A new Quinnipiac University poll points to a clear public preference on AI development: Americans are not asking for acceleration. Most respondents want the pace reduced or stopped entirely until safety has been verified.
The findings show a public mood centered on caution, independent oversight, and skepticism toward company leadership. The poll surveyed 1,202 US adults from September 24 to 27, with a margin of error of +/- 3.5 percentage points.
Most respondents want AI development slowed or paused
The headline result is direct. According to the poll, 47 percent want to reduce the pace of AI development, while 30 percent want a full stop until safety has been verified. Only five percent want faster development.
That split matters because it shows that concern is not limited to a small group calling for a total halt. The larger finding is broader: many Americans want the technology to move more carefully, even if they do not all support stopping it outright.
In plain terms, the poll suggests that speed is not the public's main priority. Safety verification is. The strongest combined signal is that respondents want AI companies and policymakers to treat AI risk as something that must be addressed before progress continues at the current pace.
Independent safety standards have broad support
The poll also found strong support for outside guardrails. 86 percent support the independent safety standards currently under debate for AI companies, even if they slow progress.
That result is important because it connects two ideas that are often treated as tradeoffs: safety and progress. Respondents were not only asked whether they liked safety standards in the abstract. The source states that support remains high even when those standards could slow progress.
For AI companies, that creates a clear message from the poll: public acceptance may depend less on promises of fast innovation and more on whether safety standards are independent. The word independent matters here because the survey results also show limited trust in AI company leaders.
Trust in AI company leaders is low
The poll found that 74 percent have little or no trust in the leaders of AI companies. That finding helps explain why independent standards received such high support.
If the public trusted company leaders to manage the risks on their own, outside standards might seem less urgent. But the poll points in the opposite direction. Respondents appear to want safeguards that do not rely only on internal judgment from the companies building AI systems.
"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.
That quote captures the larger pattern across the findings. The issue is not only whether AI development should continue quickly or slowly. It is also who gets trusted to decide when AI systems are safe enough.
Survival concerns shape the debate
The poll found that 73 percent worry that AI could eventually threaten humanity's survival. That is one of the clearest signals in the survey, because it shows that concern about AI is not limited to everyday inconvenience or industry disruption.
The source does not break down what respondents meant by that worry, so the safest reading is narrow: a large share of those surveyed see possible future stakes that are severe enough to justify caution. Combined with support for independent safety standards, the result suggests that many Americans want AI risk treated as a serious public issue.
The numbers also help explain why a small share wants faster AI development. If most respondents are worried about survival-level risk, faster development without verified safety may not look attractive to them.
What the poll signals for AI companies
The Quinnipiac University poll does not describe a public that is simply rejecting AI. It describes a public that wants AI development to be slowed, checked, or paused until safety has been verified.
The practical message is consistent across the results:
- 47 percent want to reduce the pace of AI development.
- 30 percent want a full stop until safety has been verified.
- Only five percent want faster development.
- 86 percent support independent safety standards currently under debate for AI companies, even if they slow progress.
- 74 percent have little or no trust in the leaders of AI companies.
- 73 percent worry that AI could eventually threaten humanity's survival.
Taken together, those findings point to a public that wants stronger proof, clearer safeguards, and less reliance on company assurances. For the AI industry, the challenge is not only technical progress. It is earning confidence in how that progress is governed.