White House Outlines AI Policy Agenda in New National Framework

The White House released a set of legislative recommendations March 20, 2026, as part of its National Policy Framework for Artificial Intelligence. The document signals the administration's priorities but is not law and creates no binding obligations. Recommendations across seven sections address child safety, infrastructure and economic growth, intellectual property and digital replicas, free speech, workforce development and limits on state by state AI regulation. On child safety the paper calls for age-assurance measures on AI services likely accessed by minors, stronger parental controls and features to reduce risks of exploitation and self-harm. It states that existing child privacy protections should apply to AI systems with limits on data collection for training and advertising. The recommendations urge Congress to protect residential ratepayers from electricity costs linked to AI data centers and to streamline federal permitting for power development. They also call for stronger tools against AI-enabled fraud and support for small-business AI adoption through grants and incentives. On intellectual property the administration states that training AI models on copyrighted material does not violate copyright law though courts should resolve fair-use questions. It recommends avoiding legislation that would affect judicial resolution and suggests considering licensing frameworks for rights holders. The document supports a federal framework to protect against unauthorized commercial use of AI-generated digital replicas of voice or likeness with First Amendment exceptions. It urges Congress to prohibit federal coercion of AI providers on content based on partisan agendas and to provide redress for censorship efforts. The paper calls for regulatory sandboxes and broader federal dataset access while arguing against a new federal AI rulemaking body. It favors sector-specific oversight by existing agencies and industry-led standards. It endorses federal preemption of state AI laws that impose undue burdens while preserving state authority over generally applicable laws, zoning and a state's own use of AI.