Real Action on Artificial Intelligence
Last week, the White House released an Executive Order (EO) Promoting Advanced Artificial Intelligence (AI) Innovation and Security. The new EO is the latest in a collection of orders focused on AI, and shares the same stated goal: making sure that the United States leads the world in AI. With every new EO, it becomes increasingly evident that no executive mandate can fill Congress’s responsibility to act on behalf of American AI leadership and innovation.
Leaked drafts of the new order were less than promising, and rumors swirled about a Food and Drug Administration (FDA) style approach to regulating AI development. The consequences would have been dire for American innovation and would have vested immense power into the hands of federal agencies for determining the direction of American technology. The draft EO was amended and released, and corrects some of the more glaring issues of the first. However, as Neil Chilson, head of AI policy at the Abundance Institute, pointed out in his analysis on X, the EO still contains gaps that could be used to pick winners and losers. “The EO may improve on the status quo. But it deserves close and ongoing scrutiny, especially from Congress, which bears the ultimate responsibility for writing the rules that govern AI.”
Executive orders and state-by-state legislation both have their limitations. Particularly with a technology progressing as rapidly as AI, an EO can only offer so much detail and a state law cannot extend into another state without violating principles of interstate commerce. Federalism prescribes these matters to Congress, and in doing so, eases the friction between scope and specificity.
Congress is not ignorant to this need. Last week, a draft of bipartisan federal legislation from U.S. Representatives Jay Obernolte (R-CA) and Lori Trahan (D-MA) was released for discussion. The “Great American Artificial Intelligence Act of 2026” is a sprawling 296 pages. The proposal does preempt state laws, answering the increasingly urgent problem of a patchwork that leaves innovation dead in the water. But the draft is far from perfect, and aspects of it mirror some of the most zealous state-based regulations by imposing strenuous criteria for compliance and assurance—the process by which models will be approved for use. The Great American Artificial Intelligence Act, in its current form, also does not solve some of the most pressing questions regarding how state laws that impact interstate commerce should be fixed, leaving this area fairly ambiguous. These flaws are not a reason to despair, because the legislative process in Congress ideally operates in a refining manner. The draft can be amended and improved with a flexibility not available to other forms of regulation or frameworks.
AI is not passive. The technology is developing rapidly, growing the economy, changing the landscape of different industries, and becoming increasingly vital to national security and leadership. AI policy cannot be passive, either. A smart national public policy standard will create space for the free market to refine the way the technology is developed and used, and Congress, with its representatives elected by the American people, is uniquely positioned to chart the course.
