Judge strikes down Trump admin sanctions against Anthropic

A federal judge has ruled that the Trump administration illegally punished AI company Anthropic after it refused to let the Pentagon use its Claude chatbot for autonomous weapons and domestic surveillance.

Judge Rita Lin blocked federal agencies from enforcing the administration’s restrictions and overturned Defense Secretary Pete Hegseth’s decision to label Anthropic a “supply chain risk.” 

Lin said national security cannot justify retaliation against government critics.

The dispute began in February after Anthropic refused to allow its Claude AI models to support autonomous weapons or domestic surveillance.

The company argued that current AI systems are not reliable enough for such sensitive uses and that domestic surveillance could violate rights.

The Pentagon ended its $200 million defense contract with Anthropic and embraced rival OpenAI.

Anthropic sued, claiming the government retaliated against its AI safety position and violated its First Amendment rights. The ruling makes a March suspension of the sanctions permanent. The government can appeal.