US Court Halts Pentagon's Exclusion of AI Firm Anthropic from Military Contracts

A US court has delivered a significant victory to artificial intelligence firm Anthropic, halting the Department of Defense's attempt to exclude the company from military contracts. The ruling, issued by Judge Rita Lin, declared the Pentagon's classification of Anthropic as a national security risk to be illegal and baseless.

"The mere reference to national security is not a free pass to punish government critics and retaliate against them," the judge wrote in her decision. The Department of Defense, under Secretary Pete Hegseth, had previously labeled Anthropic a supply chain risk. This action followed the company's refusal to make its AI model available for US surveillance purposes or for use in autonomous weapons systems.

Anthropic has argued that AI models are not sufficiently reliable for autonomous weapons and that domestic surveillance violates civil liberties. The company filed a lawsuit in March, alleging that the government violated its rights to free speech and due process. Anthropic stated that the exclusion could cost the company billions of dollars.

The Department of Justice, according to court documents obtained by Reuters, countered that Anthropic's refusal could endanger military systems during ongoing operations. The Pentagon also emphasized that private companies should not restrict military actions, asserting that the classification was based on the refusal of contract terms, not on Anthropic's views on AI safety. Anthropic is pursuing a separate legal challenge in Washington against a similar exclusion from civilian government contracts.