A federal judge has ruled that the Pentagon acted illegally and without basis when it designated artificial intelligence company Anthropic as a supply chain risk earlier this year. U.S. District Judge Rita Lin issued an order Thursday night stating that the government's actions were driven by a desire to penalize Anthropic for criticizing the Department of Defense's views on AI use, rather than on any evidence of actual sabotage risk.
The dispute began in February when the government accused Anthropic of endangering national security. Anthropic CEO Dario Amodei had expressed concerns that the company's products could be used for mass surveillance or autonomous armed drones, and refused to allow unrestricted military use of its technology. Anthropic subsequently sued the Pentagon in March, alleging an "unlawful campaign of retaliation."
In her 59-page ruling, Lin, an appointee of former President Joe Biden, stated that neither the Constitution nor the federal statute invoked by the government permitted "sweeping penalties based principally on Anthropic’s critique of the Administration’s views." The government is expected to contest the ruling.
An Anthropic spokesperson said the company welcomes the judge's decision and remains focused on working with the government to leverage AI for national security. The White House did not immediately respond to a request for comment.
The legal challenge has amplified a public debate about the deployment of AI in warfare and surveillance. Notably, OpenAI, Anthropic's rival and maker of ChatGPT, finalized its own agreement with the Pentagon just hours after the government's punitive actions against Anthropic.
Earlier in the legal proceedings, Judge Lin had temporarily blocked the Pentagon from labeling Anthropic a supply chain risk and halted the enforcement of a directive ordering federal agencies to cease using Anthropic and its chatbot, Claude. During a hearing on July 30, Lin expressed concern that the government's position appeared to contradict the First Amendment and stated that the evidence had increasingly weakened the government's case.
Lawyers for the Department of Justice argued during that hearing that the vast and opaque nature of AI models prevents the Defense Department from evaluating them similarly to physical hardware. Anthropic's legal team countered that the government's actions inflicted significant harm and broadly threatened to stifle speech and debate on critical issues.
Anthropic is also pursuing a separate, narrower case in the federal appeals court in Washington, D.C., concerning a different Pentagon rule used to designate the company as a supply chain risk.