On Thursday (local time), a US judge reversed the Pentagon's decision to blacklist the AI firm Anthropic. Earlier, Defense Secretary Hegseth had added Anthropic to the blacklist, citing "national security supply chain risks." This action came after Anthropic declined the military's request to utilize its AI chatbot, Claude, for surveillance or autonomous weapons systems, leading to its exclusion from specific military contracts. Anthropic contended that this blacklisting would result in billions of dollars in losses and tarnish its reputation. The company maintains that current AI models are ill-suited for autonomous weapons and is firmly against surveillance activities. Conversely, the Pentagon asserts that private companies should not meddle in military affairs.

In reaction, Anthropic initiated legal proceedings, accusing the government of overstepping its bounds in retaliation. The company claimed that its rights to free speech and due process were violated and that the Pentagon's decision lacked factual justification. Ultimately, the judge ruled that the Pentagon's action was both illegal and baseless, highlighting that national security cannot be invoked as a pretext to retaliate against those who criticize the government. This case represents the first instance of a US company being publicly designated as a supply chain risk under a little-known government procurement rule. The US Department of Justice had previously argued that imposing restrictions on Anthropic could introduce uncertainty and potentially paralyze military operations.