Judge Blocks Anthropic Security Risk Designation

A federal judge in San Francisco temporarily blocked the Trump administration's designation of AI company Anthropic as a national security supply chain risk on March 26, 2026, halting enforcement of orders that would bar federal agencies and contractors from using the company's Claude AI models.

U.S. District Judge Rita F. Lin issued the preliminary injunction after Anthropic filed suit earlier in the month, alleging the government's actions violated its First Amendment rights, due process and federal procedures. The ruling, which takes effect after a seven-day stay to allow for a possible appeal, restores the status quo and permits continued use of Anthropic's technology in government operations while the case advances. Lin's 43-page order described the Department of Defense's measures as likely retaliatory, noting they followed the company's public refusal to lift ethical restrictions on its AI during contract talks.

The dispute stemmed from negotiations over a potential expansion of Anthropic's existing $200 million contract with the Defense Department, which began in late 2024 through a partnership with Palantir Technologies. By fall 2025, the department sought unrestricted...