The administration of Donald Trump has introduced a new national artificial intelligence policy framework aimed at giving the federal government sweeping authority to override state-level AI regulations. The proposal, released on March , 2026, calls on Congress to establish a unified national approach to AI governance to avoid what officials describe as a fragmented regulatory landscape.
At the core of the framework is a push for federal pre-emption of state AI laws, which would prevent individual states from imposing their own rules on AI development and deployment. The administration argues that inconsistent state regulations could slow innovation and weaken the United States’ ability to compete globally, particularly against rivals like China.
The policy outlines a “light-touch” regulatory approach, focusing on promoting innovation while addressing key concerns such as child safety, intellectual property, and energy consumption. It encourages Congress to rely more on existing laws and judicial systems rather than introducing strict new regulations, signalling a preference for flexibility over heavy oversight.
One of the major components of the proposal includes protections for children using AI systems, such as parental control tools, safeguards against harmful content, and measures to reduce risks of exploitation. The framework also addresses the growing energy demands of AI infrastructure, supporting initiatives that allow data centres to manage or generate their own power.
The plan also touches on broader issues such as free speech and censorship, emphasizing that AI tools should not be used to restrict lawful political expression. At the same time, it avoids taking a firm stance on contentious topics like copyright disputes, instead suggesting that courts should resolve such issues over time.
However, the proposal has sparked concerns among lawmakers and state governments. Several U.S. states, including California, Texas, Colorado, and Utah, have already introduced their own AI regulations, and officials fear that a federal override could undermine these efforts. Critics argue that the framework lacks strong accountability measures for AI companies and does not sufficiently address risks such as algorithmic bias or misuse.
Political challenges are also expected as the proposal moves to Congress. While some Republican leaders have supported the initiative, gaining bipartisan approval may prove difficult due to differing views on regulation, state rights, and the role of government oversight in emerging technologies.
The framework is not legally binding and will require congressional approval to be implemented. Nonetheless, it represents a significant step in shaping the future of AI policy in the United States, highlighting the growing importance of balancing innovation, regulation, and national competitiveness in the rapidly evolving AI landscape.
