Anthropic CEO Dario Amodei on Thursday said the company “cannot in good conscience” agree to allow the Department of Defense to use its models in all lawful use cases without limitation, adding that the agency’s threats do not change its position.
The artificial intelligence startup has been engaged in tense negotiations with the Pentagon in recent weeks, and it said Thursday that those discussions are still ongoing. Defense Secretary Pete Hegseth has threatened to label Anthropic a “supply chain risk” or to invoke the Defense Production Act to force the company to comply with its demands.
Anthropic wants assurance that its models will not be used for fully autonomous weapons or mass domestic surveillance of Americans, while the DoD wants to be able to use the models without those restrictions.
“It is the Department’s prerogative to select contractors most aligned with their vision,” Amodei wrote in a statement. “But given the substantial value that Anthropic’s technology provides to our armed forces, we hope they reconsider.”
Hegseth met with Amodei at the Pentagon on Tuesday, and he gave Anthropic until Friday evening to agree to his agency’s demands. The DoD sent Anthropic its “last and final offer” on Wednesday night, according to a senior Pentagon official.
Chief Pentagon Spokesman Sean Parnell said Thursday that the DoD has “no interest” in using Anthropic’s models for fully autonomous weapons or to conduct mass surveillance of Americans, which he noted is illegal. He emphasized that the agency wants the company to agree to allow its models to be used for “all lawful purposes.”
“This is a simple, common-sense request that will prevent Anthropic from jeopardizing critical military operations and potentially putting our warfighters at risk,” Parnell wrote in a post on X on Thursday. “We will not let ANY company dictate the terms regarding how we make operational decisions.”
Anthropic signed a $200 million contract with the DoD in July, and it was the first lab to integrate its models into mission workflows on classified networks.
The startup’s rivals OpenAI, Google and xAI were also granted contract awards of up to $200 million from the DOD last year. Those companies have agreed to let the DoD use their models for all lawful purposes within the military’s unclassified systems, though xAI also agreed to allow its models to be used in classified settings this week.
“Our strong preference is to continue to serve the Department and our warfighters—with our two requested safeguards in place,” Amodei said. “Should the Department choose to offboard Anthropic, we will work to enable a smooth transition to another provider, avoiding any disruption to ongoing military planning, operations, or other critical missions.”
WATCH: Anthropic’s AI safety contradiction
