White House, GOP Again Trying to Enact Federal Preemption of State AI Laws

According to the Post, White House Special Adviser for AI and Crypto David Sacks, Director of the White House Office of Science and Technology Policy Michael Kratsios, House Majority Leader Steve Scalise and House Speaker Mike Johnson huddled throughout last week to devise a strategy for slipping the preemption measure into the online safety package without attracting too much opposition.

One tactic under discussion, per the Post’s sources, is to also include the NO FAKES Act in the package in an effort to garner support from Sen. Marsha Blackburn (R-TN). NO FAKES, which would hold people and companies liable for creating or distributing non-consensual deepfakes, has been a priority for Blackburn, who previously represented the music capital of Nashville while in the House, but who has been a strong opponent of federal preemption.

Around 20 states have passed comprehensive privacy laws covering AI and several others have passed more limited measures. AI companies and investors have clamored for Congress to enact a single, federal standard that would override the growing patchwork of often conflicting state regulations.

“What I fear is, if we don’t take action today to stop this growing patchwork of AI laws, we’re going to end up in the same place as we are today,” David Grossman, VP of regulatory affairs at the Consumer Technology Association told the Post.

Republican leaders in Congress sought to include a 10-year moratorium on state passage or enforcement of AI laws in the One Big Beautiful budget reconciliation bill passed last summer but were forced to drop the provision in the face of dissention within their own ranks. They tried again in September by attaching the moratorium to the National Defense Authorization Act but were again forced to scuttle the provision to hold their narrow majority together.

Read more: Anthropic Sues Pentagon to Block National Security Blacklist Over AI Restrictions

President Trump then threatened to withhold federal funds from states with “unduly burdensome” AI regulations by executive order in January. But the EO does not have the force of law, and some states could be targeted are controlled by Republicans leading to pushback within the GOP.

While the package of children’s online safety bills has a measure of bipartisan support on Capitol Hill, raising Republican hopes of slipping through federal AI preemption, passage of the package is not assured. The current version does not include a “duty of care” element that would hold companies directly responsible for harms their technology causes children and had significant bipartisan support.

Nor have GOP leaders yet devised a strategy to attracting the seven Democratic vote needed in the Senate to overcome a filibuster, per the Post. The top Democrat on the Senate Commerce committee, Sen. Maria Cantwell (D-WA), has previously led the negotiations on AI policy for her party but to date there has been little bipartisan engagement on the latest GOP effort, according to the Post’s sources.

“One, we don’t take orders from the White House. Two, they’re tried zero engagement, so good luck with that,” said Rep. Ted Lieu (D-CA), chair of the newly formed House Democratic commission on AI.

With Congress now confronted with the war in Iran, and the ongoing shutdown of the Department of Homeland Security, there is currently no timeline for bringing the online safety package to a vote, with or without the AI preemption provision.