Anthropic sues Pentagon over Claude AI use ban

Anthropic, a leading artificial intelligence firm, has initiated legal action against the U.S. Department of Defense (DoD) in San Francisco, challenging a Pentagon-wide directive that prohibits the department and its contractors from utilizing the company's technology.

 

The lawsuit stems from Anthropic's refusal to remove safety constraints embedded within its Claude AI model, a decision the company asserts led directly to the government's restrictive measure.

 

DoD Designates Anthropic as Supply Chain Risk

 

The Department of Defense formally designated Anthropic as a "national security supply chain risk," effectively blocking all DoD offices and vendors from deploying its AI tools. Anthropic states that Defense Secretary Pete Hegseth issued this designation on March 3, with the subsequent ban taking effect last month.

 

Anthropic's Legal Challenge

 

Filed on March 9, Anthropic's complaint argues that the government's actions infringe upon the company's constitutional rights, specifically its protections for free expression and due process. The company characterizes the administration's move as unprecedented and lacking a legal basis.

 

Safety Dispute at Core of Conflict

 

Anthropic maintains it declined to permit what it describes as the unrestricted military use of its AI for fully autonomous weapons systems and mass domestic surveillance. The company contends that its advocacy for guardrails around advanced AI systems constitutes protected speech, and the Pentagon's response amounts to punishment for this stance.

 

Government Disputes Anthropic's Framing

 

The White House has disputed Anthropic's interpretation, arguing in a recent filing that the conflict arises from contract negotiations and national security considerations, rather than retaliation for free speech. U.S. District Judge Rita Lin is presiding over the proceedings and will determine whether the Pentagon's restriction can be temporarily halted while the case progresses.

 

Scrutiny Over Potential Overreach

 

The dispute has drawn concern from legal specialists and lawmakers regarding potential government overreach and efforts to pressure U.S. companies to compromise AI safety practices. Senator Elizabeth Warren (D) has reportedly raised questions concerning the government's approach in this matter.

 

Authority of Designation Questioned

 

Legal analysts have highlighted a social media post from February 27, where Secretary Hegseth directed the Defense Department to apply the supply-chain risk label. Experts suggest this directive may exceed statutory authority. The legality of the designation process appears central to the case, though court records do not yet provide a resolution.