Americans Worry About AI Harming Humanity, Advocate Slowing Development Rather Than Halting It

Americans Favor Controlled Pace for AI Development Amid Competitiveness and Safety Concerns A CBS News/YouGov poll conducted September 16 to 18, 2026, reveals a nuanced national consensus on artificial intelligence: while Americans recognize the technology potential for medical and scientific advancement, prevailing concerns over safety, misuse, and employment disruption drive a strong preference for deliberate development pacing. The survey of 2,051 U.S. adults, with a margin of error of ±2.5 percentage points, indicates that most citizens oppose halting AI entirely but advocate for regulated slowdowns to mitigate emerging risks. The findings highlight a pragmatic balance between innovation and oversight. Respondents widely acknowledge that the United States must advance AI capabilities to maintain strategic competitiveness, particularly relative to China. This geopolitical imperative is heavily offset by safety anxieties. Majorities fear that autonomous AI systems could independently cause harm, with an even larger share worried about malicious human application. Consequently, the public has largely dismissed the feasibility of corporate self-regulation, instead demanding robust government intervention to restrict deployment patterns and enforce accountability frameworks. Economic displacement remains a primary catalyst for public pessimism. Individuals anticipating significant job losses exhibit heightened skepticism across broader AI metrics, reinforcing a prevailing perception that the technology currently yields more societal friction than net benefit. Notably, while many workers expect AI to streamline daily tasks and improve individual productivity, they simultaneously predict a net reduction in overall employment opportunities. This operational optimism clashes with macroeconomic anxiety, explaining why positive outlooks on specific applications remain overshadowed by cautious sentiment. Demographic analysis shows minimal divergence along age or partisan lines, suggesting a cross-cutting national consensus that prioritizes containment over acceleration. The data points to a clear policy expectation: federal and state authorities should emphasize safety mandates and development restrictions rather than industry subsidies or promotional campaigns. For technology developers and regulators, the survey outlines a precise mandate. Public acceptance hinges on demonstrable safety controls, transparent governance, and proactive workforce transition strategies. Industry leaders must align development roadmaps with these measured expectations, ensuring that continued innovation in strategic and scientific sectors proceeds within strictly bounded operational and ethical frameworks. As AI integration expands, securing societal consent will require prioritizing risk mitigation and regulatory clarity alongside technical progress.