Amid an escalating fight over how the Pentagon wants to use artificial intelligence, Anthropic’s chief executive said the company cannot accept the Defense Department’s latest demands for broader, fewer-constraints access to its technology. Dario Amodei said Thursday that Anthropic “cannot in good conscience accede” to the Pentagon’s requests, deepening a public clash that is threatening to pull Anthropic’s contract and trigger other steps by Friday.
The dispute has sharpened after Defense Secretary Pete Hegseth met with Amodei and delivered an ultimatum Tuesday, according to the report. The terms at issue centered on new contract language from the Defense Department that Amodei described as pushing for unrestricted use, with military officials warning that they could go further if the parties do not reach agreement—up to taking actions tied to supply-chain risk or invoking a Cold War-era law known as the Defense Production Act.
Anthropic responded by saying it is not walking away from negotiations, but that the new contract language produced “made virtually no progress on preventing Claude’s use for mass surveillance of Americans or in fully autonomous weapons.” The company’s stated position is that its own policies prevent its models from being used for those purposes. It also said the Pentagon’s contract dispute is becoming more pressing because Anthropic is the last of its AI peers under government deals to not supply its technology to a new U.S. military internal network.
Sean Parnell, the Pentagon’s top spokesman, said earlier on social media that the military has “no interest in using AI to conduct mass surveillance of Americans (which is illegal)” and no interest in developing autonomous weapons that operate without human involvement. In his account, the Pentagon wants to use Anthropic’s model for “all lawful purposes,” while opening up the technology in a way that Pentagon officials say would avoid “jeopardizing critical military operations.” Parnell also said the government would not allow any company to dictate terms for how operational decisions are made.
Amodei, in his statement, framed the ultimatum and related threats as a contradiction. He said Thursday that the later threats were “inherently contradictory: one labels us a security risk; the other labels Claude as essential to national security.” He also wrote that the Pentagon has discretion in choosing contractors aligned with its vision, adding that Anthropic hope the department will reconsider given what it said its technology provides to the armed forces.
The public disagreement spilled into direct comments by lawmakers. Sen. Thom Tillis, a North Carolina Republican who is not seeking reelection, said the Pentagon has been handling the matter unprofessionally while Anthropic is “trying to do their best to help us from ourselves.” Tillis told reporters, “Why in the hell are we having this discussion in public? This is not the way you deal with a strategic vendor that has contracts.” He added that when a company resists a market opportunity out of fear of negative consequences, lawmakers should listen to it and then “behind closed doors figure out what they’re really trying to solve.”
Sen. Mark Warner, the ranking Democrat on the Senate Intelligence Committee, said he was “deeply disturbed” by reports that the Pentagon is “working to bully a leading U.S. company.” Warner said in a statement that the dispute “further underscores the need for Congress to enact strong, binding AI governance mechanisms for national security contexts.”
Beyond the contract fight itself, the report also pointed to changes in legal oversight within the military. It said Hegseth told Fox News last February that the department ultimately wants lawyers who provide “sound constitutional advice” and do not serve as “roadblocks.” The report said Hegseth also fired top lawyers for the Army and Air Force without explanation and that the Navy’s top lawyer resigned shortly after the election in late 2024.