Anthropic has initiated legal action in federal court in California, asking a judge to set aside a Pentagon designation that labels the company a national security supply-chain risk and to bar federal agencies from enforcing that classification.
The complaint, filed on Monday, contends the designation was unlawful and infringes on Anthropic’s free speech and due process rights. The company argues the government is effectively punishing protected corporate speech and has asked the court to undo the formal classification.
According to the filing, the Department of Defense moved to place Anthropic on the national security blacklist last Thursday, a step that restricts how the company’s technology can be used by federal entities. A source cited in related reporting said the technology had been used in military operations in Iran, a point that figures in the Pentagon’s decision to limit U.S. government use.
Defense Secretary Pete Hegseth is reported to have designated Anthropic a national security supply-chain risk last week, a move that followed the company's refusal to remove guardrails that prevent its AI from being applied to autonomous weapons or domestic surveillance. Anthropic says those protective measures reflect its policies, and the firm asserts the designation runs afoul of constitutional limits on government authority.
In its court filing Anthropic stated: "These actions are unprecedented and unlawful. The Constitution does not allow the government to wield its enormous power to punish a company for its protected speech." The company seeks injunctive relief to stop enforcement actions and to have the designation vacated.
The designation has implications for Anthropic’s ability to do business with the federal government. The company has warned that the classification poses a direct threat to its government contracts and engagements. The case may also set a precedent for how other AI companies negotiate the terms under which their technology may or may not be used by the military.
Anthropic’s CEO, Dario Amodei, emphasized that the designation covers a limited area of use. On Thursday he clarified the ruling had "a narrow scope" and that businesses could continue to use Anthropic’s tools in projects not associated with the Pentagon.
The lawsuit frames the dispute as both a legal challenge to a specific administrative action and a test of the bounds on government power to restrict commercial use of AI technology when constitutional protections are invoked.