President Donald Trump's administration is pressing a gridlocked Congress to enact a national artificial intelligence framework, warning that a patchwork of state regulations threatens U.S. innovation and global leadership.

The White House unveiled the four-page "National Policy Framework for Artificial Intelligence: Legislative Recommendations" on March 20, 2026. It outlines six pillars aimed at balancing AI advancement with targeted protections. Key proposals include empowering parents with tools to manage children's online accounts and devices, requiring AI platforms to reduce risks of sexual exploitation or self-harm promotion, and streamlining permitting for data centers to generate on-site power without burdening ratepayers.

The framework also calls for respecting intellectual property rights while affirming fair use for AI training on copyrighted materials, preventing AI censorship of lawful speech, removing barriers to innovation, and expanding workforce training for an AI-driven economy. It explicitly urges Congress to preempt state laws that impose undue burdens on AI development or penalize developers for third-party misuse.

Michael Kratsios, head of the White House Office of Science and Technology Policy, emphasized the need for uniformity. "We want to create an environment where innovators have certainty about the way that they can develop their products, and it's something only Congress can provide," he said. "The first step is to create one national framework so we can avoid a patchwork."

Despite Republican support, including from Sen. Marsha Blackburn, who called it a "roadmap for crafting legislation" and introduced the TRUMP AMERICA AI Act, Congress has made no progress. House Speaker Mike Johnson and Leader Steve Scalise praised it as providing "much-needed certainty" for innovators.

States have filled the void with their own measures. Utah Republicans proposed transparency rules for AI consumer protections, but a White House memo helped stall them. Pennsylvania's SAFECHAT Act mandates safeguards against harmful chatbot content, while Texas requires AI disclosures and bans on discriminatory uses. State leaders argue federal inaction justifies their steps. "Congress is in a gridlock and they not only will not act, but they can't act," said Utah Rep. Doug Fiefia.

This push builds on a December 2025 executive order directing federal efforts to block obstructive state AI rules. A prior attempt for a 10-year state moratorium failed amid bipartisan opposition. The administration maintains that national standards will shield communities from AI-enabled scams, bolster energy reliability, and ensure U.S. dominance against competitors like China.

As midterm elections approach, prospects for bipartisan passage remain uncertain, with Democrats criticizing the plan for insufficient accountability and states resisting preemption.