Full story
Export Control Shuts Models
Anthropic disabled its two most powerful AI models, Claude Fable 5 and Mythos 5, after a U.S. government export-control directive ordered the company to suspend access to the models by any foreign national, including foreign national Anthropic employees.
“The AI firm Anthropic has blocked access to its newly released cutting-edge software, following an order by the United States government”
Time reported that the Trump administration cited national security concerns when it issued the directive on Friday and that users reported being unable to access the models on Saturday.
Anthropic said it had to “abruptly disable” access to the models for all customers to comply with the order, and the shutdown followed the models’ public rollout just days earlier.
The BBC said the “net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance,” linking the action to U.S. authorities’ security fears shortly after the models’ release.
Jailbreak Dispute and Debate
Anthropic said the government’s concern involved a method of “jailbreaking” Fable 5, and the company argued it had received only verbal notice of a “potential narrow, non-universal jailbreak.”
In its statement, Anthropic said it had “reviewed a demonstration of this specific technique being used to identify a small number of previously known, minor vulnerabilities,” and it said other publicly available models could discover the same issues without requiring a bypass.
The BBC quoted Anthropic saying “Our understanding is that the government believes it has become aware of a method of bypassing, or 'jailbreaking' Fable 5,” while the company also said it disagreed that the finding should trigger a recall.
Gina Neff, Professor of Responsible AI at Queen Mary University London, told the BBC that restricting access could limit “the development and safe testing of these AI systems,” and she warned it could restrict collaboration with governments around the world.
Sovereignty, Access, and Fallout
The BBC reported that the European Union said the latest development further underlined “Europe's need for technological sovereignty,” after Thomas Regnier said the Commission was assessing the situation.
“Anthropic suspends new AI tools over US government security concerns Anthropic has suspended its powerful new AI model after US authorities raised security concerns just days following its public release”
Time quoted Anton Leicht, a fellow at the Carnegie Endowment for International Peace, saying the shutdown “shows how irrelevant most other countries have become to AI policy,” and he argued that “neither access to foreign markets nor any retaliation options held by any other country factored into the administration's decision.”
Nextgov/FCW said the shutdown would likely complicate near-term plans to test or deploy Anthropic’s most capable cyber-focused systems for federal agencies and critical infrastructure partners, while raising unresolved questions about balancing trusted access with fears of misuse.
CNN reported that the Commerce Department issued the restriction and that it meant “many foreign nationals working for Anthropic will not be able to touch those models,” as Anthropic said it believed the government’s standard would “essentially halt all new model deployments for all frontier model providers.”
