Trump bans Anthropic from federal use, designates AI firm a supply-chain risk

President Trump on Saturday, 28 February 2026, ordered every federal agency to stop using Anthropic's artificial intelligence technology and directed the DOD to designate the company a supply-chain risk, a label normally reserved for firms from adversary nations such as China, reports the Wall Street Journal. 

The dramatic escalation followed Anthropic's refusal to grant the Pentagon unconditional authority over its Claude AI models for all lawful uses. Anthropic CEO Dario Amodei said the company could not "in good conscience" remove its prohibitions on domestic mass surveillance and autonomous weapons, calling the demand a line it would not cross. DOD officials countered that the military would never use AI unlawfully and that Anthropic needed to relinquish control.

Hours after the ban, OpenAI announced it had reached its own classified-use agreement with the Pentagon. CEO Sam Altman said the deal includes the same guardrails Anthropic had sought, prohibitions on mass surveillance and autonomous weapons plus technical safeguards, and urged the administration to extend the same terms to all vendors. Elon Musk's xAI also recently won classified-setting approval, though agencies have raised reliability concerns about its Grok model.

Defense Secretary Pete Hegseth's supply-chain risk designation could force government contractors to prove they do not use Claude in any capacity, potentially cutting into Anthropic's broader commercial business. Anthropic said it will challenge the designation in court, calling it "legally unsound" and a "dangerous precedent." Dean Ball, a former Trump AI adviser now at the center-right Foundation for American Innovation, warned the move could chill private-sector willingness to work with the Pentagon and said partners including Nvidia, Amazon and Google may face pressure to divest from Anthropic.

The dispute traces back to a July 2025 executive order on "woke AI" widely seen as targeting Anthropic, as well as the company's hiring of former Biden officials and its advocacy for AI regulation. Industry analysts say the clash is less about substantive policy differences than about political alignment. Each company's Pentagon contract is valued at up to $200 million.

GSA has already removed Anthropic from its procurement offerings, and agencies face a six-month phaseout, a transition AI experts say could set back government modernization efforts.