Court Rules on Pentagon's Stance
A recent 2-1 decision by the U.S. Court of Appeals for the District of Columbia Circuit has upheld the Department of Defense's authority to place artificial intelligence startup Anthropic on a national security blacklist. This ruling marks a significant legal setback for the San Francisco-based creator of the Claude AI chatbot, affirming the Pentagon's assessment that integrating Anthropic’s technology into military systems could pose a national security threat.
The Core of the Dispute
The conflict originated from negotiations surrounding a proposed $200 million deal between the Department of Defense and Anthropic. The agreement aimed to prototype advanced AI capabilities. However, talks collapsed after Anthropic declined to grant the Pentagon unrestricted access for “all lawful uses” of its technology. The company insisted on establishing explicit “red lines” to prevent its AI from being deployed in mass surveillance or fully autonomous weapons systems.
Legal Justification and Dissent
Writing for the majority, Judge Gregory Katsas stated that the federal government acted lawfully in designating Anthropic as a “supply chain risk.” The court rejected Anthropic's argument that the blacklisting infringed upon its First Amendment rights, concluding that the government's action was legitimately “based on the company's refusal to assent to a contract term that the Department deemed essential.”
Despite the majority opinion, one judge on the three-member D.C. panel dissented, arguing that there was no legitimate concern that Anthropic could manipulate its technology to create a genuine security risk.
Implications for Anthropic and Military AI
The D.C. ruling effectively prohibits the use of Anthropic's Claude AI within the Department of Defense, presenting a considerable challenge for the company, especially as it prepares for a planned initial public offering. This decision, however, does not overturn an earlier August ruling by a California federal court, which partially invalidated a parallel government designation and permits other civil agencies and contractors to continue collaborating with the startup.
In response to the ruling, an Anthropic spokesperson stated, “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.” Meanwhile, defense officials celebrated the court’s backing, with Undersecretary Emil Michael posting on X that “warfighters will sleep better knowing that no private company will insert their opinions in the chain of command.”
Industry Reactions and Next Steps
The standoff had previously prompted Secretary of War Pete Hegseth to accuse Anthropic of attempting to dictate military policy, leading President Donald Trump to direct federal agencies to cease business with the firm. Anthropic now has the option to request an en banc rehearing before the full 11-judge D.C. Circuit bench or escalate its appeal to the U.S. Supreme Court.