U.S. District Judge Rita F. Lin ruled that the Defense Department's designation of AI company Anthropic as a national security supply-chain risk was unlawful retaliation that violates the First Amendment and fails adequate due process under the Fifth Amendment. The court ordered all directives against the company rescinded, dealing a significant setback to the Pentagon's use of procurement blacklists to pressure AI developers.
The Dispute
The Pentagon blacklisted Anthropic after the company declined to allow its AI models to be used for mass surveillance and autonomous weapons systems. In her ruling, Judge Lin criticized the "empty invocation of national security" as a "blank check" to punish government critics, and noted that the government continued using Anthropic's technology even as it pursued the designation.
First Amendment Grounds
The court found that the designation functioned as retaliation for protected speech and association, crossing the line from procurement policy into punishment of a contractor's policy positions. The ruling does not force the Pentagon to buy Anthropic products; it invalidates the punitive designation and restores the company's ability to compete for federal business on equal terms.
What Comes Next
A second, related case against Anthropic remains pending in Washington, D.C., and the government could appeal this decision. The ruling arrives amid a broader policy debate over how the U.S. military should procure AI while respecting the rights and policy choices of frontier model developers, and it may reshape how agencies use blacklist authorities to pressure contractors over the use of their technology.
