Trump Moves to Cut Anthropic From Federal Agency Work

President Donald Trump said he is moving to ban federal agencies from working with the artificial intelligence company Anthropic after a weeks-long fight between the company and the Pentagon.

Trump is giving federal agencies that currently use Anthropic’s AI tools six months to stop using their models and choose a different provider. This announcement comes after Dario Amodei, Anthropic’s CEO, declined to give the Department of Defense unfettered access to his company’s AI models due to concerns about software being used for mass surveillance and fully automated weapons.

“The Leftwing nut jobs at Anthropic have made a DISASTROUS MISTAKE trying to STRONG-ARM the Department of War,” Trump wrote in a Truth Social post. “I am directing EVERY Federal Agency in the United States Government to IMMEDIATELY CEASE all use of Anthropic’s technology.”

“We don’t need it, we don’t want it, and will not do business with them again! There will be a Six Month phase out period for Agencies like the Department of War who are using Anthropic’s products, at various levels. Anthropic better get their act together,” he added.

Trump and congressional Republicans have been eager to integrate AI technology into federal agencies. Defense Secretary Pete Hegseth has been one of the loudest advocates for adopting AI for military work — even using Anthropic’s AI models to aid in the capture of Venezuelan President Nicolás Maduro earlier this year.

Anthropic and the Pentagon have been locked in tense negotiations after the company tried to create a carveout for certain uses for its technology, like the military using AI for mass surveillance or having completely autonomous weapons be able to kill without a human in the loop.

Other AI companies working with the Pentagon allow the military to use their software for “any lawful use,” Axios reported. Anthropic’s AI models are, so far, the only models allowed to access classified information due to the reliability of its software.

The Pentagon has been pursuing a deal with Anthropic because of that reliability. Dave Lawler at Axios reported that despite Trump’s post, a deal with Anthropic is still on the table.

Amodei said in a Thursday statement that even the most advanced AI models right now cannot be trusted to perform certain high-stake tasks.

“We believe AI can undermine, rather than defend, democratic values. Some uses are also simply outside the bounds of what today’s technology can safely and reliably do,” Amodei wrote.

For Republicans, Amodei’s refusal to accede to the Pentagon’s conditions represents an attempt to impose private safety commitments over the Pentagon’s priorities.

“If Anthropic has concerns about how the Department of Defense will use its AI technology for warfighting, then it has the right not to do business. It is good to have this debate because it is all very new,” Rep. Don Bacon told NOTUS in a statement.

Anthropic did not immediately respond to a request for comment.

A six-month period to phase out all Anthropic software is less onerous than the more extreme measures the Trump administration reportedly had considered. But even after Trump’s post, Hegseth declared Anthropic a supply-chain risk. That would bar the company from working with the Pentagon, or any other defense contractor and is a measure that has traditionally been used against foreign companies suspected of espionage.

“Effective immediately, no contractor, supplier, or partner that does business with the United States military may conduct any commercial activity with Anthropic. Anthropic will continue to provide the Department of War its services for a period of no more than six months to allow for a seamless transition to a better and more patriotic service,” Hegseth wrote in a post on X.

Anthropic’s AI models are widely popular across the technology industry and financial corporations. Dean Ball, a former Trump administration adviser and author of Trump’s AI Action Plan, criticized the move as highly disruptive.

“Nvidia, Amazon, Google will have to divest from Anthropic. This is simply corporate murder. That’s what you’re watching. I could not possibly recommend investing in American AI to any investor; I could not possibly recommend starting an AI company in the United States,” Ball posted on X.

The administration also considered the legally untested approach of invoking the Defense Production Act — a law that gives the president authority to force companies to accept or prioritize certain government contracts — to force Anthropic to provide an unrestricted version of their model.

While Trump has been a reliable ally to the tech industry since returning to office, many on the right like Elon Musk and Katie Miller, who are both part of Anthropic competitor xAI, alleged that Anthropic’s leadership supports left-wing ideologies.

For Joel Thayer, a Republican lawyer and senior AI fellow at the America First Policy Institute, this issue goes beyond ideology into a question of whether tech giants should be able to steer defense policy.

“Outside of whether something is woke or not woke, I think what it comes down to is a question of who is in control,” Thayer told NOTUS.

“If we’re going to entrust a select group of companies with this incredibly powerful tool that will have all sorts of applications,” he said, “what does this concentration of power in these markets actually involve? And how much control should they have, especially if you’re going to be a government contractor?”

Morgan Plummer, vice president for policy design at Americans for Responsible Innovation, an advocacy group that supports regulating AI, said the relationship between Anthropic and the administration is an example of how fraught the negotiations between the government and technology companies have become.

“Frustratingly, the President’s directive to cease use of all Anthropic products an hour before the established 5:01 deadline seems to be yet another bad faith action on the part of the government in these strained negotiations,” Plummer wrote in a statement to NOTUS minutes after Trump’s post.

“The president’s tweet seems to reinforce the ways in which the situation became personal for some of the parties involved rather than keeping the talks focused on maintaining access to key capabilities that our warfighters clearly need. This is more evidence of the vast intellectual and philosophical gulf that still very much exists between DC and the Valley,” he added.

—
Editor’s Note: This story has been updated with additional reporting.