US export directive forces Anthropic to pull AI models; cybersecurity experts urge reversal
5 verified4 unconfirmed
The U.S. government issued an export control directive to Artificial Intelligence company Anthropic, effectively forcing the company to pull its latest AI models, Fable 5 and Mythos 5, offline. The directive, which bans non-Americans from accessing the models, cited an unspecified national security concern. Anthropic stated it does not believe the government’s action was warranted. In response, more than 100 cybersecurity executives and experts urged the Trump administration to lift the directive, arguing it could harm U.S. cybersecurity and give adversaries an advantage. The experts warned that taking away the best cyber defense capabilities without good reason is dangerous. The directive marks one of the U.S. government’s most significant steps to restrict access to advanced AI models and has raised questions about the government’s reasoning and broader implications for the tech industry.
What’s verified
The U.S. government issued an export control directive to Anthropic regarding its Fable 5 and Mythos 5 AI models.
Anthropic took both models offline to comply with the directive.
The directive bans non-Americans from accessing the models, citing national security concerns.
Over 100 cybersecurity executives and experts urged the administration to lift the directive, arguing it could undermine U.S. cybersecurity.
The move has been described by researchers and experts as potentially dangerous and counterproductive.
Not yet confirmed
The specific national security concern cited in the directive has not been publicly detailed by the government.
The Commerce Department has not responded to requests for comment on the directive.
A dispute between Anthropic and the Pentagon led Defense Secretary Pete Hegseth to seek a supply chain risk declaration against the company, according to one source.
Questions remain about whether the government misinterpreted a research paper describing a guardrail bypass, or if other political factors influenced the directive.
Misconceptions
Sources indicate the directive was not based on a genuine security vulnerability or jailbreak of the models, as some observers initially assumed. Researchers stated that the alleged bypass was a simple difference in prompt wording and should not have triggered export controls. Anthropic has also expressed doubt that the government’s concerns warranted such a step.
Key figures
Anthropic (AI company)
U.S. Commerce Department
President Donald Trump
Defense Secretary Pete Hegseth
Katie Moussouris (cybersecurity researcher)
Justin Hendrix (editor, Tech Policy Press)
Sources: TechCrunch, abcnews.com, theregister.com
