Reps. Jay Obernolte (R-Calif.) and Lori Trahan (D-Mass.) on Thursday released a 269-page discussion draft of the Great American Artificial Intelligence Act of 2026, a bipartisan attempt to settle the federal-versus-state question on AI governance by trading a three-year freeze on state AI development laws for a binding federal audit regime on the largest model developers.

The bargain is the point. States would lose authority to legislate how advanced systems are built but keep their power over how those systems are used and deployed. California’s AB 2013, which requires developers to publicly summarize training data, would be preempted, along with a portion of AB 1008, according to an accompanying document from Trahan’s office reported by Roll Call.

In exchange, “large frontier developers,” defined as those with more than $500 million in gross revenue in the prior calendar year, would face independent third-party audits every six months and have to publish frontier AI frameworks. Civil penalties run up to $1 million per violation per day. The bill defines “catastrophic risk” as foreseeable risk of death or serious injury to more than 50 people, or more than $1 billion in property damage.

The draft also codifies a Center for AI Standards and Innovation inside the Commerce Department, authorized at $100 million per year for fiscal 2027 through 2029, and extends the Cybersecurity Information Sharing Act of 2015 through fiscal 2035. Reps. Scott Franklin (R-Fla.), Suhas Subramanyam (D-Va.), Erin Houchin (R-Ind.) and Scott Peters (D-Calif.) joined the release.

Not everyone is buying the trade. Brendan Steinhauser, chief executive of the Alliance for Secure AI, said the federal framework “does not justify preempting states’ ability to pass their own AI safeguards.”

The lawmakers are soliciting public feedback before formal introduction, which is itself a tell: the preemption fight, deferred all year in the Senate, has finally arrived in legislative text.

Sources