A federal court has determined that the Trump administration’s decision to blacklist Anthropic, a prominent AI technology firm, was illegal. The ruling dismissed government measures aimed at curbing the company’s operations, which were deemed retaliatory and unconstitutional.

In a decision delivered by Judge Rita Lin in the U.S. District Court for the Northern District of California, it was established that the government’s actions violated the First Amendment. The court found that the administration had unjustly labelled Anthropic as a supply-chain risk to national security after the company refused to retract its restrictions on using its technologies for lethal autonomous warfare and surveillance of American citizens.

Judge Lin stated, “The undisputed record shows that the challenged actions constituted unlawful retaliation in violation of the First Amendment,” as she ruled in favor of key parts of Anthropic’s motion for summary judgment.

National Security Claims Insufficient

The ruling highlighted that former President Donald Trump and Defense Secretary Pete Hegseth had instructed all federal agencies to cease using Anthropic’s products and prohibited defense contractors from engaging with the company, regardless of the nature of their business. Judge Lin ruled that these actions were not only illegal but also lacked substantial justification.

“While the Department of War has the right to choose its AI vendors, the evidence shows that the measures taken against Anthropic were unfounded and illegal,” she remarked, referring to the Trump administration’s preferred designation for the Department of Defense. She emphasized that invoking national security cannot be a blanket justification for punitive actions against dissenting entities.

Judge Lin criticized the government’s rationale for the blacklisting, noting that the justification provided was minimal. The defendants acknowledged that their concern over Anthropic having potential backdoor access to its technologies was unfounded and that, in fact, the company’s AI solutions do not pose any more risk to national security than any other “black box” AI model.