A federal judge appeared skeptical Tuesday that the Pentagon lawfully designated Anthropic a "supply chain risk," suggesting the move may have been retaliation for the AI company's public opposition to Trump administration contracting terms. U.S. District Judge Rita Lin heard arguments in San Francisco on Anthropic's motion for a preliminary injunction that would nullify the Defense Department's designation and restore the company's ability to work with federal agencies.
Lin pressed government lawyers on why the department went beyond cutting off Anthropic's own military work and instead barred the company from all government business. "What is troubling to me about these reactions is that they don't really seem to be tailored to the stated national security concern," she said. "If the worry is about the integrity of the operational chain of command, [the Department of War] could just stop using Claude. It looks like defendants went further than that because they were trying to punish Anthropic."
The judge said the broad ban "looks like an attempt to cripple Anthropic and, specifically, my concern is whether Anthropic is being punished for criticizing the government's contracting position in the press." She concluded the hearing without ruling, saying she would decide the preliminary injunction motion within days.
The government's national security argument
Deputy Assistant Attorney General Eric Hamilton argued that Anthropic poses a risk because it might try to "sabotage" the military through a future software update. He suggested the company could build a "kill switch" into its Claude chatbot if it disagreed with how the Department of War was using the technology.
"The worry is that Anthropic will, instead of merely raising concerns and pushing back, it will decide we have a problem with what DOW is doing, and we are going to manipulate the software in a way so that we can control how DOW is using it," Hamilton said. He could not confirm whether Anthropic actually has the technical ability to push such an update.
Lin called that "a pretty low bar" for banning a company from government work. "It's enough if an IT vendor is stubborn and insists on certain contracting terms, and it asks annoying questions, then it can be designated as a supply chain risk because they might not be trustworthy?" she asked.
Contradicting the secretary's public statement
In one notable exchange, Hamilton appeared to contradict Defense Secretary Pete Hegseth's public statement about the scope of the ban. Hegseth posted last month that "no contractor, supplier, or partner that does business with the United States military may conduct any commercial activity with Anthropic."
"I'm not aware of any authorities that would permit DOW to categorically bar contractors from using a company's products for non-DOW work," Hamilton said. When Lin pressed him to explain why Hegseth would make that claim if it wasn't accurate, Hamilton said, "I don't know."
Anthropic's attorney, Michael Mongan, argued that the department's rationale "continues to shift" and that the designation is an attempt to punish the company after contract negotiations failed. "I think that the reality is that this is a supply chain designation in search of a justification or a rationale and that is a problem, under the Administrative Procedure Act," he said.
Mongan urged the judge to issue a preliminary injunction quickly, saying the Pentagon's actions have caused "immediate, irreparable and ongoing harm to Anthropic, to its constitutional rights, to its reputation in the eyes of partners, customers and investors and to its business interests, not just in the national security sector."
Why this matters for legal professionals
This case tests how far the government can go in using a national security designation to exclude a company from federal contracting - and whether that power can be used in response to a vendor's public criticism. The Administrative Procedure Act claims here are standard fare for attorneys who handle government contracts or procurement disputes, but the novel question is whether a supply chain risk designation can survive judicial review when the government's own witnesses cannot articulate a concrete threat. For lawyers advising AI vendors or other technology contractors, the ruling will signal whether the courts will police the line between legitimate security exclusions and retaliatory action. Legal teams working with AI procurement issues may also want to track how this case affects the broader contracting environment for companies like Anthropic. Related training on Claude AI Courses & Certifications and AI for Government Training can help practitioners understand the technology at the center of the dispute.
Your membership also unlocks:
