- A Quinnipiac University poll of 1,397 U.S. adults found that 76 percent trust AI “rarely” or “only sometimes,” while adoption continues rising — only 27 percent have never used AI tools, down from 33 percent a year earlier.
- Seventy percent of Americans believe AI will decrease job opportunities, up from 56 percent in April 2025, with Gen Z the most pessimistic at 81 percent.
- Fifty-five percent say AI will do more harm than good in daily life, up from 44 percent a year ago, and 74 percent believe the government is not regulating AI enough.
- AI usage for research jumped to 51 percent from 37 percent, and data analysis usage rose to 27 percent from 17 percent over the same period.
What Happened
A Quinnipiac University poll published on March 30, 2026, revealed a widening gap between Americans’ use of AI tools and their trust in the results. The survey of 1,397 U.S. adults, conducted March 19-23 with a margin of error of 3.3 percentage points, found that 76 percent trust AI “hardly ever” or “only sometimes,” while just 21 percent trust it “most of the time” or “almost all of the time.” TechCrunch’s Rebecca Bellan reported on the findings.
“The contradiction between use and trust of AI is striking,” said Chetan Jaiswal, Ph.D., one of the poll’s analysts. “Americans are clearly adopting AI, but they are doing so with deep hesitation, not deep trust.”
Why It Matters
The poll captures a paradox that has significant implications for AI companies and policymakers. Americans are rapidly integrating AI into work and personal tasks — only 27 percent reported never having used AI tools, down from 33 percent in April 2025. Yet their confidence in the technology’s outputs is declining simultaneously. Fifty-five percent now say AI will do more harm than good in daily life, up from 44 percent a year ago.
The trust deficit is especially pronounced among younger Americans who use AI the most. Tamilla Triantoro, Ph.D., noted that “younger Americans report the highest familiarity with AI tools, but they are also the least optimistic about the labor market.” Gen Z respondents were the most pessimistic about job impacts, with 81 percent believing AI will decrease opportunities, compared to 66 percent of Baby Boomers.
Technical Details
AI usage increased across every category measured. Research use jumped to 51 percent from 37 percent in April 2025. Data analysis rose to 27 percent from 17 percent. Image creation climbed to 24 percent from 16 percent. Twenty percent of respondents reported using AI for medical advice, and 15 percent for personal advice. Among employed adults, 32 percent reported using AI at work, and 30 percent expressed concern about their jobs becoming obsolete — up from 21 percent a year earlier.
The survey also tested specific trust boundaries. Eighty-one percent of respondents preferred a combination of AI and human doctors for reading medical scans, even if AI were proven more accurate. Brian O’Neill, Ph.D., observed that “it’s telling that most people would still want a human involved in reading medical scans even if it were proven the AI tool was more accurate.” Eighty percent said they would be unwilling to have an AI supervisor at work.
Who’s Affected
The findings present challenges across the AI industry. Seventy-six percent of respondents said businesses are not transparent enough about their AI use, and 74 percent believe the government is not regulating AI enough, up from 69 percent a year ago. On political content, 45 percent favored requiring disclosure of AI-generated material while 38 percent supported banning AI use in political content entirely.
Attitudes toward AI infrastructure also skewed negative. Sixty-five percent opposed building AI data centers in their communities, citing electricity costs (72 percent), water use (64 percent), and noise (41 percent). Among supporters, job creation (77 percent) and tax revenue (53 percent) were the primary motivations.
What’s Next
Jaiswal summarized the overall sentiment: “Americans are not rejecting AI outright, but they are sending a warning: Too much uncertainty, too little trust.” The survey results suggest that increasing adoption alone will not build public confidence. AI developers face pressure to improve transparency and explainability, while regulators face growing public demand for oversight frameworks that have yet to materialize at the federal level.
