Security

Anthropic’s Last Stand: Fighting to Salvage Pentagon Deal

Published

7 hours ago

March 5, 2026

Anthropic makes last-ditch effort to salvage deal with Pentagon after blowup

Anthropic CEO Negotiating with Department of Defense to Salvage Relationship

Reports indicate that Anthropic CEO Dario Amodei is back at the negotiating table with the Department of Defense in a bid to save the company’s ties with the US military. This move comes as efforts are made to prevent the company from being excluded from defense projects due to concerns about being a “supply chain risk.” Tensions escalated between the two parties last Friday following a prolonged public dispute over Anthropic’s reluctance to grant the Pentagon unrestricted access to its AI technology. Competitors like OpenAI quickly stepped in to fill the void.

Discussions with Defense Under-Secretary for Research and Engineering

Amodei is reportedly engaged in discussions with under-secretary of defense for research and engineering Emil Michael regarding a potential new contract. This contract would enable the US military to continue utilizing Anthropic’s Claude AI models, as per sources cited by the Financial Times. The situation had escalated when Michael publicly criticized Amodei, accusing him of endangering national security by withholding AI access from the military.

Stakes High for Anthropic Amid Supply Chain Risk Designation

The outcome of the negotiations holds significant importance for the US startup. Secretary of Defense Pete Hegseth announced on Friday his intent to designate Anthropic as a supply chain risk, a classification usually reserved for entities with associations that pose security threats to the United States. Such a designation could have far-reaching consequences within the US tech sector, leading firms to sever ties with Anthropic to maintain defense contract eligibility.

Leaked Memo Exposes Tensions with Trump Administration

A leaked memo from Amodei to Anthropic staff, recently disclosed by The Information and FT, is expected to further strain relations between the company and the Trump administration. The memo criticizes OpenAI’s collaboration with the Pentagon as mere “safety theater” and accuses both parties of propagating falsehoods. Amodei suggests that Anthropic’s unfavorable standing with the government is linked to a lack of political support, contrasting the cozy relationship between OpenAI executives and the Trump administration.

Defense Department’s Demands and Anthropic’s Principles

The conflict between Anthropic and the Pentagon revolves around the Defense Department’s insistence on unrestricted access to the company’s technology. Anthropic, however, stands firm on its conditions for military use, specifically prohibiting mass surveillance of Americans and lethal autonomous weapons. Hegseth advocates for the availability of AI technology for “any lawful use,” a stance Anthropic rejects due to concerns about potential ethical violations. In contrast, xAI and OpenAI have reportedly agreed to the Defense Department’s terms.