A federal district court in California has ruled that the Pentagon's designation of artificial intelligence developer Anthropic as a national security supply chain risk was unlawful and unconstitutional. In a 59-page decision, U.S. District Judge Rita Lin determined that the Department of Defense retaliated against the company in violation of the First Amendment and the Fifth Amendment's Due Process Clause.
The permanent injunction prohibits federal agencies named in the complaint from enforcing the supply chain restriction, which had barred defense contractors and procurement offices from deploying Anthropic's Claude models.
Origins of the Procurement Dispute
The conflict between Anthropic and defense leadership originated during contract negotiations over deployment boundaries for frontier language models. Anthropic insisted on contractual safeguards prohibiting the use of its technology in fully autonomous lethal weapon systems and mass domestic surveillance. In contrast, Pentagon officials, led by Defense Secretary Pete Hegseth, required unrestricted access to the models for all lawful military applications.
Following the impasse in negotiations, the Department of Defense designated Anthropic as a supply chain risk. The designation, historically applied to foreign adversaries and compromised hardware vendors, canceled Anthropic's active government contracts and prohibited third-party defense suppliers from incorporating Anthropic API endpoints into defense software pipelines.
Judicial Findings on First Amendment and Due Process
In the decision, Judge Lin rejected the government's position that statements made during procurement discussions and public statements on AI safety were exempt from First Amendment protection.
Key findings from the court include:
- Pretextual Justification: The court held that the Department of Defense used the supply chain risk classification as a pretext to penalize Anthropic for its public positions on safety guardrails and critical commentary regarding defense AI deployment policies.
- Constitutional Limits on Executive Discretion: While acknowledging the government's broad discretion in selecting commercial contractors, the court ruled that statutory national security authorities do not grant the executive branch power to retaliate against companies for exercising constitutionally protected speech.
- Due Process Violations: The administrative process leading to the supply chain blacklist failed to provide adequate procedural safeguards, notice, or an evidentiary record demonstrating technical or operational security risks.
Industry Implications and Parallel Litigation
The ruling resolves Anthropic's district court challenge in California, though broader legal disputes over federal AI procurement authority continue. A related case addressing separate procurement restrictions remains pending before the U.S. Court of Appeals for the District of Columbia Circuit.
The decision establishes an important legal precedent for commercial AI developers negotiating acceptable-use policies with government agencies. It clarifies that federal procurement authorities cannot leverage national security supply chain designations to bypass standard contracting procedures or punish vendors over policy disputes regarding AI safety constraints.
