Anthropic Secures Initial Legal Victory Against Pentagon’s Supply Chain Risk Designation
A federal judge in California announced on Thursday that the Trump administration’s classification of Anthropic as a supply chain threat was unlawful.
U.S. District Judge Rita Lin stated that Defense Secretary Pete Hegseth’s designation of Anthropic as a national security risk amounted to “unlawful retaliation” that violated the First Amendment. Additionally, she labeled the decision as “arbitrary and capricious” and noted that Anthropic was denied the due process guaranteed by the Fifth Amendment.
Earlier this year, Hegseth and former President Donald Trump identified Anthropic as a supply-chain threat, instructing all federal agencies, even those not linked to defense, to cease collaboration with the firm known for its Claude AI system.
The conflict arose when Anthropic imposed strict limitations on certain safety protocols that would enable the Pentagon to utilize its models for autonomous weapons and broad surveillance of U.S. citizens. Although the Pentagon asserted that it would only use Anthropic models for lawful activities, it accused the company of potentially attempting to influence how the military could utilize those models, which it had purchased.
In her ruling, Lin indicated that the government’s actions demonstrated a desire to publicly reprimand Anthropic for its boldness in critiquing the government.
She highlighted the inconsistency between the supply chain label and other government activities, such as Hegseth’s plan to invoke the Defense Production Act concerning Anthropic, which would imply that the company was necessary for national security rather than a risk. She also noted the Department of Defense’s ongoing efforts to secure a contract with Anthropic and its collaboration with the company’s new cybersecurity model, Mythos.
Lin further emphasized that it is evident that Anthropic “undisputedly lacks” any backdoor access to its technology after it is provided to the Department of Defense.
“While the Department of War has the right to choose its AI vendor, the evidence clearly shows that the extensive restrictions placed on Anthropic were illegal and unfounded,” Lin remarked. “The mere claim of national security cannot justify punitive actions against critics of the government,” she added.
An Anthropic spokesperson expressed their approval of the court’s decision, stating, “We are pleased that the court ruled this supply chain risk classification was unlawful. Our focus remains on working effectively with the government to leverage AI for national security in a manner that benefits all Americans.”
In March, Anthropic filed two lawsuits against the Department of Defense in California and Washington, D.C., with the D.C. case still ongoing.
Media outlets have attempted to reach the Department of Defense for a statement.
