Anthropic is now officially labelled a "supply chain risk" in the US (Updated)

As announced last weekend, the Pentagon has now officially labeled Anthropic a “supply chain risk.” This is the culmination of a highly publicized standoff between the AI company and the US government that was sparked over security guardrails in using Anthropic’s Claude models for military operations. Anthropic is now officially the first-ever domestic company to be designated a “supply chain risk" status in American history.

President Donald Trump and Defense Secretary Pete Hegseth made the announcement after Anthropic refused to drop specific restrictions on its Claude AI model. Two main points of conflict were using Anthropic’s AI technology for mass domestic surveillance and fully autonomous weapons.

In an official statement, Anthropic said it doesn’t believe that today’s frontier AI models are reliable enough to be used in fully autonomous weapons. The company also stated that mass domestic surveillance of American citizens violates fundamental rights.

Because CEO Dario Amodei refused to bend the terms of service, the government is cutting ties and terminating its existing $200 million defense contract with Anthropic. This contract was signed in July 2025 as part of a broader government plan to utilize technology from leading AI vendors for military missions. Several major tech firms were awarded a two-year deal with a ceiling of up to $200 million for supplying the government with frontier AI models. That initial group included Google, OpenAI, xAI, alongside Anthropic.

The supply chain risk label carries massive hurdles for Anthropic. President Trump directed all federal agencies to immediately stop using Anthropic technology. Defense Secretary Pete Hegseth also added that any company doing business with the government risks violating the presidential directive and DoD supply-chain risk rules if they continue to use Anthropic’s AI. This could easily result in contract breaches or funding cuts. Contractors might also face audits and severe penalties under federal acquisition laws.

However, in a rather contradictory move, the Pentagon gave itself a six-month grace period before it stops using Anthropic’s technology, “to allow for a seamless transition to a better and more patriotic service.” The military even used Claude through Palantir’s Maven system for recent military operations in Iran, which happened only two days after the Pentagon’s announcement.

Anthropic plans to challenge this decision in court, citing that the Defense Department does not actually have the statutory authority to block private contractors from using Claude for their other commercial clients.

In the midst of the fallout between Anthropic and the government, OpenAI quickly struck a deal to supply its own AI models to classified military networks for “all lawful purposes.” This development triggered another chain reaction, most notably with users of both AI platforms.

Refusing to comply with Pentagon requests proved to be a significant marketing boost for Anthropic. People were praising CEO Amodei for standing his ground against the Trump administration, which resulted in Claude becoming the most popular app on Android and iOS in just a few days.

At the same time, OpenAI’s agreement triggered the exact opposite reaction. CEO Sam Altman received massive public backlash for striking the deal with the Pentagon just hours after Anthropic refused. This initiated the beginning of the “Quit ChatGPT” movement and a massive exodus of users from the platform. Well over 2.5 million users have already stopped using ChatGPT in the last couple of days.

As Bloomberg reports, Anthropic was officially given the “supply chain risk” status by the government today. We’ll follow the further course of action from all parties involved.

Update (March 5, 4:45 PM PST): Anthropic CEO Dario Amodei released a new statement confirming the company officially received the supply chain risk designation letter from the government. Anthropic is doubling down on its ethical boundaries and will take the Department of Defense to court, as the company believes the government's decision isn't legally sound.

The Department’s letter has a narrow scope, and this is because the relevant statute (10 USC 3252) is narrow, too. It exists to protect the government rather than to punish a supplier; in fact, the law requires the Secretary of War to use the least restrictive means necessary to accomplish the goal of protecting the supply chain. Even for Department of War contractors, the supply chain risk designation doesn’t (and can’t) limit uses of Claude or business relationships with Anthropic if those are unrelated to their specific Department of War contracts.

Amodei further stated that Anthropic has had productive conversations with the Department of War over the last several days and that he's proud of the work his company has done for the government before, but draws the line at mass domestic surveillance and fully autonomous weapons.