Most Americans Now See AI as Bad for Society and Want More Regulation, Poll Finds
Key Takeaways
- •Fifty-seven percent of US adults say artificial intelligence is negatively influencing society, with nearly one-third describing the technology as extremely or very harmful.
- •Sixty-seven percent of respondents believe the federal government is not doing enough to regulate AI, including 80% of Democrats, 66% of independents, and 54% of Republicans.
- •Two-thirds of Americans disapprove of building new data centers in the United States, making AI's physical infrastructure a tangible issue in local communities.
- •The findings echo earlier 2026 research, including an Annenberg survey in which only 17% of respondents expected AI to have a positive impact over the next decade and 65% said the government had done too little to regulate it.
- •Seventy percent of participants in a SurveyMonkey study consider AI policy a vital voting issue, and 67% disapprove of President Trump's handling of AI technology.
A new poll from The Washington Post, ABC News, and Ipsos has found that 57% of US adults believe artificial intelligence is having a negative influence on society. The same survey shows that 67% of respondents think the federal government is not doing enough to regulate how the technology grows and develops. The findings arrive as debate over AI oversight continues in Washington and state capitals.
What the Poll Found
The survey ran from October 9–13, 2026, and reached 2,545 US adults. Its headline finding is the 57% who view AI as bad for society overall, with nearly one-third of respondents describing the technology as either “extremely” or “very” harmful.
Frustration over regulation crosses party lines. Among Democrats, 80% said Washington is falling short on AI oversight, while independents landed at 66% and Republicans at 54% — a breadth of agreement that spans the political spectrum.
The poll also found that two-thirds of Americans disapprove of building new data centers in the United States. That finding speaks to AI’s physical footprint: data centers are the infrastructure behind AI services, and their construction is becoming a tangible local issue in communities across the country.
Not a One-Off Result
The findings are consistent with earlier research. A survey from the Annenberg Public Policy Center, conducted between February and March 2026, found that only 17% of respondents expected AI to have a positive impact over the next decade, while 42% predicted a negative one.
About two-thirds of respondents, 65%, said the government has done “too little” to regulate AI, and that view was held by 53% of Republicans in the survey.
A separate SurveyMonkey study found that 70% of participants consider AI policy a vital voting issue, while 74% support stricter regulations. Studies from Gallup and SurveyMonkey also show majority support for safety rules even if those rules slow AI development. For an industry pushing rapid expansion, that describes a public willing to trade speed for guardrails.
The 2026 polling also shows 67% disapproving of President Trump’s handling of AI technology — a verdict that carries political weight with seven in ten Americans calling AI policy a vital voting issue.
Why People Are Worried
The research points to three recurring concerns: job loss, safety, and the sheer pace of development. Whether that breadth of unease — and the bipartisan appetite for oversight behind it — translates into federal legislation, or leaves AI governance to state-level rules for now, is the question to watch.
