Reps. Jay Obernolte (R-CA) and Lori Trahan (D-MA) released a 269-page discussion draft of the Great American AI Act on June 4, the first bipartisan House effort to build a binding federal frontier-AI regime and the first to put real teeth into the preemption fight Washington has been ducking since last July’s 99-1 Senate vote rejecting a sweeping 10-year state-AI moratorium.

The draft is organized into four titles covering frontier governance, workforce, cybersecurity, and R&D, according to Tech Policy Press. It defines “large frontier developers” as companies above $500 million in annual revenue that have trained a frontier model, and subjects them to independent audits every six months with full access to records, personnel, and systems. Civil penalties run up to $1 million per violation per day. Catastrophic risk is defined concretely: foreseeable death or serious injury to more than 50 people, or over $1 billion in property damage.

The preemption is narrower than what the Senate rejected, but pointed. For three years, states can’t pass laws “specifically regulating the development” of frontier models, which would knock out California’s AB 2013 and chunks of SB 53. Consumer protection and civil rights authority is meant to survive, though Trahan herself isn’t sure the text accomplishes that. “If there are genuine ambiguities in the text that could swallow state consumer protection law or civil rights enforcement, that’s a problem I want to fix,” she told Tech Policy Press.

Other provisions: a WARN-style 60-day notice when AI is a “substantial factor” in mass layoffs (per FPF), $300 million over three years for the Commerce Department’s Center for AI Standards and Innovation, and an extension of the Cybersecurity Information Sharing Act of 2015 through fiscal 2035. Four co-sponsors signed on: Scott Franklin, Suhas Subramanyam, Erin Houchin, and Scott Peters. Feedback goes to [email protected] before formal introduction.

The lobbying split is already visible. The Alliance for Responsible Innovation is running Massachusetts ads pressuring Trahan against any state-law carve-out. The Alliance for Secure AI’s Brendan Steinhauser put the counter-position plainly: “A national AI standard should protect at least as much as it preempts.” That sentence is the entire fight.

Sources

Sources