Nearly 50 House Democrats say no to unauthorized AI in Trump administration

April 16 (Reuters) - Members of Congress have called on the Trump administration to cease applying any unauthorized artificial intelligence system toward its effort to slash government spending, according to a letter they sent on Wednesday.

Nearly 50 Democrats in the U.S. House of Representatives signed onto the message, seen by Reuters, expressing concern about AI's use by the Department of Government Efficiency. Republican President Donald Trump set up DOGE and tasked billionaire Elon Musk to cut what they say is waste, fraud and abuse in the federal bureaucracy.

Sign up here.

While the Democrats acknowledged AI's potential to modernize the U.S. government, they raised concerns about reported employee monitoring and access to sensitive data by DOGE and its AI tools.

"These present serious security risks, self-dealing, and potential criminal liability if not handled correctly, and have the potential to undermine successful and appropriate AI adoption," said the letter, spearheaded by Representatives Donald Beyer, Mike Levin and Melanie Stansbury.

Trump officials have said that AI is being used to look for worker communications considered hostile to the administration, and DOGE is heavily using Grok AI technology from Musk's startup xAI, Reuters reported last week.

The White House Office of Management and Budget, to which the Democrats addressed their letter, did not immediately respond to a request for comment. The Environmental Protection Agency previously said of Reuters' reporting that it was looking for AI to make the EPA more efficient but not using it for "personnel decisions in concert with DOGE.”

The House Democrats said in the letter that the administration should terminate the use of any AI systems that have not gone through formal approval processes such as FedRAMP or that do not meet existing legal requirements.

In particular, they questioned OMB on whether the administration had used technology powered by Musk's xAI. They sought an account of new software deployed by the Trump administration, along with information on how such tools complied with certain laws and whether they had various authorizations.

The letter warned against potential conflicts of interest, which the Democrats claimed would become "exponentially worse if Musk pursues further contracts to become a major provider of government AI services."

As a special government employee, Musk may not involve himself in federal activities that would benefit his companies, in keeping with ethics laws.

He did not immediately respond to a request for comment.

The Trump administration issued directives this month to streamline AI's acquisition and adoption in government, while requiring agencies to manage risks from any high-stakes uses of the technology, like hiring and firing.

Reporting by Jeffrey Dastin in San Francisco and Marisa Taylor in Washington; Editing by Mark Porter

Our Standards: The Thomson Reuters Trust Principles., opens new tab