Anthropic faces fallout across federal agencies from DOD clash

The high-stakes dispute between Anthropic and the U.S. military led to a sweeping decision Friday by President Donald Trump to remove the AI startup’s technology from all federal agencies. 

Already, several agencies are taking action. The General Services Administration, Department of State, and Department of Health and Human Services immediately indicated in public statements, comments, or internal emails that they were moving to boot Anthropic. The fallout is sure to continue as agencies untangle the Claude maker from their workflows.

The clash centered on the Defense Department wanting Anthropic to remove stipulations that limited the military’s use of the startup’s technology in real-world operations, DefenseScoop previously reported. Anthropic CEO Dario Amodei said in a statement Thursday that the company could not accede to the request “in good conscience.”

In alignment with Trump’s announcement, the GSA said it is removing Anthropic from the sandbox testing platform USAi.gov and its Multiple Award Schedule procurement opportunities. 

“GSA stands with the President in rejecting attempts to politicize work dedicated to America’s national security,” GSA Administrator Edward Forst said in a statement Friday. “Building resilient, secure, and scalable AI solutions demands alignment, trust, and a willingness to make hard calls.”

Similarly, a State Department spokesperson told DefenseScoop that the agency was “taking immediate steps to implement the directive and bring our programs into full compliance.” 

At HHS, Deputy Chief AI Officer Arman Sharma emailed staff that the agency would be “disabling enterprise Claude” as a result of the directive. That email, which was obtained by FedScoop, also stated that user history would be preserved and a communication to all staff was expected the following week. 

Notably, HHS rolled out Claude to all of its staff in December as part of the GSA’s OneGov strategy. Under that deal, Anthropic offered its platform to the federal government at a deeply discounted rate of $1 per agency for a year.

FedScoop reached out to HHS, the White House, and nearly 20 other federal agencies Friday night for further details on plans, but did not receive any immediate responses. Anthropic also did not respond immediately to a request for comment prior to publication. 

Anthropic’s technology is present across federal agencies, according to the latest AI inventories. Workers in the Department of Commerce are deploying Anthropic’s Claude for Government to generate analytical reports, data visualizations and documentation for analysis workflows. The Energy Department’s Idaho National Laboratory is piloting Claude to assist with coding, among other tasks. Anthropic is also one of the vendors powering document summarization and content generation for the Department of Homeland Security’s Customs and Border Protection. 

In a social media post Friday, Trump said he is directing every federal agency to “immediately cease” all use of the AI startup’s technology. For agencies that are using the technology at various levels, there will be a six-month phase-out period, he added. 

Secretary of Defense Pete Hegseth said that in conjunction with the president’s directive, he is directing the agency to designate Anthropic as a supply-chain risk to national security. 

“Effective immediately, no contractor, supplier, or partner that does business with the United States military may conduct any commercial activity with Anthropic,” Hegseth said in a post to X. “This decision is final.”

Some advocacy groups expressed concern over the White House’s regulatory actions toward Anthropic. 

“It chills private companies’ ability to engage frankly with the government about appropriate uses of their technology, which is especially important in national security settings that so often have reduced public visibility,” Alexandra Givens, CEO and president of the Center for Democracy and Technology, said in an emailed statement. “Retaliating against a company for setting tailored, principled conditions on its product’s use undermines basic market freedoms and makes us all less safe.”

Givens said the presidential directive Friday set a “dangerous precedent,” a sentiment shared by 50% of the nearly 2,000 U.S. adults surveyed by Morning Consult earlier this week. 

Sen. Elizabeth Warren, D-Mass., called for Hegseth to testify before lawmakers regarding the situation. 

“Is the Trump administration punishing Anthropic because it’s refusing to help mass surveil American communities or build killer robots?” Warren said Friday in a post to X. “The American people deserve to know what Trump officials are planning at the Pentagon.”

DefenseScoop reporter Brandi Vincent contributed reporting.

This story is developing and will be updated if new information becomes available.