Here we go again.
The Once And Future Fable
The United States Department of Commerce, as per a letter from Commerce Secretary Howard Lutnick, apparently in response to a narrow jailbreak identified by Amazon, has classified Fable 5 and Mythos 5 as being subject to US export controls. That explicitly means cutting off access to all ‘foreign nationals,’ even within the United States, even if they are Anthropic employees.
Given Anthropic has no means to verify citizenship at this time, that meant complete shutdown of the model, at least for the time being.
This Action And Its Implementation Are Absurdly Stupid
If you take the action at face value, rather than as an attempt to lash out at Anthropic, there is no way to pretend this is not deeply, deeply stupid.
What Happens Now?The United States Department of Commerce, as per a letter from Commerce Secretary Howard Lutnick, apparently in response to a narrow jailbreak identified by Amazon, has classified Fable 5 and Mythos 5 as being subject to US export controls. That explicitly means cutting off access to all ‘foreign nationals,’ even within the United States, even if they are Anthropic employees.
Given Anthropic has no means to verify citizenship at this time, that meant complete shutdown of the model, at least for the time being.
Anthropic: The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees. The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance. Access to all other Anthropic models will not be affected.The justification for this appears to be rather flimsy, at best, and based on lack of understanding of what even is a jailbreak or how defense in depth works.
Dean W. Ball: I can’t tell if this is lawfare against Anthropic in particular or extreme national-security hawkery. Regardless, it is simply cartoonish.
Anthropic: We received the directive from the government today at 5:21pm (ET). The letter did not provide specific details of its national security concern. Our understanding is that the government believes it has become aware of a method of bypassing, or “jailbreaking” Fable 5.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.
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.
We apologize for this disruption to our customers. We believe this is a misunderstanding and are working to restore access as soon as possible.That left Anthropic with no options but to entirely withdraw it from the market, at least for the time being, since they have no way to verify who is and is not a United States citizen. [...]
This Action And Its Implementation Are Absurdly Stupid
If you take the action at face value, rather than as an attempt to lash out at Anthropic, there is no way to pretend this is not deeply, deeply stupid.
Dean W. Ball: If this is true, it is just baffling. An administration whose posture is that we *should* export advanced AI chips to China, which also wants to ban… Britain (and every other non-American on Earth)… from using our best models? I have no words.
zooko ⓩ: Judging from [the announcement], I imagine that some senior government official was shown a jailbreak—something they had never seen before and didn’t know about—and this was their kneejerk reaction.
Dean W. Ball: If implemented as this reporting suggests, Anthropic’s latest models would be subject to export controls to all *non-Americans,* including non-American nationals based in the US. This means you should expect to have to prove your citizenship to use Anthropic models. [...]
It is a regular thing for the Executive Branch of the United States Government, these days, to issue declarations of policy that are, to use the technical term, absolutely bonkers and stunningly destructive with no reasonable way to implement them, often without stopping to realize what they are doing.
It is also a regular thing for them to then quietly walk those policies largely or entirely back, once the consequences become clear, leaving only relatively minor total devastation in their wake.
Alas, it is also a regular thing for them to leave at least a substantial portion of the new stupid and destructive policy in place indefinitely, and sometimes we keep all of it, or they even keep going further.
Or Anthropic could give the White House what it wants, no matter who is right about whether doing so makes any sense.
We are not short on examples of any of this.
One thing that must now be considered is that many employees of OpenAI, Google and Anthropic, and other AI labs, are not United States persons.
Yo Shavit (OpenAI): Unless this changes, OpenAI researchers on visas need to plan for the fact they’ll probably lose access to internal models, and therefore their ability to do their jobs moving forward, sometime in the next couple months.If we drive all foreign talent out of our AI labs, and otherwise actually go down the current road, that is one of the few things that could put China and other competitors back in the game in earnest, both slowing us down and speeding them up.
I hope the company acts to prevent that.
dave kasten: Uhhh so incidentally, does anyone have a plan to prevent all the non-US citizen AI scientists from going to join foreign labs after they get bored of playing Wordle at work for a month, or are we just sort of planning on having the greatest counterproliferation failure since we deported Qian Xuesen in 1955 and gave Mao a rocket program?
At Anthropic, Amanda Askell and Andrej Karpathy are examples of employees who suddenly are unable to work with Claude Mythos 5, even after Anthropic sorts out a new access control system.
by Zvi Mowshowitz, DWAtV | Read more:
***
On Friday evening the United States Government has forced Anthropic to take down all access to Fable and Mythos.It’s been a rough weekend. [...]
1. More details have come to light. There remains some fog of war, but we now have a rather good idea why Claude Fable and Mythos were, deeply stupidly, taken down.When Anthropic did not do so, the White House hit them with an export restriction that they knew would force Fable and Mythos down for everyone.
2. A narrow jailbreak was discovered, of the type Anthropic warned in advance obviously existed. All demonstrated outputs are things GPT-5.5 can not only produce, but produce without any sort of jailbreak or bypass.
3. The White House demanded Anthropic take down Fable to ‘fix’ the situation, and did not listen when Dario tried to explain that there was no situation to fix.
A lot of nihilists are justifying this decision, and blaming Anthropic, all of whom are very much confirming that they adhere to Dean Ball’s portrait of the United States Government as a dying NPC hospice patient we have to properly placate with the proper vibes and genuflection so they don’t lash out at us. Except they equate this with strength and righteousness, because might makes right, power and vibes.
This is a fast developing story with a large speed premium, so I apologize for any errors, and for the structure likely not being ideal. We do the best we can.
What we do not know is:
1. What was motivating the government to make these decisions.
2. How deeply they were confused about how any of this works.
3. Whether they demanded and are demanding a narrow fix or a global fix. Narrow fix is probably easy. Global fix is probably impossible.
4. What they intend to do next and what they are trying to accomplish.The good outcome would be that this is a terrible misunderstanding, a reflection of a panic reaction, which can be sorted out quickly, after which we can restore access. Or where they otherwise face enough pressure they quickly realize they made a mistake, or Anthropic can do something to quickly assuage their concerns even if it is dumb. There will still be a terrible precedent set, which comes with a lot of permanent damage to trust in American AI, to our business climate, to our ability to employ vital foreign AI talent, to America’s relationships to its allies, to the progress of Project Glasswing and our cyber security, and to the rule of law.
***
[ed. In addition, see: Seductive Salience (the inevitable politization of AI regulation).]
