the hype doesn't land comfy with others. It's a freaking text generator, not god in a pocket.

As I only signed up to check out the fable, I just did this.

"Refund of €96.84 processed. Expect it in 3–10 business days."

Let's see how long it takes. Funny that it never takes this long to be deducted from my card.

(Ok gotta spend my upvote points somewhere)

“Wait what do you mean you’re banning it?”

They had better give me a refund!

The Trump administration should focus on things like the UFC fight etc.

This also looks like the perfect China shaped gap in the market if there ever was one.

On the one hand someone will subscribe $4.99 a month for TODO.app or calendar.com because they are paying for a solo dev or a small team to work on constant development and improvement of products filling a particular niche.

On the other hand, Linux, Django, PyTorch, React, Zed, Helix, Postgres, Arch, Chromium, Firefox, Rust, Python etc. ship continually improving, solid pieces of enormous infrastructure for free, to be used freely by all, off the back of hundreds if not thousands of active core developers. These projects and large and complicated. They are also commodities.

Then, ahem*, on the final hand there are of course Windows, Office, Adobe, macOS and iOS, et al which span both categories: monster projects that are also commercial and also commodities and yet they have hooked themselves into the world in such a way that most folks gotta pay for ‘em.

LLMs feel like they want to be in the same category as the OSs of yesteryear, with all the fanfare of major release versions named like 95, 98, 2000, XP… or like Leopard, Tiger, Yosemite, Sequoia. The training and evaluation pipelines might feel like they fall into those categories, but the models themselves — after all, distillations of someone else’s public or private IP — do not.

”In 1991, the United States Supreme Court in Feist Publications, Inc. v. Rural Telephone Service Co ended a seventy year struggle among federal circuits concerning copyright protection of factual compilations. Prior to this decision, courts allowed copyright protection for works if the compiler labored over his project, whether or not the work involved originality or creativity.” **

It might seem like a trivialization, but aren’t LLMs just telephone directories? Except instead of phone numbers of a public phone system they contain weights of a mind that’s read a public library? Such works might or might not be proprietary based on “sweat of the brow” copyright laws.

* after Niven/Pournelle https://en.wikipedia.org/wiki/The_Gripping_Hand …

** https://digitalcommons.law.uga.edu/cgi/viewcontent.cgi?artic...

Why do people still want to build businesses in the US or in Silicon Valley? California taxes are already punishingly high, especially after recent rate increases and the 2017 cap on SALT deductions. And now we have a Machiavellian, authoritarian, fascist, tech-illiterate administration interfering with the operation of free markets.

I'm speaking rhetorically, of course. I know Silicon Valley still has the densest concentration of talent and venture capital. The network effects are real. But it is long past time for that to change. I hope entrepreneurs around the the world see this and think twice before moving to the US or starting a business there.

Perhaps they will. I used to work in Silicon Valley and was very much in demand. Now I run my own business from a tax-free state, and my income is high enough that moving back to California would impose a huge financial penalty. I am originally from Europe, and California's marginal tax rate is now so high that I would pay less tax back home. When I moved to the US, the opposite was true.

I'm sure I'm not the only one doing that calculation.

Any other ostensibly AI companies that have just gone public?

Dario must be popping a champagne tonight as regulatory capture has successfully been initiated.

We've all been debating what moat these labs possess. Today we learn the moat is regulation blocking the usage of foreign open source models.

misAnthropic will soon ask for IDs to give you all a nerfed model, and you have no other choice because deepseek/qwen/Kimi will soon be banned thanks to misAnthropic's efforts to lock us into their regulated product.

We will soon realize the long game that Dario has been playing by painting his own product as an existential risk.

Either deliver on your fuckass promise to end the world and replace everyone and make everything shitty forever or fuck off.

Shit or get off the pot already, clowns.

Similar things will happen with China, and the EU has zero-chance of developing frontier models. We are just fucked now.

The funny thing is that the model isn't even impressive. I'd still use ChatGPT over it for anything other than design work. As soon as OpenAI cracks design, I'll never touch an Anthropic model again.

https://news.ycombinator.com/item?id=48498573

Any company that uses Magaland LLMs should be aware of the very real Trump related risk.

What happens if the LLM your firm runs on is disabled tomorrow, because Trump wakes up feeling slightly annoyed...

This is just cost of doing business in corrupt Soviet vessel states like the USA.

Aerospace, defense, semiconductors, telecom, advanced manufacturing equipment, etc

SpaceX entire operation is under ITAR, because even though their rockets are not weapons, rockets are treated as weapons for export purposes

But what about the pelicans ?

for what id expect to be an in memory switch, 2-3h is a while

For Nvidia chips you could've deluded yourself that the US is just anti-china, that position is harder to argue for right now.

Remember, it is all a grift.

American companies have, and continue to, gather data for free from across the globe and, despite our willingness to subscribe, we non Americans will be restricted from latest tech.

This is a big middle finger to me, and my gut reaction is to take my subscription to Mistral and not believe a dime of statements from Americans-- people, companies and government.

Biggest "Meh!" moment of my life so far...

It was over before this announcement. After a couple of days, even though the model was set to fable, it felt like opus. We are back to sticks and stones.

Europe really needs to get some useful sovereign capability and right quick.

There is something called the Streisand effect and they are about to unintentionally get a bunch of more token gamblers into their casino.

We'll see if this backfires hard, but then again constant doomsaying will get yourself under scrutiny and self exclusion (due to the 30+ day retention clause) and this is exactly what Anthropic wants for free marketing.

If I make some statement, and post it on the internet, and someone downloads it in another country, I haven't exported anything. I haven't _moved_ anything. It's the same mythology that casts copying of bytes as tantamount to stealing; it tells a lie about the nature of physical matter as contrasted with the nature of information.

So, OK, let's say this is so - that this activity is not a legitimate target of the export control - why doesn't Anthropic just tell the US government to pound sand, and that they'll choose to ignore this directive?

Is it just because they fear violent reprisal from agents of the state?

And if so - if the reason that we tolerate censorship and damage to the internet, a global collaborative project specifically designed to evolve above the whim of any legacy state is that the actors in question fear violence - haven't we departed democratic notions of decision-making in favor of a "might makes right" approach?

I'm not convinced that the US government can ever embody (or has ever embodied) the republic framework set forth in its founding documents, but at a minimum, for it to do so, and for it to be constrained to those functions, its constituents need to somehow overcome this fear of telling the government, "no, we won't do that. See you in court."

"How dare you release this model to poor people? This belongs only with the ultra-rich!"

They can say whatever they want... but I just have this gut feeling that this is part of it.

pigs get fat, hogs get slaughtered

Gives the word "weights" a more literal meaning.

"I think they are lying to you"

https://youtu.be/zfYsSFY4l18

I really have no clue how Anthropic released this thing without doing any real testing. I did use it on claude.ai with no issues; talking just for code.

They aren't counter to the government, this is all kayfabe to introduce precedence for the US government to be justified in putting controls on AI, expect that by the end of the month there are discussions to regulate Deepseek.

It could be the case that Anthropic created this whole situation on their own, I figured they'd release a "dangerous" model at some point then piggy back off of bad outcomes to dig their regulatory moat

It could also be the case that Altman has close ties to the white house and is using regulatory levers on his competion.

I stand by that its all Kayfabe to make AI look more dangerous than it is (it cant even center a div reliably) to justify controls on Open Source.

You bribe someone in the admin to restrict access after a couple of days of media blitz and user approval, locking in the honeymoon period that new model releases get (remember when GPT-4 was new?). The spooky factor gives it even more marketing, and just before the IPO the Trump admin frees Mythos and they make nice after the DoD debacle.

By thinking for ourselves and writing the code with the keyboard.

Laying the groundwork to limit access to high capability models

2. Get treated like you actually did what you claimed, and face consequences

3. ???

4. Profit

Why do people still want to build businesses in the US or in Silicon Valley? California taxes are already punishingly high, especially after recent rate increases and the 2017 cap on SALT deductions. And now we have a Machiavellian, authoritarian, fascist, tech-illiterate administration interfering with the operation of free markets.

I'm speaking rhetorically, of course. I know Silicon Valley still has the densest concentration of talent and venture capital. The network effects are real. But it is long past time for that to change. I hope entrepreneurs around the the world see this and think twice before moving to the US or starting a business there.

Perhaps they will. I used to work in Silicon Valley and was very much in demand. Now I run my own business from a tax-free state, and my income is high enough that moving back to California would impose a huge financial penalty. I am originally from Europe, and California's marginal tax rate is now so high that I would pay less tax back home. When I moved to the US, the opposite was true.

I'm sure I'm not the only one doing that calculation.

They'll make a book with mythos's code and sell it like Phil Zimmerman did with PGP?

Anthropic: "Oh... American access only, you say? I'm sorry, we can't promise that (because VPNs and US-local cloud hosting and all that), so we need to turn it off completely."

...probably.

If so, I wonder what turn the political shenanigans will take next?

Based on the actions of the current administration and the short-sighted tech oligarchs who have been consistently pushing towards neo-fascism/neo-feudalism, probably one that further degrades trust all around and gives China even more of a leg up.

Let's see!

Wonder how many US-based early-stage startups are using Opus to research incorporating and moving overseas at this very moment.

EU isn’t tenable, UK is iffy. Australia? Thailand? Who wants to be innovation-friendly?

Even if they negotiate a way out of this particular spat, this is just the start of securing this technology in the name of national security. Does this pop the bubble? What happens to the trillions invested in this AI craze? When do we outlaw Chinese models?

Although it's gonna be more difficult to come up with a Fable competitor than a m365 one

We are witnessing the largest stock manipulation by the United States government in history.

The big one here really is "including foreign national Anthropic employees." Funny, I called out that there probably is a Chinese spy at Anthropic in my previous comment a few days ago. Looks like they are catching on, good luck getting rid of all of them.

The bad news is that they will probably start imposing restrictions on Chinese models.

This fits the model established with RSA, PGP, and the Sony PS3. They were export controlled for quite some time. I don't think there was ever any actual danger with any of those things, and today it feels especially quaint, but they fit the model of "corporation makes wild-ass claims of superiority of their tech and USGov takes them at their word."

My big problem with this is that it's applied so narrowly to Anthropic. This should be levied against OpenAI, Google, and xAI as well. There is systemic risk with generative AI being used for deep fakes and other propaganda generation at scale that needs to be addressed.

But unfortunately, that's not what is happening here. What's happening here is a political hit job. There's one of two things happening: either USG has been roped into burying a competitor for (OpenAI/xAI/whatever), or it's been roped into creating a superiority narrative for Anthropic, such that in two years when this admin is finally ousted, Anthropic gets to enjoy a floodgate of new attention as the new regime bulk CTRL+Zs everything Trump's lackeys did. It might even be both at the same time, given the connections of the major investors. This all could get well be a stirring of the pot to see what comes out.

I can't decide which is worse.

I was this close to predicting the (complete) suspension in some sense but for different reasons (compliance)

but this is a bit of shit-show at this point and I am unsure how involved Anthropic is.

Maybe Anthropic gave us access to Fable 5 for some point so that we can all discuss it and see how Anthropic is relevant as compared to gpt 5.5 (not that I like ClosedAI more but fact is that I have heard decent things about it)

So I am not sure if this suspension can lead to an idea like me. Anthropic showed us a really competent model and then removed it and now you might have to form a custom deal with Anthropic or similar maybe similar to mythos if you wish to access the models and they can rake in extra dollars from top clients.

but they were already doing that with mythos, then what was the point of Fable. PR support that Anthropic hasn't fallen off?

Or maybe I am just overthinking and Anthropic is genuinely hurt by this decision given that they did release the model but US govt said no and US govt and Anthropic has some beef with each other.

There are so many factors for this news and the narrative/implications of that, that it is hard to understand what really happened unless some more news comes (IMO)

Dario, yesterday: "I am grateful to see the Trump administration’s Executive Order move incrementally towards a greater role for government in AI, though Anthropic’s proposal recommends even further action."

Trump, today: Further action

Dario, today: "Waaaah! This petard I asked President Trump for hoisted my ass halfway to the Moon! Nobody warned me he'd do something like this! No fairrrr!"

> In fact, our safeguards are so strong that many users have complained that they are overly broad.

When you ask for regulation, you get regulated. Welcome to the real world

The US has long been catastrophically corrupt, with a pay-to-play government, but this army of grifters and thieves have turned every dial to 11.

And now this. How would they even enforce this restriction when they can't know what nationality the end user behind some API query belonging to a company account has? It seems like nobody is thinking things through anymore and the end result is total unreliability from every angle. What a huge mess all of it, sigh.

1. Likely: This is a completely contrived marketing stunt. Release a spooky sounding model, market it as such, then use the narrative to regulate open source models and dig your regulatory moat. Notice the emphasis on "foreign nationals" here.

2. Likely: Dario has a meeting at the White House this next week (confirmed by Trump this week), and this is being used to get leverage over him.

3. Uncertain: Altman has closer connections to the Trump regime and is pulling in favors to level the playing field and slow down competition.

Regardless, This is a win for Anthropic and Dario.

1. This will jump start a more serious discussion of regulation around LLMs (which are ultimately useless, regulation is just there to make them more money).

2. They can then only serve these models at their high, usage based pricing and bleed less money while serving up tons of interest because people are going to want to pay more for the "spooking banned model".

3. This will probably come with the perks of verifying everyone's identity who uses it (to comply with no "foreign nationals rule). I'll leave that up to your imagination for how that's beneficial to all the powers that be, including Anthropic. I expect this to be used as an excuse for pushing ID requirements across the AI product landscape.

4. There will be a more serious discussion about sanctioning Chinese AI labs, expect that to start happening very soon.

Either way, its all dumb. Don't use an LLM to do your work for you and save your brain. Your brain is literally atrophying by using these models the way most of you guys do. You don't need them.

And LLMs? I’m just going to run open source models on local hardware, it all seems like the 1980s with compilers. Why not just submit by prompts to a high quality model running on so-so-hardware overnight, like the devops cycle with compiling a big codebase? And oh look, nobody pays for compilers anymore, who compiles their code in the cloud?

The funniest part of all of this is that the very people hyping all this - they’re the ones that AI could most easily replace. They have zero specific detailed knowledge - they just orchestrate. Agents are great at orchestration, right? But then, who needs the shareholders, anyway.

Good. 4.8 is good enough.