US Survey Reveals Strong Bipartisan Support for AI Regulation
A recent survey reveals a striking bipartisan consensus among U.S. residents on the urgent need for government regulation of artificial intelligence, a finding that directly challenges the potential limitations posed by the federal government’s newly unveiled AI Action Plan. For years, policymakers have grappled with the complex question of how to oversee AI development and deployment. During the previous administration, a shift in strategy emerged, culminating in a proposed budget reconciliation bill provision that sought to restrict states’ ability to regulate AI for a decade. This measure, however, faced broad bipartisan opposition and was ultimately withdrawn. Now, the current administration’s AI Action Plan introduces a new dynamic, potentially compelling states to choose between enforcing their existing state-level AI regulations or risking the loss of federal funding.
The comprehensive survey, conducted by the Program for Public Consultation (PPC) at the University of Maryland’s School of Public Policy, indicates that a significant majority of both Republicans and Democrats endorse various proposed regulations for AI systems. Steven Kull, Director of the PPC, articulated the public’s sentiment, stating, “Clearly, Americans are seriously concerned about the current and potential harms from AI. And while the public is wary of government regulation, they are clearly more wary of the unconstrained development and use of AI.” While respondents acknowledged that regulation might influence the pace of innovation, the overwhelming majority expressed support for five key proposals currently under consideration by the U.S. Congress. These include three distinct regulations targeting the use of AI in decision-making processes, two focused on deepfakes, and a call for the U.S. to lead in establishing an international treaty to ban autonomous AI-powered weapons.
Delving into specific areas, the survey found strong support for regulating AI used in critical decision-making contexts, such as hiring or healthcare. A substantial majority—84 percent of Republicans and 81 percent of Democrats—favored a requirement for AI programs to pass a government test before being implemented. Similarly, over eight in ten Republicans and nearly eight in ten Democrats supported empowering the government to audit AI programs already in use and compelling companies to rectify any identified issues. Given that the effectiveness of AI systems hinges on the quality of their data, 74 percent of all respondents also backed a proposal requiring companies to disclose how their decision-making AI models are trained, upon government request.
The proliferation of deepfakes, or AI-manipulated imagery, poses escalating security risks, a concern highlighted by a recent incident when the president shared a deepfake video in July. The survey revealed that a large majority of respondents—83 percent of Republicans and 78 percent of Democrats—advocated for prohibiting the use of deepfakes in political advertisements. Furthermore, 80 percent of all respondents supported a mandate that all publicly shared deepfakes be clearly labeled. One respondent powerfully encapsulated the public’s concern, stating, “Purposefully putting out misleading videos and pictures to the public as the truth will be the downfall of any democratic society.”
Perhaps the most extreme risk posed by AI is its potential use in autonomous weaponry. The United Nations Secretary-General has urged nations to ban the development of lethal autonomous weapons, a stance the U.S. government has thus far opposed, even as AI increasingly reshapes military strategies globally. Despite the U.S. government’s position, the survey found robust public backing for an international ban: 75 percent of Republicans and 77 percent of Democrats favored the U.S. establishing an international treaty to prohibit the development of such weaponry.
The survey, which was offered in both English and Spanish, engaged 1,202 adults nationally from July 30 to August 7. These findings build upon a previous PPC survey conducted in March 2024, collectively underscoring a clear and unified public demand for proactive government oversight in the rapidly evolving landscape of artificial intelligence.