Trump Administration Blocks Foreign Access to Anthropic's Advanced AI Models

The Trump administration has dramatically escalated its conflict with one of the world's leading artificial intelligence companies, placing sweeping export controls on Anthropic's most advanced AI models.[1][2]

In a letter sent Friday by Commerce Secretary Howard Lutnick to Anthropic CEO Dario Amodei, the administration ordered that the company's newly developed Mythos 5 and Fable 5 models be immediately restricted from any location outside the United States.[1][8]

The directive goes further than standard export bans, explicitly blocking access for all foreign governments, overseas companies, and even foreign nationals currently residing within the United States.[1][2]

The move marks a historic shift in how Washington regulates commercial technology, effectively treating cutting-edge artificial intelligence systems as classified national security assets rather than consumer software products.[1][3]

The export controls are the latest and most severe salvo in a bitter, months-long standoff between the White House and Anthropic over how the military is permitted to use artificial intelligence.[4][7]

The rupture began in early 2026 when the Department of Defense demanded unrestricted access to Anthropic's Claude models for "all lawful uses."[7]

Anthropic, a company founded specifically on principles of AI safety, refused the Pentagon's demands. Executives cited their corporate constitution, which strictly prohibits the use of their models for mass domestic surveillance or for autonomous weapons systems that operate without human oversight.[4][7]

The administration's response was swift and unprecedented. In late February, Defense Secretary Pete Hegseth designated Anthropic a "supply chain risk"—a classification typically reserved for foreign adversaries like Huawei or companies linked to the Chinese military.[4][7]

President Trump subsequently ordered all federal agencies to cease using Anthropic technology, effectively killing a $200 million contract the company had secured to operate on classified military networks.[4]

"I fired Anthropic," Trump boasted to reporters in March. "Anthropic is in trouble because I fired them like dogs, because they shouldn't have done that."[4]

Anthropic fought back, filing a high-stakes lawsuit in federal court arguing that the government was unlawfully retaliating against the company and violating its First Amendment rights by punishing it for its safety policies.[5]

In late March, U.S. District Judge Rita F. Lin issued a preliminary injunction temporarily blocking the Pentagon's blacklisting, setting the stage for a protracted legal battle over whether the government or private developers control the ultimate use of AI technology.[5]

However, Friday's export controls bypass the court's injunction entirely by utilizing the Commerce Department's authority over international trade and national security, rather than the Pentagon's procurement rules.[1][2]

The blockade coincides with the issuance of National Security Presidential Memorandum 11, a sweeping new directive signed by President Trump that aims to accelerate the military's adoption of AI.[3][6]

The memo explicitly orders the termination of contracts with any company that attempts to restrict how the government uses its artificial intelligence, while simultaneously revoking Biden-era guardrails that the current administration views as overly restrictive.[3][6]

For the broader tech industry, the administration's actions serve as a stark warning. The government is signaling that AI companies must choose between maintaining strict ethical guardrails and participating in the lucrative federal contracting ecosystem.[3][7]

The requirement to block foreign nationals inside the U.S. from accessing the models presents a massive logistical hurdle. Tech companies rely heavily on foreign talent via H-1B and F-1 visas, and enforcing such a ban will require unprecedented identity verification for commercial software.[1][8]

As Anthropic prepares for a rumored initial public offering, the loss of its entire international customer base for its flagship models threatens to cause billions of dollars in financial damage.[5]

The standoff now moves to two fronts: the ongoing federal lawsuits in California and Washington, D.C., and the immediate scramble by Anthropic to comply with export controls that fundamentally alter the landscape of the global AI race.[5][8]