Reps. Jay Obernolte (R-Calif.) and Lori Trahan (D-Mass.) dropped a 269-page discussion draft of the Great American AI Act on June 4, the most ambitious bipartisan federal AI framework yet floated in the House, and the first one carrying a preemption clause sharp enough to void California’s training-data and watermarking statutes. Co-sponsors Scott Franklin (R-Fla.), Suhas Subramanyam (D-Va.), Erin Houchin (R-Ind.), and Scott Peters (D-Calif.) signed on, lending it the rare patina of cross-aisle seriousness.

The substantive guts are real. Frontier developers with more than $500 million in prior-year gross revenue would publish AI frameworks disclosing “catastrophic risk,” which Roll Call reports the draft defines quantitatively as foreseeable risk of 50 or more deaths or $1 billion in property damage. The bill codifies the Commerce Department’s Center for AI Standards and Innovation, authorizing $100 million per year across fiscal 2027–2029, per FedScoop. The Future of Privacy Forum notes mandates for critical safety incident reporting, whistleblower protections, and a 60-day notice when AI is a “substantial factor” in mass layoffs. It also extends the Cybersecurity Information Sharing Act of 2015 through fiscal 2035.

Then comes the bomb. A three-year sunset preemption clause would freeze state laws “specifically regulating the development” of AI models, voiding California’s AB 2013 training-data disclosure law and part of SB 942 on watermarking, according to Trahan’s office. Deployment laws survive; development laws don’t.

The blast radius was immediate. Brendan Steinhauser of the Alliance for Secure AI said the federal framework “does not justify preempting states’ ability to pass their own AI safeguards.” Brad Carson of Americans for Responsible Innovation called it a “generational mistake.”

Sponsors are taking written feedback at [email protected]. No introduction date is set. Influential Democrats have already denounced the draft, which means the bipartisan label is doing more rhetorical work than legislative.

Sources

Sources