Reuters/Ipsos poll finds most US voters say Washington isn't taking AI risk seriously

A Reuters/Ipsos poll found that most US voters think the Trump administration and Congress are not taking the risks of AI seriously enough: 57% of registered voters, including a third of self-identified Republicans, said the administration had not taken AI risks seriously enough or at all, and 54% said the same of Congress. The poll was fielded online over six days ending 5 October, surveying 4,506 US adults including 3,526 registered voters, with a margin of error of 2 percentage points.

Sixty-two per cent of respondents said AI could get out of control and threaten the future of humankind — roughly unchanged from a Reuters/Ipsos poll taken in August 2025. Eighty-four per cent saw AI as a threat to American workers, a higher share than said the same of illegal immigration, foreign-worker competition or large corporations. A bipartisan majority, 80% of Democrats and 61% of Republicans, said they supported stricter government regulation of AI, and 56% backed limits on building new data centres, including 69% of Democrats and 43% of Republicans. The poll also found 57% opposed letting the US government use AI to select the target of a military strike, up from 49% in the August 2025 survey.

Reuters reported that Trump had announced, the week before the poll’s release, that executives from six leading AI developers — Nvidia, SpaceX, OpenAI, Anthropic, Meta and Google — had agreed to voluntary safety principles including outside auditing, with no stated consequence for non-compliance; a White House official pointed to a new administration task force intended to keep “America continues to lead the world” on AI. Daniel Kokotajlo, a former OpenAI employee who left the company in 2024 over safety concerns, told Reuters that control of AI “is an important concern but it’s not top of mind for most Americans, nor should it be,” arguing voters were more immediately worried about jobs and safety.

The result echoed a Quinnipiac poll published a week earlier, which had found 73% of Americans concerned that advanced AI could threaten human survival and a majority wanting development slowed or stopped — part of a broader run of 2026 surveys showing hardening public unease with AI, now surfacing as a campaign issue ahead of the November midterm elections.