- Pew Research data cited in the report found only 10% of Americans said they are more excited than concerned about AI in daily life, compared with 56% of AI experts who expect a positive national impact over the next 20 years.
- The gap is widest on employment: 73% of experts expect AI to have a positive effect on how people work, against 23% of the general public—a 50-point spread.
- 64% of Americans believe AI will lead to fewer jobs over the next 20 years; U.S. public trust in government to regulate AI responsibly stands at 31%, the lowest of all nations Ipsos surveyed.
- Globally, the share of respondents who said AI offers more benefits than drawbacks rose from 55% in 2024 to 59% in 2025, but the share describing AI as making them “nervous” also grew, from 50% to 52%.
What Happened
Stanford University’s AI Index released its 2026 annual report on April 13, documenting a pronounced divergence between how AI researchers and industry experts assess the technology’s societal impact and how the general public does. The report, covered by TechCrunch reporter Sarah Perez, aggregates survey data from Pew Research, Ipsos, and Gallup to track sentiment across employment, healthcare, economic growth, and government regulation. Its release coincided with social media reactions to a recent attack on OpenAI CEO Sam Altman’s home, which drew comments that AI insiders said revealed the depth of public hostility they had not previously registered.
Why It Matters
The expert-public divide documented by Stanford carries practical weight because AI deployment and policy decisions are largely made by the optimistic expert cohort, not the skeptical public. The findings are reinforced by a Gallup poll cited in the report, which found that Gen Z—roughly half of whom use AI daily or weekly—is growing less hopeful and more angry about the technology despite high usage rates. Caroline Orr Bueno, Ph.D., commenting on the report’s release, wrote on X: “I think a lot of AI leaders are just out of touch with normal people and don’t realize that fears of Skynet are not what is primarily driving anti-AI sentiment. Most people are way more concerned with their paycheck and the cost of utilities.”
Technical Details
Pew Research data published in March 2026, cited extensively in the Stanford report, measured the expert-public gap across three domains. On employment, 73% of AI experts expected a positive impact on how people do their jobs, compared with 23% of the U.S. general public—a 50-point gap. On medical care, 84% of experts anticipated benefit versus 44% of the public. On the economy, expert optimism stood at 69% against 21% among the public.
Ipsos data in the same report placed U.S. public trust in government to regulate AI responsibly at 31%—the lowest of any nation surveyed. Singapore ranked highest at 81%. Within the U.S., 41% of respondents said federal AI regulation would not go far enough, while only 27% said it would go too far. A separate data point showed that 64% of Americans believe AI will lead to fewer jobs over the next 20 years, while experts surveyed were considerably less pessimistic on that question.
Who’s Affected
Workers in sectors with high AI automation potential—including administrative roles, customer service, and knowledge work—are the demographic most represented in the skeptical public cohort, based on survey data about job and economic concerns. AI companies including OpenAI, Google, and Anthropic, whose public communications have emphasized long-term benefits, now operate in an environment where only 10% of Americans report being more excited than concerned about AI in daily life. Policymakers considering federal AI legislation face an electorate that, by a 14-point margin, believes current regulatory efforts will fall short of what is needed.
What’s Next
Stanford’s AI Index is used as a reference document in Congressional hearings and by regulatory bodies including the European Commission, which is in the process of implementing the EU AI Act. The Pew Research survey underlying several of the report’s core comparisons was published in March 2026 and covers U.S. demographic breakdowns not fully detailed in the summary reporting. Stanford has not announced a public forum or follow-up release tied to the 2026 findings.
