What smart people are saying about the sudden ban on foreign use of Anthropic’s new AI models

• Anthropic said the Trump Administration ordered it to block foreign access to two powerful AI models.

• In response, the company cut off all access to Fable 5 and Mythos 5.

• The technology world reacted quickly and surprisingly to this situation.

Anthropic’s announcement Trump Administration His order to block foreign access to Fable 5 and Mythos 5 caused the company to completely cut off access to these AI models, sending a shockwave through the tech world.

Here are some reactions on social media to the clashes between the two sides: White House and Anthropic.

Dean W. Ball, senior fellow at the American Foundation for Innovation

“If this is true, it’s just confusing. This is an administration whose stance is us.” should do Should we export advanced AI chips to China, which wants to ban Britain (and every non-American in the world) from using our best models? I have no words.

“I can’t tell if this is specifically an anti-Anthropic law or extreme national security smuggling. Either way, it’s just cartoonish.”

Peter Girnus, senior threat researcher at the Zero Day Initiative

“Two things are true at the same time.

“First: For months Anthropic marketed Mythos as too dangerous to release. Sam Altman said ‘it was incredible marketing to say we built a bomb. The Department of Commerce has now officially admitted that it was a bomb. If you describe your product as a munition in every press release, eventually the government will take your word for it. They wrote the legal basis themselves and named it the trademark.

“Second: we’ve done this experiment before. In the 90s, the government classified encryption as munitions under ITAR. Activists defeated this by printing PGP’s source code as books, because books are speech-proof and floppy disks are weapons exports. A T-shirt with three lines of RSA Perl was legally munitions. The controls collapsed because the math didn’t end up at customs.”

“The new wrinkle is the ‘deemed export’ rule: Showing controlled technology to a foreign national inside the United States is tantamount to exporting it abroad. So Anthropic’s own foreign national employees are now excluded from the model they’re building. The munitions are in the building, and the people who built it are not allowed to look at it.”

Marc Andreesen, partner at Andreessen Horowitz

Chris McGuire, senior fellow for China and emerging technologies at the Council on Foreign Relations

“I actually think targeted export controls on model access are prudent. However, blanket checks on all countries on a single model without warning are highly questionable. Imposing equally far-reaching export controls that also restrict access to foreign nationals is simply absurd and would clearly result in the model being withdrawn from distribution as has just happened.”

“Export controls are a critical tool and an extremely powerful tool. Used correctly, they have the potential to greatly expand the US lead in AI. Used incorrectly, they will hinder AI development. The Commerce Department’s export control strategy has been completely inconsistent and sabotaging. It ships powerful AI chips to China, fails to enforce controls to prevent Chinese smuggling, creates massive loopholes that allow AI chips to be shipped to China, and prevents US AI companies from publishing their own models.”

“This needs to end. We urgently need a smart export control strategy that implements strong export controls while giving U.S. companies an advantage by denying our competitors access to advanced technology. Commerce and BIS consistently do the opposite. If BIS doesn’t understand how to use its powers or what the consequences of its actions are, then it needs to find some new personnel who can actually implement a competent export control strategy. The current strategy is incoherent and self-defeating.”

Matthew Pines, CEO of Physical Superintelligence

“This will send shockwaves through every lab and neolab… US export control laws operate under a strict liability standard… they are a very sharp sword…”

Dan Shipper, CEO of Every

“My take on this situation right now is that they will lift the ban in a few days and the net effect will be increased demand for Fable

“But these things are extremely disruptive and distracting for people inside the company. The only comparable scenario I can remember is Sam Altman’s firing, and that was resolved relatively quickly. Although things bounced back, I think it disrupted their momentum for a while.”

“I hope for a good result here!”

Josh Pigford, Founder of Baremetrics

“Anthropic hasn’t done themselves ANY favors with their hype over the last 6-12 months. But I also guarantee it has nothing to do with national security.”

Peter Barnett, researcher at the Machine Intelligence Research Institute

“If it is true that USG exports controlled Fable because another company said it could jailbreak it, then we could be facing a regime where AI companies regroup each other’s models before USG allows distribution.”

Yusuf Mahmood, Director of Artificial Intelligence Policy at America First Policy Institute

“The US government has ordered Anthropic to suspend all foreign national access to Fable 5 and Mythos 5 within or outside the US. As a reminder, a large percentage of technical workers at all frontier AI labs (including Anthropic) are likely foreign nationals.”

Jeremy Howard, co-founder of fast.ai

“I don’t agree with this decision and I don’t like it.

“But also…

“HOW DID ANTHROPIC NOT SEE THIS ARRIVAL?

“This the “It’s an obvious answer to ‘This is too dangerous for anyone but us to use’, because it’s based on a premise that almost no one accepts (‘we are uniquely good’).”

Ryan Brewer, member of the OpenAI technical team

“If the US government continues in this direction, eventually you will only be able to access border intelligence from a small set of buildings in the Bay Area. It’s a shame.”

Ketan Ramakrishnan, Yale Law professor

“The federal government will aggressively regulate AI developers. The question is whether this regulation will be done wisely; whether there will be a role for Congress and public debate, or whether it will just be opaque executive action; etc.”

Yann LeCun, executive chairman of AMI Labs

“Dario Amodei’s ridiculous fearmongering about Mythos/Fable (and AI in general) finally pays off:

“The US government prohibits use by non-Americans, Including foreign employees in the US

“A man reaps what he sows.”

Peter Harrell, visiting scholar at Georgetown Law School

“I find it ridiculous and un-American for the government to tell me that, as an American, I cannot use an advanced AI model because of an unspecified and non-public security threat. We should regulate AI, but it should be transparent and based on neutral rules, and not at 5pm on a Friday.”

Alan Rozenshtein, research director at the Law Institute

“This could be the first major First Amendment AI case.”

Aaron Levie, CEO of Box

“This is a huge turning point for AI regulation.

The government is beginning to think that some models are too powerful for certain uses, setting a precedent for a possible range of controls in the future.

I agree that this is unnecessary and that we should first regulate the use of artificial intelligence as opposed to basic models. But equally, there are many people who prefer this outcome.

“Either way, we are unlikely to return to a world without a much more meaningful government involvement in the pace of AI advancement.”

Gary Marcus, artificial intelligence researcher

“Commerce’s shocking executive order this afternoon — effectively shutting down Anthropic by cutting off access to Mythos 5 and Fable 5 for many of its own employees — appears to be both overly dramatic and counterproductive for the U.S. AI industry.

Maybe he’ll do China a favor. Of course, every Chinese person working at a US AI company (and there are many) will consider returning to competition in China as soon as possible.

Investors will begin to wonder whether American AI companies can thrive in this atmosphere.

“If you want an example of AI regulation that could stifle innovation, this is it.”

Timnit Gebru, founder of the Distributed Artificial Intelligence Research Institute

“Lol Anthropic made such a big deal about their model being ‘dangerous’ that they thought the only result would be hype + something stopping China from doing something.”

If you like this story, don’t forget to follow Business Content on Yahoo.