Most Americans Want AI Development Slowed or Halted, Quinnipiac Poll Finds
A new Quinnipiac University survey reveals that over three-quarters of Americans want artificial intelligence development slowed down or stopped completely, citing deep distrust in tech leadership and existential safety concerns.
A vast majority of Americans favor putting the brakes on artificial intelligence progress, according to a recent Quinnipiac University poll. The findings reveal pervasive skepticism toward the technology's societal benefits and an overwhelming demand for independent regulatory oversight.
Broad Support for a Cooldown
The survey found that 47 percent of respondents want to slow down the current pace of AI development, while another 30 percent favor a complete stop until safety can be verified. In contrast, only 5 percent of respondents said they want development to move faster.
Public hesitation is fueled by a severe lack of confidence in tech executives. Nearly three-quarters of respondents—74 percent—reported having little or no trust in the leaders of AI companies.
"Taken together, these results suggest that the public wants safeguards that do not depend simply on taking company leaders at their word," said Chetan Jaiswal of Quinnipiac University.
Overwhelming Demand for Independent Guardrails
Concerns about safety extend directly into policy preferences. An overwhelming 91 percent of participants agreed that guardrails for AI systems in the United States are important, and 86 percent expressed support for independent safety standards currently under debate for AI companies, even if those rules slow technological progress.
Existential risks also weigh heavily on the public mind. Roughly 73 percent of respondents expressed worry that AI could eventually threaten humanity's survival. When asked about specific threats, 52 percent stated they are more concerned about humans misusing AI than about autonomous AI operating on its own.
Local Resistance and Global Competition
The poll highlighted a notable friction between national competitiveness and daily life. While 69 percent of respondents agreed that it is important for the United States to keep pace with China on AI development, 53 percent believe AI will ultimately cause more harm than good in everyday life.
That skepticism extends into physical infrastructure. As tech firms rapidly build out computational capacity across the country, 72 percent of Americans said they would oppose having an AI data center built in their own community.
The Quinnipiac University poll surveyed 1,202 U.S. adults between September 24 and September 27, carrying a margin of error of plus or minus 3.5 percentage points.
