Are Efforts to Pre-empt State AI Regulation Actually Dead?

Amid pushback earlier this year, Congress scrapped a proposed decadelong ban on state and local governments enforcing regulations on artificial intelligence. But that idea, while shelved, isn’t dead, cautioned speakers at the National Conference of State Legislatures (NCSL)’s recent summit.

Opponents to the 10-year moratorium had successfully advocated against the idea, largely on the grounds that it would remove state- and local-level protections without creating anything at the federal level to fill that void. So far, there hasn’t been much momentum to address that Achilles’ heel and craft a comprehensive federal AI measure, said Barrie Tabin, NCSL senior legislative director.

“It's going to be interesting when this moratorium comes back — which we believe it's going to, in one shape or form or another — how that conversation is going to play out, because there still won't be anything comprehensive at the federal level,” Tabin said.

Federal lawmakers may be considering putting a pre-emption measure in the National Defense Authorization Act, or drafting a standalone bill that would pre-empt some state laws, said Connecticut state Sen. James Maroney during the panel.

Any future standalone law imposing restrictions would require more votes to pass than this year’s provision, which was stuffed into a budget reconciliation measure, Tabin noted.

Whatever the avenue, the administration doesn’t seem keen to give up on freeing up companies developing this technology. The White House released an AI action plan in July 2025 instructing several federal agencies to find ways to withhold AI-related funding to states with “burdensome” AI regulations. The administration argues that such “bureaucratic red tape” holds the private sector back and threatens America’s ability to become the global leader on AI. The action plan adds, however, that the federal government should “not interfere with states’ rights to pass prudent laws that are not unduly restrictive to innovation.”

Specifically, the AI action plan recommends the Office of Management and Budget help ensure that agencies with AI-related discretionary funding programs consider a state’s regulatory climate when making funding awards, and “limit funding if the state’s AI regulatory regimes may hinder the effectiveness of that funding or award.” How it will play out remains to be seen.

“We don’t really quite know right now what these discretionary funds are,” Tabin said.

And the administration hasn’t yet defined what, exactly, counts as “overly burdensome,” said Maroney. That vagueness raises questions about whether having a state data privacy law could be deemed overly burdensome to AI companies that want access to a lot of data on which to train their models. The phrasing of the original, defeated moratorium would have blocked such laws, Maroney said.

Experts, meanwhile, praised other parts of the federal AI action plan, especially its focus on helping workers displaced by AI. The plan calls for the federal government to guide states on finding affected workers and provide funding for rapidly retraining workers who’ve been, or are likely to be, displaced.

And states haven’t been idle on this issue, either. Connecticut passed legislation calling for the state to measure AI’s impact on the workforce — including tracking the kinds of jobs being created and eliminated as a result of the technology — and to provide retraining opportunities. New York is also tracking impacts, with a law that requires employers to inform the state if AI or other technological innovations or automations contributed to layoff decisions.

Getting such data is an important first step as states work to respond, Maroney said.

“All of our states need to be thinking about workforce development, how we make sure we give everyone the opportunity to succeed,” Maroney said. “We need to be working on reskilling and upskilling our population.”

It remains to be seen, however, whether any state policies aimed at regulating this technology will survive the test of time — or the test of future federal legislation.