US judge rules Pentagon unlawfully targeted AI firm Anthropic over weapons refusal

A federal judge has ruled that the US Department of Defense acted unlawfully by designating artificial intelligence startup Anthropic as a supply chain risk. Judge Rita Lin condemned the move as "illegal and baseless", noting it is the first time an American company has faced such a public designation.

The legal dispute originated after the Claude-maker refused to let the military utilise its AI models for autonomous weapons or domestic surveillance. Anthropic argued that Defense Secretary Pete Hegeth overstepped his authority by invoking legislation designed to shield military systems from foreign sabotage.

Judge Lin rejected the government’s justification, stating that citing national security "is not a blank check to punish and retaliate against government critics". During an earlier phase of the lawsuit, she observed that the government appeared to be trying to "cripple" the company and "chill public debate" over military AI use, adding that this looked like "classic First Amendment retaliation".

The Pentagon had previously argued that private enterprises should not be permitted to constrain military action. The White House escalated the rhetoric, previously labeling Anthropic "a radical left, woke company" attempting to control military activity, and insisting the armed forces are beholden to the US Constitution, "not any woke AI company's terms of service".

Anthropic welcomed the decision, with a spokesperson stating the company remains focused on "working productively with the government to harness AI for our national security so all Americans benefit from this technology". The firm had maintained its models were not reliable enough for safe deployment in weapons systems and opposed domestic surveillance as a violation of rights.

For European markets and technology policymakers, this ruling establishes a vital legal boundary for the AI sector. It demonstrates that tech companies can successfully resist military contracts on safety and ethical grounds without facing retaliatory supply-chain blacklisting that impacts their business and freedom of speech.

As the European Union continues to refine its own AI governance frameworks, this US precedent underscores the growing friction between national security ambitions and corporate safety mandates. Investors and tech firms on both sides of the Atlantic will closely monitor how governments attempt to enforce compliance in the rapidly evolving defence AI sector.