SAN FRANCISCO — A federal judge issued a preliminary injunction Thursday temporarily blocking the Trump administration from banning federal agencies from using artificial intelligence technology developed by Anthropic PBC.
U.S. District Judge Rita F. Lin in the Northern District of California paused the Pentagon's designation of Anthropic as a supply-chain risk to national security and halted a presidential directive ordering all federal agencies to cease use of the company's AI tools, including its Claude chatbot. The order was placed on hold for seven days to allow the government time to appeal.
The administration's actions followed a dispute with Anthropic over restrictions the company placed on the use of its technology. The Pentagon had labeled Anthropic a supply-chain risk earlier this month, citing national security concerns after the company sought to limit certain military applications, including fully autonomous weapons and surveillance of Americans. President Donald Trump then directed federal agencies to immediately stop using Anthropic's products, with some officials referencing a six-month phase-out period for existing contracts.
Anthropic sued the Defense Department and other agencies, arguing the ban could result in billions of dollars in lost revenue and that the designation was punitive. During a hearing Tuesday, Judge Lin described the government's rationale as troubling and questioned whether the measures appeared tailored to stated national security needs. She noted that the supply-chain risk label is typically reserved for adversaries.
The decision affects government contracts involving Anthropic's AI models, which have been used across various federal agencies for tasks ranging from data analysis to other applications. The company, based in San Francisco, has faced the prospect of being removed from government procurement lists, including through the General Services Administration.
The Trump administration has vowed to defend the ban in court, emphasizing the importance of trust in relationships with vendors providing services to the military. Officials have argued that Anthropic attempted to dictate policies on AI use, which they said undermined national security priorities.
As of Friday, the preliminary injunction remains in effect pending any appeal. The broader lawsuit challenging the supply-chain risk designation and related directives continues in federal court in San Francisco. Both sides are expected to file further arguments as the case proceeds.
