Reuters/Ipsos Poll: Most US Voters Say Washington Is Not Taking AI Risks Seriously

Computer-generated English audio

AI regulation becomes a voter concern

A Reuters/Ipsos survey released on October 7, 2026, found widespread concern among American voters about the risks of artificial intelligence and whether elected officials are responding adequately. The findings suggest that AI is becoming more than a debate among technology companies and researchers. For many people, it is increasingly connected to jobs, public safety and decisions made by government. The poll arrives as candidates prepare for the November midterm elections, when economic security and confidence in institutions are already major issues.

What the poll found

According to Reuters, 57% of registered voters surveyed said President Donald Trump’s administration was not taking AI risks seriously enough, while 54% said the same about Congress. The survey included 4,506 US adults, of whom 3,526 were registered voters, and was conducted over six days ending October 5. These figures measure respondents’ views of government action; they do not establish that a particular policy is effective or ineffective. Polling results should be interpreted with attention to wording, methodology and sampling uncertainty.

Jobs and economic anxiety

One important source of concern is the possibility that AI systems will automate tasks currently performed by people. Technology can also create new jobs and improve productivity, but the benefits and costs may not be shared evenly. Workers in administrative, customer-service and creative roles may encounter changes sooner than people in some other occupations. The pace of adoption varies by employer and industry. Voters’ fears therefore reflect uncertainty about future opportunities as much as confirmed job losses. A responsible policy debate should distinguish between forecasts, company announcements and observed labor-market outcomes.

Concerns about safety and misuse

AI can help analyze information and support useful services, but systems can also generate inaccurate material or be misused. Risks discussed in public debate include deceptive content, cyberattacks, privacy violations and unreliable automated decisions. Not every application poses the same level of danger. A tool that summarizes documents is different from software used to make consequential decisions about healthcare or security. Regulation may need to reflect these differences rather than treating all AI products identically. Transparency about testing, oversight and responsibility can help users understand where safeguards are necessary.

Data centers and local communities

The expansion of AI computing has increased attention on data centers, which require electricity, cooling and physical infrastructure. Communities considering new facilities may weigh construction jobs and tax revenue against electricity demand, water use and effects on nearby residents. The Reuters/Ipsos findings suggest that some voters want a stronger role for government in evaluating those trade-offs. Local conditions matter: the impact of a facility depends on the power grid, water supply, project design and environmental rules. Broad national claims cannot substitute for project-specific evidence.

A debate that crosses party lines

Public concern about AI does not fit neatly into a single political party’s agenda. People with different views on taxation, business and government may nevertheless share questions about job displacement or technology safety. At the same time, they may disagree sharply about the right response. Some favor stronger federal rules, while others worry that burdensome regulation could slow innovation or disadvantage smaller firms. The poll highlights a public demand for attention, but it does not resolve how lawmakers should balance safety, competition and economic growth.

The challenge for Congress

Federal lawmakers face difficult choices about which AI uses require specific rules and which can be addressed under existing laws. Issues include consumer protection, privacy, discrimination, cybersecurity and the government’s own deployment of automated systems. Legislation can take time to negotiate, especially when industries and public-interest groups disagree about enforcement. Voluntary commitments from companies may help establish practices, but critics argue that enforceable standards are needed for high-risk applications. Supporters of flexible approaches counter that overly rigid rules can become outdated as technology changes.

What the numbers do not prove

A survey is a snapshot of opinion, not a referendum or a forecast of election results. Respondents may express concern about AI while prioritizing inflation, healthcare or other issues when they vote. The share of voters dissatisfied with government action also does not necessarily equal the share supporting any one proposed law. It is important to distinguish registered voters from all adults and to avoid comparing polls that ask different questions as though they were identical. The poll offers valuable evidence of public sentiment but cannot predict the future of technology policy.

Sources and next developments

The original findings were reported by Reuters on October 7. Readers interested in federal policy can follow Congress.gov for introduced legislation and official bill status. For broader US political and technology coverage, visit NewsNationOnline. The next questions are whether candidates propose concrete AI safeguards, how agencies respond to specific risks and whether voters view those proposals as credible. Any claim that a bill has passed should be verified through the official legislative record.

What voters can look for in campaign proposals

Campaign promises about artificial intelligence can sound similar even when they describe very different policies. Voters can ask whether a candidate supports independent testing for systems used in sensitive settings, disclosure when synthetic media appears in political advertising, or clear procedures for challenging automated decisions. Another useful question is who would enforce a proposed rule and whether smaller businesses would receive practical guidance. A proposal with measurable responsibilities and a realistic implementation plan is easier to evaluate than a broad pledge to make AI safe. Citizens can also compare the positions of federal candidates with decisions made by state governments, which may regulate particular uses of technology separately.

Companies developing and deploying AI also face questions about accountability. Publicly available documentation, incident reporting and meaningful opportunities for human review can make claims about safety easier to assess. None of these measures eliminates every risk, and their value depends on the application. The debate is likely to continue well beyond the election as new products reach workplaces, schools and public services. For now, the Reuters/Ipsos survey is evidence that many Americans expect elected leaders to engage with these questions rather than leave them entirely to technology companies.

Related reporting

For additional context on AI policy and the federal government’s response, see our NewsNationOnline artificial intelligence coverage. Readers can compare the policy debate with the official records linked above rather than relying on campaign statements alone.

Survey methodology and additional findings

The poll was conducted from September 30 to October 5, 2026, among 4,506 US adults, including 3,526 registered voters. Reuters reported a margin of error of approximately two percentage points. The survey found that 84% of registered voters regarded AI as a threat to American workers, while 56% supported limits on building AI data centers. Those results describe respondents’ opinions, not established predictions of economic damage or the effectiveness of specific policies. Readers can consult the Ipsos survey information and methodology alongside the original Reuters report.

Related US technology coverage

For additional context, read NewsNationOnline’s report on AI development and robotics research. This related coverage provides background and should be read alongside the dated official sources in this article.