Democrats plan to introduce a wave of legislation on artificial intelligence and data centers once they take control of Congress in January, according to lawmakers, congressional aides and advocates cited by Reuters.

The proposals include bills to shield consumers from rising electricity costs tied to data centers, require independent testing of advanced AI systems, and clarify when AI companies can be held liable for harm caused by their products, the report said.

Any such legislation would face steep odds while President Donald Trump remains in office. The report says he has repeatedly dismissed calls for mandatory AI rules. On Friday, Trump said he would work with US states that "want to become RICH" by embracing AI, without offering details.

Democrats said committee control would let them advance proposals, compel testimony from industry leaders and build support for action if AI systems cause real-world harm. Representative Greg Casar, a Texas Democrat, said AI executives should be called to testify under oath.

Representative Lori Trahan of Massachusetts plans to push legislation requiring independent experts to evaluate leading AI models against safety standards, according to her deputy chief of staff, Francis Grubar. Representative Zoe Lofgren of California said Democrats would focus on "responsible regulations for AI while also fostering innovation."

The report also cites a Reuters poll published in October, which found that most US voters think the Trump administration and Congress are not taking AI risks seriously enough. Ed Mills, who manages policy research at Raymond James, said "the probability of legislation is much higher than people anticipate." Congress could override a presidential veto with two-thirds support in both chambers, though the report does not say whether that is likely.

Democrats also plan to scrutinize major AI companies, including OpenAI and Anthropic. In October, Lofgren and other House committee leaders asked OpenAI, Anthropic, Meta, Alphabet's Google and SpaceX to detail their security practices and report instances of their AI tools acting in ways their designers did not intend.