For the past two years, states have been writing their own AI rules. California's SB 1047 vetoed or not, Colorado's AI Act, Texas's proposals - a growing web of local regulation that tech companies have loudly complained about. Today, the White House told Congress to shut it all down and replace it with a single federal standard.
The Trump administration released what it calls a "National AI Legislative Framework," a blueprint for how Congress should regulate artificial intelligence. The document lays out six principles, but the one that will generate the most debate is federal preemption - the idea that federal AI law should override state-level AI regulations entirely.
The Six Principles
The framework covers:
- Child protection and parental controls - giving parents tools to manage AI interactions with minors, with specific focus on preventing exploitation and self-harm
- Community safety - protecting Americans from AI-related harms
- Intellectual property - protecting creators' and publishers' work "in the age of AI" while still allowing AI systems to learn from existing content under fair use
- Free speech - preventing AI censorship (a recurring theme in this administration's tech policy)
- American AI dominance - keeping the U.S. ahead in the global AI race
- Workforce development - preparing Americans for AI-driven job changes
The framework also pushes for faster permitting so data centers can generate power on-site, which addresses a real bottleneck. AI data centers are hungry for electricity, and current permitting processes can take years.
The State Preemption Fight
This is where things get contentious. Trump signed an executive order in December attempting to block states from crafting independent AI regulations, arguing that "a patchwork of rules would hurt growth in the sector." The legislative framework doubles down on that position.
The White House framing is that "strong federal leadership" is needed so "the public can trust how artificial intelligence is being used in their lives." Translation: let us write one set of rules instead of letting 50 states write 50 different ones.
There is a real argument here. Companies building AI tools do face genuine compliance headaches when every state has different requirements. But consumer advocates and some members of Congress from both parties have pushed back, arguing that federal preemption often means weaker protections, not better ones. When Washington preempts state law, it tends to set a floor that looks more like a ceiling.
The framework is deliberately light on enforcement details. It reads more like a wish list than draft legislation - heavy on principles, thin on mechanisms. Congress will have to fill in the gaps, and given the current partisan dynamics around tech regulation, that process could take a while.
What This Means for AI Companies
For companies building and selling AI tools, federal preemption would be a significant win. One compliance framework instead of dozens. For users, the question is whether federal rules will actually address the specific harms that motivated state legislation in the first place - algorithmic discrimination, deepfake abuse, lack of transparency.
The framework does not mention open-source AI policy, deepfake-specific provisions, or AI in hiring - all areas where states have been most active. Those gaps will need to be filled before this becomes law, if it ever does.