White House publishes National AI Legislative Framework

The framework takes a pro-innovation and pro-free-speech approach, aiming to protect children, safeguard communities, respect intellectual property, promote U.S. AI leadership, prepare the workforce, and establish a national standard that pre-empts conflicting state laws while preserving certain state powers.   

The framework is a high-level, directional document, and its practical impact will depend on whether and how Congress acts on its recommendations. 

The framework is structured around seven pillars, summarized below. 

1. “Protecting children and empowering parents.”

The framework calls on Congress to give parents greater control over their children's privacy, screen time, and content exposure on AI platforms, and recommends that platforms likely to be accessed by minors adopt "commercially reasonable, privacy protective, age-assurance requirements" to reduce risks of exploitation and self-harm.   

2. “Safeguarding and strengthening American communities.”

The framework seeks to shield residential ratepayers from increased electricity costs driven by AI data center construction and recommends streamlining federal permitting processes to enable AI developers to build or procure on-site power generation, thereby accelerating infrastructure development and enhancing grid reliability. It also proposes strengthening law enforcement capacity to combat AI-enabled scams, particularly those targeting vulnerable individuals, and ensuring national security agencies can assess frontier AI models. Congress should also support small business with implementing a wider deployment of AI tools by offering AI resources, grants, and tax incentives.  

3. “Respecting intellectual property rights and supporting creators.”

The Administration recognizes the need to strike a balance between protecting the work of American innovators and allowing AI models to make fair use of available material, and therefore it supports letting the courts decide. It also encourages Congress to consider licensing frameworks that would allow rights holders to negotiate compensation with AI providers and proposes federal protections against unauthorized AI-generated replicas of an individual’s voice or other identifiable attributes.

4. “Preventing censorship and protecting free speech.”

The framework proposes measures to prevent the U.S. government from coercing AI providers to alter content on ideological grounds, and to give Americans a right of redress against government censorship on AI platforms. 

5. “Enabling innovation and ensuring American AI dominance.”

The framework recommends regulatory sandboxes and AI-ready federal datasets. Notably, it recommends against creating any new federal rulemaking body to regulate AI, preferring to rely on existing sector-specific regulators and industry-led standards. 

6. “Educating Americans and developing an AI-ready workforce.”

The framework emphasizes that American workers should benefit directly from AI-driven growth through skills training, new job creation, and expanded opportunities across sectors. It calls for non-regulatory methods to integrate AI training into existing education, workforce, and apprenticeship programs. It also recommends bolstering capabilities at land-grant institutions (specific U.S. universities) to provide technical assistance, launch demonstration projects, and develop AI youth development programs. 

7. “Establishing a federal policy framework, pre-empting cumbersome state AI laws.”

The framework recommends that Congress enact a unified federal standard for AI regulation that would pre-empt state-level AI laws the administration considers to “impose undue burdens” and replace them with a streamlined national approach. Specifically, the framework urges Congress to bar states from enacting laws that govern how AI models are developed and to shield developers from liability for downstream third-party misuse of their models. It also directs lawmakers to refrain from establishing any new federal regulatory body dedicated to AI oversight, instead favoring sector-specific regulation through existing agencies supplemented by industry-led standards.

The press release is available here and the full framework here.