# Most Registered U.S. Voters Say AI Risks Are Taken Too Lightly By Riza Dagoc Canonical URL: https://www.tokenpost.com/news/regulation/27553 Published: 2026-10-07T16:59:48.000Z Updated: 2026-10-07T16:59:48.000Z Section: Regulation > The poll found 69% support for stricter AI regulation, with majorities also concerned about job losses, data-center growth and uncontrolled effects. Most registered U.S. voters said the Trump administration and Congress have not taken artificial intelligence risks seriously enough, while broad bipartisan support for tighter oversight highlights the technology’s growing political importance ahead of the 2026 midterm elections. The poll of 4,506 U.S. adults, including 3,526 registered voters, was conducted from Sept. 30 to Oct. 5. The margin of error was ±1.7 percentage points for registered voters and ±1.5 points for all adults. Among registered voters, 57% said the Trump administration had not taken AI risks seriously enough or at all. Another 54% gave the same response about Congress. Among registered voters, 69% supported stricter government regulation of AI. Support reached 77% among Democrats, 60% among Republicans and 55% among independent or other voters, showing that tougher rules have backing across party lines. Concerns extended beyond regulation. Some 77% of registered voters were concerned that AI could eliminate jobs in certain professions, while 76% were concerned that the technology could make it harder for younger Americans to obtain well-paying entry-level jobs. The expansion of AI infrastructure also drew opposition. Fifty-six percent supported pausing or limiting construction of new U.S. data centers that support AI and computing services, linking the policy debate to electricity demand and local infrastructure. Military use and long-term safety were additional areas of concern. Sixty-two percent agreed that uncontrolled AI effects could become harmful enough to risk humanity’s future, while 57% said the government should not use AI to determine the target of a military strike. The results come ahead of the Nov. 3, 2026, U.S. midterm elections, when control of Congress is at stake. Registered Democrats were more likely than Republicans to say the administration and Congress had failed to address AI risks, but majorities in both parties favored stricter regulation. The Trump administration has emphasized voluntary action by AI companies while warning that broad federal rules could weaken U.S. competition with China. That approach has featured in the wider policy debate over AI safety and development, including [the administration’s call for AI companies to manage safety risks voluntarily](). ## Links in this article - [the administration’s call for AI companies to manage safety risks voluntarily](https://www.tokenpost.com/news/regulation/25525)