US Lawmakers Unveil 269-Page Draft AI Bill Targeting Frontier Developers

A Bipartisan Push for Federal AI Oversight

A bipartisan coalition in the US House of Representatives has released a comprehensive 269-page draft of the Great American Artificial Intelligence Act of 2026. Spearheaded by Rep. Jay Obernolte (R-CA) and Rep. Lori Trahan (D-MA), the legislation aims to establish a unified federal framework for artificial intelligence governance. Currently in a public feedback phase, the draft outlines stringent requirements for leading AI developers, introduces novel consumer safeguards, and addresses the contentious issue of state versus federal regulatory authority.

Targeting Frontier AI Developers

The draft legislation places its primary regulatory focus on large frontier developers, defined as companies with over $500 million in annual revenue that have trained advanced AI models. This threshold effectively captures industry leaders such as OpenAI, Google DeepMind, and Anthropic. Under the proposed rules, these entities would be required to publish detailed safety frameworks outlining protocols for mitigating catastrophic risks, hire independent verification agencies to conduct biannual compliance audits, and face civil fines of up to $1 million per day for violations.

The State Preemption Debate

One of the most polarizing provisions in the draft is a three-year state law preemption clause. If enacted, the bill would temporarily block state and local governments from enacting or enforcing laws specifically targeting AI model development, though states would retain authority over AI deployment and use. Proponents argue that a unified federal standard is essential to prevent a fragmented regulatory landscape that could stifle American innovation. Critics, however, warn that the preemption could undermine existing state-level consumer protection and child safety regulations, effectively lowering standards to a federal baseline.

Expanding Protections for Workers and Consumers

Beyond corporate compliance, the draft introduces several measures aimed at protecting individuals and strengthening institutional resilience. Key provisions include whistleblower safeguards for AI company employees and contractors who report violations, stricter sentencing guidelines for AI-enabled financial fraud, and the creation of a specific criminal offense for using AI to impersonate federal officials. The bill also mandates federal studies on AI's impact on employment, funds K-12 AI literacy programs, and extends the Cybersecurity Information Sharing Act through fiscal year 2035 to maintain threat intelligence sharing.

Industry Reaction and Legislative Outlook

The tech industry's response has been cautiously mixed. Organizations like NetChoice have endorsed the bill's overarching direction but raised concerns that mandatory audits and data-sharing mandates could inadvertently expose proprietary trade secrets. The legislative path forward remains uncertain. Previous attempts to impose a decade-long pause on state AI regulations were overwhelmingly rejected by the Senate. By shortening the preemption window to three years and coupling it with transparency requirements, lawmakers hope to build broader consensus. Whether the draft can secure enough support before the August congressional recess will depend on intense negotiations and stakeholder feedback in the coming months.