Bipartisan Supermajorities Back AI Regulation in New National Poll
Summary
- • University of Maryland poll finds 79–89% of both Republicans and Democrats support multiple federal AI regulation proposals
- • Large bipartisan majorities favor a dedicated federal AI oversight agency (Dems 85%, Reps 79%) and pre-deployment government testing for AI used in high-stakes decisions like hiring and loan approvals
- • Deepfake labeling draws the strongest bipartisan consensus — 88% of Dems and 89% of Reps support mandatory labeling; 83–85% favor banning campaign deepfakes entirely
- • Poll covered 11 states and 28 competitive House districts ahead of November midterms, giving results direct legislative relevance
Details
Federal AI Agency: Dems 85%, Reps 79%
Bipartisan supermajorities favor creating a dedicated federal agency to monitor AI, anticipate emerging risks, recommend best practices, and enforce regulations
Pre-Deployment AI Testing: Dems 82%, Reps 78%
Americans support requiring AI systems used in critical decisions — hiring, health insurance, loans — to pass government-designed tests before deployment and allowing government audits of deployed systems
Deepfake Labeling: Dems 88%, Reps 89%
The strongest consensus across all tested proposals: bipartisan supermajority supports mandatory labeling of all AI-generated images, audio, and video distributed publicly
Campaign Deepfake Ban: Dems 83%, Reps 85%
Large bipartisan majority supports prohibiting political campaigns from using any AI-generated deepfakes — one of the few areas where Republicans outpace Democrats in support
International AI Rules: Dems 82%, Reps 78%
Americans want the US to work with other nations to establish international rules governing the most powerful AI systems
Autonomous AI Weapons Ban Has Cross-Party Backing
Strong bipartisan support for an international agreement prohibiting AI-powered weapons capable of selecting and firing on targets without direct human control
Survey Targeted 28 Competitive Congressional Districts
Poll was conducted nationally and deliberately in 11 states and 28 competitive House districts ahead of November midterm elections — making results directly actionable for campaigns and legislators
University of Maryland Program for Public Consultation (PPC) national survey, published USA Today, August 2, 2026.
What This Means
A comprehensive University of Maryland survey spanning 11 states and 28 competitive congressional districts reveals that Americans hold a rare area of genuine bipartisan consensus: robust supermajorities of both Republicans and Democrats — in most cases 79–89% — support significant federal AI regulation across every tested proposal. The proposals, drawn directly from bills already introduced in Congress and frameworks from the UN and ICRC, range from creating a dedicated federal AI oversight agency to mandatory deepfake labeling and banning autonomous AI weapons. This polling data gives legislators meaningful political cover to advance AI regulation without fear of partisan backlash — a rare alignment in today's polarized environment. Whether Congress translates this unusually broad public mandate into legislation ahead of November's midterm elections remains the open question.