Democratic Sen. Elizabeth Warren of Massachusetts participates in her Senate Armed Services Committee confirmation hearing on Thursday, January 15, 2026 in Washington, DC.
Aaron Schwartz | Bloomberg | Getty Images
Sen. Elizabeth Warren of Massachusetts said the Pentagon’s decision to designate artificial intelligence startup Anthropic as a supply chain risk “sounds like retaliation.”
In a formal letter to Defense Secretary Pete Hegseth on Monday, Warren noted that the department “could have chosen to terminate the contract with Anthropic or continue to use its technology in unclassified systems.”
“I am particularly concerned that the Pentagon is attempting to strongly intimidate American companies into providing the Pentagon with the tools to spy on American citizens and deploy fully autonomous weapons without appropriate safeguards,” Warren wrote.
As the war in Iran continues and the conflict enters its fourth week, U.S. senators are seeking more answers from the Pentagon and its contracts with tech companies.
In the days leading up to the war, the Pentagon and Anthropic clashed because Anthropic wanted unfettered access to its models for any “lawful purpose,” while Anthropic wanted guarantees that the models would not be used for fully autonomous weapons or domestic “mass surveillance.”
Hegseth posted on February 27 that he was directing the Department of Defense to apply a “supply chain risk” label to the company. The official notification came a week later as the ministry continued to use Anthropic’s Claude model in Iran.
Anthropic filed a lawsuit against the Trump administration after the company was blacklisted and deemed a threat to U.S. national security. A preliminary hearing in the case is scheduled for Tuesday in the U.S. District Court for the Northern District of California.
Hours after Anthropic was blacklisted, OpenAI came on the scene and announced a deal with the Department of Defense.
The company said it was confident the Pentagon would not use its AI systems for mass surveillance or fully autonomous weapons, citing OpenAI’s “safety stack,” existing laws, and contract language that has not been fully shared.
But neither Mr. Altman nor the Pentagon has been able to allay the concerns of lawmakers, the public and some company employees.
Warren also wants answers from OpenAI CEO Sam Altman.
In Monday’s letter, Warren asked Altman for information about the terms of the agreement with the Pentagon.
“We are concerned that the terms of this agreement will allow the Trump administration to use OpenAI technology to conduct mass surveillance of the American public and build lethal autonomous weapons that could harm civilians with little human oversight,” the letter said.
Last week, Mr. Altman met with several members of Congress in Washington, D.C., where Sen. Mark Kelly, D-Ariz., raised “serious questions” about the company’s approach to war and its contracts with the Pentagon.
“Ultimately, it is impossible to assess the safeguards and prohibitions that may exist in OpenAI’s contract with the Department of Defense without seeing the full contract document, which neither the Department of Defense nor OpenAI has made public,” Warren wrote.
He added that what was published raises serious concerns about the Department of Defense’s use of AI.
Despite calls for answers, Senate Democrats have limited ability to force action because Republicans control the White House and both chambers of Congress.
