The U.S. Court of Appeals for the D.C. Circuit on Sept. 25 upheld the Pentagon’s designation of Anthropic as a supply chain risk, allowing the Defense Department (DOD) to continue excluding the company’s artificial intelligence (AI) models from its systems and defense contract work.
The 2-1 ruling rejected Anthropic’s challenge under the Federal Acquisition Supply Chain Security Act (FASCSA) 2018, which gives the government authority to address supply chain risks in federal procurement. The court found that DOD had “ample support” for concluding that continued use of Claude by the department or its contractors presented a national security risk.
The dispute stemmed from the Pentagon’s push to use Claude for “all lawful uses.” Anthropic objected to two applications: mass surveillance and fully autonomous weapons.
After negotiations between Anthropic and the Pentagon broke down, Defense Secretary Pete Hegseth directed the department to designate the company a supply chain risk. Anthropic challenged the designation, alleging that the government had unlawfully retaliated against the company, while the administration defended the action as a national security measure.
“We reject these challenges,” Judge Gregory Katsas wrote in an opinion joined by Judge Neomi Rao.
Katsas wrote that the department had sufficient support for concluding that continued integration of Claude into its information systems presented a national security risk covered by the statute.
The court pointed to restrictions Anthropic had encoded into Claude to prevent the model from performing certain tasks. According to the opinion, those restrictions had prevented Claude from performing tasks requested by government users.
The opinion also cited a dispute over whether contractual restrictions barred the use of Claude in an overseas military operation, leaving the department uncertain about whether the model would perform as expected.
“Because Anthropic was willing and able to enforce contractual restrictions through model training,” Katsas wrote, the department reasonably worried that a “critical defense system” supported by Claude might “fail to engage” as expected.
They also rejected Anthropic’s constitutional claims. Katsas wrote that Anthropic’s due process claim failed because the department promptly notified the company of the exclusion and its rationale and provided an opportunity to contest the decision.
The majority also rejected Anthropic’s First Amendment claim, stating that the Pentagon excluded the company because it refused to agree to a contract term the department considered essential, rather than because of Anthropic’s advocacy on AI policy.
Under Secretary of Defense for Research and Engineering Emil Michael praised the ruling in a post on X, writing that “the hammer of justice has smashed” Anthropic’s arguments and that “warfighters will sleep better knowing that no private company will insert their opinions in the chain of command.”
Judge Karen LeCraft Henderson dissented from the ruling, explaining that the government interpreted the supply chain risk statute too broadly.
“According to today’s decision, that contractor will have a choice: Agree to the Secretary’s demands or risk being designated a national security threat under FASCSA,” Henderson wrote.
“I cannot agree that this is the scenario the Congress had in mind when it enacted FASCSA,” she wrote, adding that Congress enacted the law to protect government systems from threats such as sabotage by foreign adversaries, rather than to address a U.S. company openly building safety restrictions into its products.
The D.C. Circuit ruling differs from an Aug. 27 decision by U.S. District Judge Rita F. Lin in California, who sided with Anthropic in a separate case challenging broader federal restrictions tied to the same dispute.
For Anthropic, the two rulings leave it with a mixed legal outcome. The company has won its California case challenging the administration’s broader restrictions but lost its challenge to the Pentagon’s supply chain risk designation in Washington. That means the Pentagon can continue excluding Claude from its systems and defense contract work, while the California ruling remains in place.
“We respectfully disagree with the court’s decision. Another federal court has already held the government’s parallel designation unlawful. We remain confident in our position and are considering all options, including further review,” an Anthropic spokesperson told MeriTalk.
Anthropic could next ask the full D.C. Circuit to reconsider the ruling or ask the Supreme Court to take up the case.