Washington. The Trump administration on Friday ordered all US agencies to stop using Anthropic’s artificial intelligence technology and imposed other penalties, escalating an unusually public clash over AI safety.

President Donald Trump, Defense Secretary Pete Hegseth and other officials took to social media to criticize Anthropic for refusing to grant the military unrestricted use of its AI technology by a Friday deadline. They accused the company of jeopardizing national security after CEO Dario Amodei declined to ease safeguards on its products.

“We don’t need it, we don’t want it, and will not do business with them again!” Trump wrote.

Hegseth labeled the company a “supply chain risk,” a designation typically applied to foreign adversaries that can disrupt key business partnerships.

In a statement Friday evening, Anthropic said it would challenge what it called an unprecedented and legally unsound action “never before publicly applied to an American company.”

Anthropic said it sought limited assurances from the Pentagon that its AI chatbot Claude would not be used for mass surveillance of Americans or fully autonomous weapons. The Pentagon said it had no interest in such uses and would deploy the technology only lawfully, but insisted on access without restrictions.

“No amount of intimidation or punishment from the Department of War will change our position on mass domestic surveillance or fully autonomous weapons,” the company said. “We will challenge any supply chain risk designation in court.”

The confrontation highlights broader tensions over AI’s role in national security and concerns about how increasingly capable systems could be used in lethal operations, sensitive intelligence work or government surveillance.

OpenAI Strikes Pentagon Deal

Hours after Anthropic’s punishment, OpenAI CEO Sam Altman announced that his company had reached a deal with the Pentagon to supply its AI to classified military networks, potentially filling the gap.

Altman said the same red lines central to Anthropic’s dispute are embedded in OpenAI’s agreement.

“Two of our most important safety principles are prohibitions on domestic mass surveillance and human responsibility for the use of force, including for autonomous weapon systems,” Altman wrote, adding that the Defense Department agrees with those principles and reflects them in law and policy.

He said he hopes the Pentagon will “offer these same terms to all AI companies” to de-escalate legal and governmental actions in favor of negotiated agreements.

Trump Escalates Rhetoric

Trump said Anthropic had erred in attempting to “strong-arm” the Pentagon. In a Truth Social post, he ordered most agencies to immediately stop using Anthropic’s AI but gave the Pentagon six months to phase out technology already embedded in military platforms.

“The United States of America will never allow a radical left, woke company to dictate how our great military fights and wins wars!” he wrote.

Private negotiations collapsed earlier this week when Amodei said his company “cannot in good conscience accede” to the government’s demands.

While Anthropic can absorb the loss of the contract, the administration’s action poses broader risks as the company rises from a San Francisco research lab to one of the world’s most valuable startups.

Pentagon spokesman Sean Parnell said Anthropic’s refusal was “jeopardizing critical military operations and potentially putting our warfighters at risk.” Hegseth said the department “must have full, unrestricted access to Anthropic’s models for every lawful purpose in defense of the Republic.”

Trump warned the company to “get their act together” during the phase-out period or face “major civil and criminal consequences.”

However, the supply chain risk designation is typically reserved for companies owned by US adversaries, raising questions about its use against a domestic firm.

Sen. Mark Warner of Virginia, the top Democrat on the Senate Intelligence Committee, said the move, “combined with inflammatory rhetoric attacking that company, raises serious concerns about whether national security decisions are being driven by careful analysis or political considerations.”

Silicon Valley Reacts

The dispute rattled Silicon Valley, where venture capitalists, AI researchers and employees of rival firms OpenAI and Google voiced support for Anthropic in open letters and online forums.

The decision could benefit OpenAI’s ChatGPT and Elon Musk’s Grok, which the Pentagon also plans to connect to classified networks. It may also serve as a warning to Google, which maintains a developing contract to provide AI tools to the military.

Musk backed the administration, writing on X that “Anthropic hates Western Civilization.” Altman struck a more conciliatory tone, voicing support for Anthropic’s safeguards while criticizing what he called the government’s “threatening” approach.

The episode marks another turn in the rivalry between OpenAI and Anthropic, which was founded in 2021 by former OpenAI executives.

Retired Air Force Gen. Jack Shanahan, who previously led Pentagon AI initiatives, wrote on social media that targeting Anthropic “garners spicy headlines, but everyone loses in the end.”

Shanahan said Claude is already widely used across government, including in classified settings, and called Anthropic’s guardrails reasonable. He added that large language models powering systems like Claude, Grok and ChatGPT are “not ready for prime time in national security settings,” particularly for fully autonomous weapons.