The U.S. Commerce Department ordered Anthropic to suspend foreign nationals from accessing its latest AI models, forcing the company to disable Fable 5 and Mythos 5 for all users. This marks the first time a federal directive has pulled a publicly deployed AI system offline, raising immediate questions about how national security concerns will dictate commercial technology availability.
The directive and the response
Anthropic received the export control order on Friday evening. The company stated it had to disable the models globally to ensure compliance, as separating foreign and domestic users was not feasible under the mandate. "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 company said in a statement.
The Commerce Department directed the action over a specific technique that could allow users to bypass narrow safeguards, known as a jailbreak. Anthropic noted the government provided only verbal evidence of this vulnerability and argued the workaround likely exists in other companies' systems.
A clash over safety and process
Anthropic pushed back against the sudden suspension. The company argued that a narrow potential jailbreak does not justify recalling a commercial model deployed to hundreds of millions of people. "We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people," the company said.
The models were released just days prior. Fable 5 includes strict restrictions on cybersecurity and biology queries for the general public. Mythos 5 was distributed without those safeguards to a select group of trusted partners, including key infrastructure firms.
A history of regulatory friction
This suspension follows a contentious relationship between Anthropic and the federal government. In February, President Donald Trump and Defense Secretary Pete Hegseth barred the company's products from federal agencies after Anthropic requested stronger guardrails on Pentagon use of its technology.
Anthropic subsequently sued the administration over free speech violations. A California federal judge ruled in the company's favor, though the case remains active in Washington, D.C.
Despite this history, federal agencies recently began integrating Anthropic's technology. Reports indicate the National Security Agency has been using Mythos to conduct offensive cyber operations. Furthermore, a June 2 executive order directed federal agencies to design a mechanism for gaining early, voluntary access to the most advanced commercial AI models.
Why this matters for government professionals
Federal employees and AI for Government stakeholders must now account for the reality that commercial AI availability can be abruptly altered by executive action. This incident establishes a precedent for direct federal intervention in commercial deployments based on unverified technical claims.
As agencies design mechanisms for early access to advanced commercial models under the recent executive order, this suspension highlights the friction between swift deployment and national security mandates. Professionals overseeing these initiatives must understand the evolving boundaries of public-private AI deployment, making structured AI Learning Path for Policy Makers resources increasingly critical for understanding these regulatory shifts.
Your membership also unlocks:
