The White House has unveiled a comprehensive AI policy blueprint urging Congress to establish national regulations that take precedence over state-level AI rules. Under the Trump administration, efforts to override stricter state AI measures have repeatedly fallen short, including a notable setback with the enactment of the 'One Big Beautiful Bill.'
This blueprint addresses numerous areas, ranging from safeguards for children's privacy to AI's integration in employment. The White House emphasizes that uniform implementation nationwide is essential for its effectiveness, warning that inconsistent state regulations could stifle U.S. innovation and hinder leadership in the international AI competition.
Regarding protections for young users' privacy, the blueprint recommends that Congress mandate companies to offer features such as monitoring of screen time, content access, and account management. It also reinforces that current child privacy standards extend to AI technologies, imposing restrictions on data gathering and utilization for model training. Additionally, it permits states to uphold their broader child protection statutes, including bans on child sexual abuse content, regardless of whether AI generates it.
Although the environmental and energy demands of AI facilities raise significant issues, the White House prioritizes concerns over data center expenses in its recommendations. The blueprint advocates for national AI laws to prevent elevated power costs from affecting nearby residents, while expediting approvals for building AI infrastructure to enable options like 'on-site and behind-the-meter power generation.' It further proposes reducing constraints on AI software innovation, including 'regulatory sandboxes for AI applications' and directing Congress to supply formatted federal data resources to businesses and researchers for AI purposes.
In contrast to a recent bill by Senator Marsha Blackburn (R-Tenn.) seeking to repeal Section 230—which shields platforms from liability for hosted user content—the blueprint takes a different stance. It advises Congress to bar the federal government from pressuring tech firms, including AI developers, to restrict, mandate, or modify content due to political or ideological biases, according to the White House.
The blueprint adopts a restrained approach to copyright matters and intellectual property in AI training. While the administration maintains that using copyrighted works to develop AI models does not infringe copyright law, it prefers judicial resolution over new statutes. Nevertheless, it encourages Congress to explore 'licensing frameworks' that allow rights holders to negotiate payments from AI companies.
A central element of the proposal is federal preemption of state authority, aiming to stop states from overseeing AI creation, imposing excessive obstacles on lawful AI-assisted activities, or holding AI firms accountable for misuse of their tools by others. This liability exemption for AI providers is contentious, as it intersects with pressing concerns like AI-generated explicit images of minors and its potential links to user suicides.
However, the blueprint's inconsistencies may diminish its practical value, according to Samir Jain, Vice President of Policy at the Center for Democracy and Technology, in comments to Engadget. He notes that while it includes valid principles, its contradictions and avoidance of core conflicts—such as those in children's online safety—limit its guidance for legislators. For instance, it opposes government mandates on content based on 'partisan or ideological agendas,' yet the administration's summer 'woke AI' executive order does precisely that. On preemption, it blocks state regulation of AI development but acknowledges federal law should respect states' enforcement roles against AI harms. States are at the forefront of shielding citizens from AI risks, and Congress has wisely rejected sweeping preemption twice.
President Donald Trump has sought to shape U.S. AI development and oversight with varying success, largely due to congressional reluctance to relinquish states' autonomy in tech regulation, as Jain highlights. Consequently, the extent to which this blueprint influences federal legislation remains uncertain.