Anecdotally, I think this is a much bigger cleanup than just talking to the administration.
I think there's a wider damage done which there is no coming back from. And this is for the USA, not Anthropic.
The chance that sovereignty and rules such as this could be applied to AI was a concern that a lot of people had, but the risk was unknown.
Speaking for myself, I had guessed this would happen at some point of time. I was expecting/hoping it'd be years away.
However given the events in the last few days it greatly increases my concern with building any product which can depend on an on an API which could go away for a number of my customers.
I've been experimenting with other open weight models hosted in favorable sovereign countries for a few months, but this accelerates something which was an experiment to now being a must-have.
I don't think it is going to be easy for any of the parties to repair this easily.
I see it exactly the same, self hosted LLM will be the future. They may not be SOTA, but it is better to have a Trabant in your garage than being denied use of shared Ferrari, because somebody outside of your control had hissy fit.
I like a good car analogy, but it doesn't hold up in this case. What am I going to do with a Ferrari that's durable and makes me money, the same way that Mythos/Fable can output me better code than Opus 4.x? I mean I guess I could take a picture of me in a Ferrari, and then when I show off that picture but can't produce the Ferrari, I look dumb, but Mythos generated code/artifacts are still downloadable and runnable, if you got far enough before the cut off. Judging from the quality/availability of the few days we had it, it's a shared resource anyway. The usage limits were too low to give everything to Fable to do.
As I understand it, ITAR regulations for export controls have just been applied to any form of Mythos.
These are overseen by U.S. Departments of State and Commerce, and forbid foreign nationals from access to any form of Mythos, either within or outside the U.S.
Only U.S. citizens and immigrants that are holders of a "green card" may now access Mythos.
It appears that Anthropic does not have internal controls to implement these restrictions in any form, so the only option was to shut Mythos down.
Penalties for ITAR violation can reach ten years in prison and a million dollars per violation. (I can post a link to those details if there is any interest.)
As long as Anthropic is a U.S. company, there is no escaping this.
> Only U.S. citizens and immigrants that are holders of a "green card" may now access Mythos.
It was my understanding that not even green card holders may access Mythos. Normally when restrictions like this are put in place, you need a exemption as a green card holder. A geen card is just a permit to live and work in the US. Its not the same as citizenship.
> Security Clearance: Green Card holders are generally prohibited from jobs that require high-level security clearances or sensitive government/military roles reserved exclusively for U.S. citizens.
That's not true. ITAR and security clearance are entirely separate regulatory regimes. For ITAR purposes, being a permanent resident is good enough. I used to work for a defense contractor, and we hired plenty of green card holders. They were not in general assigned to work that required a clearance, but plenty of defense-sensitive, ITAR-controlled work is done by green card holders.
Were those ITAR export controls chosen because they really are the most appropriate tool for this particular case, or because they could be deployed at a very short notice?
Individuals can leave, but the company cannot transfer restricted intellectual property.
Europe has extradition treaties, so the U.S. can force anyone in Europe back to the U.S. for criminal indictment who demonstrates inappropriate possession of this technology.
It's not "AI light bulbs", the point is to use smart meters for load shedding and load rearranging. Why wouldn't you want for people and companies to be able to use cheaper electricity in off-peak hours?
doesn't seem to be a political meeting, but rather a technical one. Maybe they'll review the jailbreaks and demonstrate that can be replicated in any model out there?
Yep, hard to feel bad when the whole reason they're in this mess is because they tried to introduce FYD about their own product as an argument why open models need to be banned.
After seeing headlines about the ufc fight on the White House lawn and seeing headlines about how Japanese World Cup fans clean up trash after the match I had an image of programmers doing lawn care at the White House...
i 100% thought this article was going to be about cleaning up the idiocracy festival from the past weekend. didn't even question the fact that people from anthropic would be there to help, seemed like a fitting asskiss.
That was seriously what I thought the headline is about. Anthropic flying in their employees to clean up the trash after Trumps birthday party for free to demonstrate their obedience with a good show of licking the Dear Leader's boots.
Extremely silly of course, but entirely in line with all the other news currently coming out of the US ;)
Being angry about one thing does not mean someone is ok with another.
And the context is quite different. The complaint about Trump relates more to what he is doing. Your complaint about Biden relates to something a party invitee did.
I guess a similarity is an element of virtue signalling. But if that is something you have a problem with then you must really hate Donald Trump because that seems to be his entire presidency. And the fight against homophobia is a legitimately virtuous activity. Right and wrong do exist.
I just can't get his popularity, he isn't smart, he isn't a great speaker, his opinions on actually important stuff is often... just sad to hear, hyping up a 'bro effect and thats about it. Its like listening to that old buddy from primary school who struggled to get through high school, and now as an adult gathered some interesting opinions about the world, politics and humans that you really are not interested in.
Tons of better folks online, if somebody really has nothing better going in their short lives to listen to blahs for 3 hours, repeatedly. I guess there are worse ways to spend time though
He helped Trump win by promoting the guy right before the election, together with Theo Von. For clicks, and undeniably, money. Both are tools. One of them with a kick of a mule, still a tool.
Cheaper and less harmful to long-term and international prospects to just make a gold-plated “AI Peace Prize” award and gift it in some ceremony. Couple of hundred thousand dollars, works great. Ask Tim Apple.
Anthropic wants to do things that they don’t want to write in an email.
How unsurprising.
Amodei has always been and always will be a creep. Unfortunately too many people here don’t want to hear it about their favourite billionaire, just like they didn’t want to hear it about all the other sociopaths that tech has minted in the last 20 years.
Unfortunately it's not widely distributed information, if you care so much about shedding a light on Amodei's shadiness it would help a lot folks like me to understand what you are talking about.
I know of Sam Altman's shenanigans from way before OpenAI (I used to be on reddit a lot more when `sama` appeared over there) but I have no idea about Amodei.
Help us out because otherwise it really does read as yet another conspiracy theory on the internet.
Despite already being familiar with such a person (in the form of Altman), for some reason you are trying to find minor reasons that they are different people, and it’s distracting you from the major reasons that they are actually the same people.
I can’t offer any more than that, because it’s all I have. It’s about observation not conspiracy.
I'm not trying anything, I want more information if it's available, otherwise it's all conjecture... Even though I dislike anyone at that level of power, the way you phrased things implied there was more information. You are just expressing an opinion and that's ok, just don't make it sound like there's more to it than that.
Anecdotally, I think this is a much bigger cleanup than just talking to the administration.
I think there's a wider damage done which there is no coming back from. And this is for the USA, not Anthropic.
The chance that sovereignty and rules such as this could be applied to AI was a concern that a lot of people had, but the risk was unknown.
Speaking for myself, I had guessed this would happen at some point of time. I was expecting/hoping it'd be years away.
However given the events in the last few days it greatly increases my concern with building any product which can depend on an on an API which could go away for a number of my customers.
I've been experimenting with other open weight models hosted in favorable sovereign countries for a few months, but this accelerates something which was an experiment to now being a must-have.
I don't think it is going to be easy for any of the parties to repair this easily.
Back when encryption and PGP was the hot topic, similar things happened with sovereigns.
I see it exactly the same, self hosted LLM will be the future. They may not be SOTA, but it is better to have a Trabant in your garage than being denied use of shared Ferrari, because somebody outside of your control had hissy fit.
I like a good car analogy, but it doesn't hold up in this case. What am I going to do with a Ferrari that's durable and makes me money, the same way that Mythos/Fable can output me better code than Opus 4.x? I mean I guess I could take a picture of me in a Ferrari, and then when I show off that picture but can't produce the Ferrari, I look dumb, but Mythos generated code/artifacts are still downloadable and runnable, if you got far enough before the cut off. Judging from the quality/availability of the few days we had it, it's a shared resource anyway. The usage limits were too low to give everything to Fable to do.
[dead]
https://archive.md/KJK8v
They should fly staff to EU instead
As I understand it, ITAR regulations for export controls have just been applied to any form of Mythos.
These are overseen by U.S. Departments of State and Commerce, and forbid foreign nationals from access to any form of Mythos, either within or outside the U.S.
Only U.S. citizens and immigrants that are holders of a "green card" may now access Mythos.
It appears that Anthropic does not have internal controls to implement these restrictions in any form, so the only option was to shut Mythos down.
Penalties for ITAR violation can reach ten years in prison and a million dollars per violation. (I can post a link to those details if there is any interest.)
As long as Anthropic is a U.S. company, there is no escaping this.
https://fortune.com/2026/06/14/how-a-warning-from-amazon-led...
> Only U.S. citizens and immigrants that are holders of a "green card" may now access Mythos.
It was my understanding that not even green card holders may access Mythos. Normally when restrictions like this are put in place, you need a exemption as a green card holder. A geen card is just a permit to live and work in the US. Its not the same as citizenship.
> Security Clearance: Green Card holders are generally prohibited from jobs that require high-level security clearances or sensitive government/military roles reserved exclusively for U.S. citizens.
That's not true. ITAR and security clearance are entirely separate regulatory regimes. For ITAR purposes, being a permanent resident is good enough. I used to work for a defense contractor, and we hired plenty of green card holders. They were not in general assigned to work that required a clearance, but plenty of defense-sensitive, ITAR-controlled work is done by green card holders.
Were those ITAR export controls chosen because they really are the most appropriate tool for this particular case, or because they could be deployed at a very short notice?
Mythos was advertised as being able to do massive damage to IT systems over government and the private sector.
How could restriction not be appropriate.
So Anthropic can't even move to Europe now since that would be considered a crime against national security?
Like trying to move a missile industry from USA to EU.
Individuals can leave, but the company cannot transfer restricted intellectual property.
Europe has extradition treaties, so the U.S. can force anyone in Europe back to the U.S. for criminal indictment who demonstrates inappropriate possession of this technology.
Simply move the training datasets and recreate it in the EU.
However, I think the US would treat this as treason (like they wouldn't want nuclear scientist to work for other nations).
I believe that Dario Amodei (Anthropic CEO) was born in Italy, and is already an EU citizen.
He is probably reviewing many exit options today.
no, he's american born. He would be entitled to (or already have) Italian citizenship by ius sanguinis ("blood right") tho.
You're lying.
And what would they be doing there?
...and take their cluster of twelve M1 Max's with them.
No need, the EU will just order more AI light bulbs, when everything is sorted: https://www.youtube.com/shorts/dUPJuSZF1Is
It's not "AI light bulbs", the point is to use smart meters for load shedding and load rearranging. Why wouldn't you want for people and companies to be able to use cheaper electricity in off-peak hours?
> Senior technical Anthropic staff
doesn't seem to be a political meeting, but rather a technical one. Maybe they'll review the jailbreaks and demonstrate that can be replicated in any model out there?
I hope no, otherwise also Claude and GPT will be banned XD
How could it not be both?
Wouldn't it be funny if US Gov takes a 10% stake in Anthropic.
Donald Trump, if you can hear me...
None of this would have happened if Anthropic had just bought a few million dollars worth of Trumpcoin :melting_face:
Anthropic: "Our new models are so mighty and dangerous, we'll limit access."
US admin: "Thanks for the warning, we'll do the same then"
This was an epic (and deserved) own-goal.
(btw heres what most likely happened, they refused to fix a jailbreak: https://x.com/DavidSacks/status/2065853007619588171)
> what most likely happened
sounds super biased
UPD just look up "anthropic david sacks"
> epic
are you unironically happy about what is happening right now?
Yep, hard to feel bad when the whole reason they're in this mess is because they tried to introduce FYD about their own product as an argument why open models need to be banned.
What a dumb take, life isn’t a meme
This is retaliation by the us gov for not unlocking Claude to use on their weapon. Not good for anybody
I hope they are good at cage fighting. Thats what Trump is all about
After seeing headlines about the ufc fight on the White House lawn and seeing headlines about how Japanese World Cup fans clean up trash after the match I had an image of programmers doing lawn care at the White House...
i 100% thought this article was going to be about cleaning up the idiocracy festival from the past weekend. didn't even question the fact that people from anthropic would be there to help, seemed like a fitting asskiss.
I only saw a contrast of Japanese World Cup vs NY Knicks
That was seriously what I thought the headline is about. Anthropic flying in their employees to clean up the trash after Trumps birthday party for free to demonstrate their obedience with a good show of licking the Dear Leader's boots.
Extremely silly of course, but entirely in line with all the other news currently coming out of the US ;)
[flagged]
Are you upset that the Whitehouse invited citizens to the lawn, or are you upset that they banned someone who was behaving distastefully?
You can't ban the president from the White House.
/s
Being angry about one thing does not mean someone is ok with another.
And the context is quite different. The complaint about Trump relates more to what he is doing. Your complaint about Biden relates to something a party invitee did.
I guess a similarity is an element of virtue signalling. But if that is something you have a problem with then you must really hate Donald Trump because that seems to be his entire presidency. And the fight against homophobia is a legitimately virtuous activity. Right and wrong do exist.
[flagged]
Funny how you are okay with pedophilia.
Not following? See "Please see HN rule #1, this is not the community for that kind of discourse."
Ignoring the whataboutism here, what about it was distasteful?
Please see HN rule #1, this is not the community for that kind of discourse.
[flagged]
The UFC athletes were also topless...
[dead]
And they were banned. What’s the real problem here. Spell it out for us.
Mighty fine AI you have there, be a shame if something were to happen to it...
Also, supposedly biggest MMA event Rogan doesn't want to be on.
Something is up
Rogan is a tool.
Interesting statement with various meanings :)
I just can't get his popularity, he isn't smart, he isn't a great speaker, his opinions on actually important stuff is often... just sad to hear, hyping up a 'bro effect and thats about it. Its like listening to that old buddy from primary school who struggled to get through high school, and now as an adult gathered some interesting opinions about the world, politics and humans that you really are not interested in.
Tons of better folks online, if somebody really has nothing better going in their short lives to listen to blahs for 3 hours, repeatedly. I guess there are worse ways to spend time though
He helped Trump win by promoting the guy right before the election, together with Theo Von. For clicks, and undeniably, money. Both are tools. One of them with a kick of a mule, still a tool.
[dead]
Nothing a few million dollars here and there can't fix ...
Please don't post unsubstantive comments.
Kissing the ring also helps - I won't be surprised if Anthropic's next model goes full MAGA
Cheaper and less harmful to long-term and international prospects to just make a gold-plated “AI Peace Prize” award and gift it in some ceremony. Couple of hundred thousand dollars, works great. Ask Tim Apple.
I don’t know, for all the bad people say about Amodei, everybody seem convinced he’d rather burn his company than let something like that happen
Anthropic wants to do things that they don’t want to write in an email.
How unsurprising.
Amodei has always been and always will be a creep. Unfortunately too many people here don’t want to hear it about their favourite billionaire, just like they didn’t want to hear it about all the other sociopaths that tech has minted in the last 20 years.
> Amodei has always been and always will be a creep
Like at what point do the weird and random Amodei smears start to seem like an orchestrated attempt?
They’re not weird or random when you’ve lived around many people like him before.
Apparently not enough of us have unfortunately, so you keep elevating them when you should know better.
So here I am trying to give you a heads up.
Unfortunately it's not widely distributed information, if you care so much about shedding a light on Amodei's shadiness it would help a lot folks like me to understand what you are talking about.
I know of Sam Altman's shenanigans from way before OpenAI (I used to be on reddit a lot more when `sama` appeared over there) but I have no idea about Amodei.
Help us out because otherwise it really does read as yet another conspiracy theory on the internet.
Despite already being familiar with such a person (in the form of Altman), for some reason you are trying to find minor reasons that they are different people, and it’s distracting you from the major reasons that they are actually the same people.
I can’t offer any more than that, because it’s all I have. It’s about observation not conspiracy.
Ok, I thought there was more to it.
I'm not trying anything, I want more information if it's available, otherwise it's all conjecture... Even though I dislike anyone at that level of power, the way you phrased things implied there was more information. You are just expressing an opinion and that's ok, just don't make it sound like there's more to it than that.
> I can’t offer any more than that, because it’s all I have
So this all boils down to you think he must necessarily be like Sam Altman because he runs an AI lab, and so you're extrapolating from that?
This could mean literally anything.
No it means one thing - they are friend-less sociopaths.
It could imply many things, in terms of how that actually manifests in their behaviour. But I don’t care about that at all.
I only wish us plebs would stop inflicting them on ourselves.
billionaires are all evil, we get it; do you have anything relevant to say?
Thanks for coming to my TED talk.