President Donald Trump said Friday he has ordered all federal agencies to phase out the use of Anthropic technology following the company’s public clash with the Pentagon over the military use of artificial intelligence.
Trump’s announcement came just over an hour before the Defense Department’s deadline for Anthropic to allow unrestricted military use of its AI systems or face penalties, and nearly a day after Anthropic CEO Dario Amodei said the company “cannot in good conscience accede” to the Pentagon’s demands.
Under the directive, most federal agencies must immediately stop using Anthropic technology, while the Pentagon has been given six months to remove the AI tools already embedded in military platforms.
“We don’t need it, we don’t want it, and will not do business with them again,” Trump said in a post on social media. Anthropic did not immediately respond to requests for comment.
The dispute centers on how artificial intelligence should be used in national security, particularly in high-stakes environments involving lethal force, classified information and surveillance. Anthropic has sought assurances that its chatbot, Claude, would not be used for mass surveillance of Americans or in fully autonomous weapons systems. The company said revised contract language would have allowed those safeguards to be ignored.
The Pentagon has denied it intends to deploy AI for illegal surveillance or weapons without human oversight. Defense officials said they want access to Anthropic’s model for “all lawful purposes” and warned the company it could be designated a supply-chain risk if it refused to comply.
Trump’s decision is expected to benefit rival systems, including Grok from Elon Musk, which the Pentagon plans to integrate into classified military networks. It also sends a warning signal to competitors OpenAI and Google, both of which hold military AI contracts.
The conflict has deepened divisions within Silicon Valley. Some employees at OpenAI and Google issued an open letter supporting Anthropic’s position, warning that the Pentagon is attempting to pressure companies into abandoning safety commitments.
In a surprise move, OpenAI CEO Sam Altman publicly backed Anthropic, questioning what he called the Pentagon’s “threatening” posture and saying most AI developers share similar limits on how their technology should be used in warfare.
Defense Secretary Pete Hegseth had warned that Anthropic could face contract termination or even the use of the Defense Production Act to compel access to its technology. Amodei said those threats were contradictory, arguing that the company could not be labeled both a security risk and essential to national defense.
Former defense officials and lawmakers from both parties raised concerns about the government’s approach. Retired Air Force General Jack Shanahan said Anthropic’s safeguards were “reasonable” and warned that current AI systems are not ready for fully autonomous national security roles.
Anthropic said it would cooperate in transitioning the Pentagon to another AI provider if necessary, but maintained that its refusal was rooted in preventing misuse of powerful technology.
