Only 16% of Americans expect positive AI outcomes over 20 years, Pew Research finds

American skepticism about AI’s societal future runs deep

A new Pew Research survey reveals a stark disconnect between AI’s expanding presence in American life and public confidence in its long-term trajectory. Despite widespread adoption of AI tools—particularly ChatGPT, which 44% of U.S. adults now report using—only 16% of Americans believe artificial intelligence will have a positive impact on society over the next two decades, according to the research. That optimism stands in sharp contrast to the 40% who expect negative outcomes, leaving a substantial 44% expressing neutral sentiment or uncertainty.

The pessimism extends beyond aggregate sentiment. According to Pew Research, a decisive majority believes the pace of AI advancement is unsustainable: nearly two-thirds of respondents said development is happening too quickly. This speed anxiety pairs with institutional distrust—59% do not believe companies will develop AI responsibly, and 67% hold little faith that U.S. government regulation will meaningfully constrain the technology’s risks.

Generational and gender divides in AI adoption and trust

The survey identifies two demographic groups as particularly skeptical. Americans under 30 register the lowest confidence: just 14% of this cohort expect positive outcomes, dipping below the national average by 2 percentage points. Women also express greater caution than men; while chatbot usage is rising among both genders, men are more enthusiastic and adopt non-ChatGPT platforms at higher rates (27% of men versus 20% of women use AI chatbots daily).

The gender gap widens when examining specific products. According to Pew Research, men are substantially more likely to use Copilot and Grok, while adoption of ChatGPT is roughly equal across genders. The chatbot hierarchy reflects OpenAI’s dominance: Gemini captures 24% of users, Copilot 17%, Meta AI 14%, Grok 8%, Claude 6%, and Character.ai 3%.

Usage patterns and adoption barriers

Despite skepticism, roughly half of all Americans have integrated AI into daily routines. About 25% use chatbots daily, primarily for research and work, the survey reports. A far larger cohort—60%—consumes AI-generated summaries routinely, largely through unavoidable feeds on platforms like Google. Fitness and diet recommendations powered by AI attract far smaller audiences.

The non-adopter cohort tells a revealing story: approximately 50% of Americans do not use AI, concentrated heavily among adults 65 and older (nearly 75% of this age group). Those abstaining cite lack of interest and no intention to engage—suggesting resistance rather than mere inexperience.

Why This Matters

The sentiment gap between AI adoption and trust creates policy pressure. Legislators and regulators now operate against a backdrop where only 16% of constituents expect benefits, weakening the case for pro-innovation frameworks. For companies like OpenAI and Google, the findings underscore a legitimacy challenge: raw user numbers mask underlying skepticism about motives and safeguards. The persistence of this doubt—despite ChatGPT’s penetration to 44% of the population—suggests that marketing and product expansion alone will not resolve public concerns. Expectations management and transparent governance frameworks will likely become competitive differentiators as AI adoption reaches saturation and skepticism hardens into voting behavior.