Elizabeth Warren criticises Pentagon move blocking Anthropic, calls it retaliation

Anthropic is gaining support in its dispute with the U.S. Department of Defence, which last month labelled the AI company a supply-chain risk after it declined to make concessions about how the military could use its technology.

In a letter addressed to Defence Secretary Pete Hegseth, U.S. Senator Elizabeth Warren (D-MA) described the Pentagon’s decision as “retaliation,” arguing that the Department of Defence could have ended its contract with Anthropic instead of taking broader action.

“I am particularly concerned that the DoD is trying to strong-arm American companies into providing the Department with the tools to spy on American citizens and deploy fully autonomous weapons without adequate safeguards,” Warren wrote, according to CNBC. She added that the decision to block Anthropic “appears to be retaliation.”

Warren’s criticism reflects concerns raised by a range of organisations and individuals who have publicly opposed the Defence Department’s treatment of the AI company. Several technology firms and employees — including individuals affiliated with OpenAI, Google, and Microsoft — along with civil liberties groups, have submitted amicus briefs supporting Anthropic and criticising the designation. Such classifications are typically reserved for foreign adversaries, not domestic companies.

The conflict began after Anthropic informed the Pentagon that it did not want its AI systems used for mass surveillance of U.S. citizens and stated that the technology was not sufficiently mature to be deployed in targeting or firing decisions for autonomous weapons without human oversight. The Pentagon responded by asserting that private companies should not dictate military use of technology and subsequently designated Anthropic as a supply chain risk.

This designation requires any organisation working with the Department of Defence to certify that it does not use Anthropic’s products or services, effectively limiting the company’s ability to collaborate with partners connected to the federal government.

Warren’s letter comes just ahead of a scheduled hearing in San Francisco, where District Judge Rita Lin will consider whether to grant Anthropic a preliminary injunction. The request seeks to maintain current conditions while the broader legal case against the Department of Defence proceeds.

Anthropic has filed a lawsuit alleging that the Pentagon’s actions violate its First Amendment rights and amount to punishment based on ideological differences. The Defence Department, however, maintains that the situation stems from a business dispute rather than protected speech and argues that its designation was based on national security grounds, not in response to the company’s views.

In filings submitted last week, Anthropic provided declarations claiming that the government’s reasoning is flawed, citing what it described as technical misunderstandings and concerns not previously raised during discussions with the Pentagon.

Separately, Warren has also reached out to OpenAI CEO Sam Altman, requesting details about OpenAI’s agreement with the Department of Defence. This inquiry followed closely after the Pentagon decided to place Anthropic on its restricted list.