AI adoption is at an all-time high. Trust is at an all-time low. This contradiction creates opportunity.
What Is This?
A national poll from Quinnipiac University released March 30, 2026, surveying nearly 1,400 American adults about their attitudes toward AI.
Professor Chetan Jaiswal summarized it: "51% use AI for research, but only 21% trust AI-generated information most of the time". People use AI as a tool but don't deeply trust its output.
Usage by category: research 51% (up from 37%), writing 28%, work/school 27%, data analysis 27%, image creation 24%. Only 27% have never used AI tools, down from 33% last year.
But sentiment tells a different story. Only 6% are "very excited" about AI. 62% are "not excited," and 80% are "concerned". Negative views have grown compared to last year.
What Changed?
| April 2025 | March 2026 | |
|---|---|---|
| AI research usage | 37% | 51% (+14pts) |
| Never used AI | 33% | 27% (-6pts) |
| Job loss concern | 56% | 70% (+14pts) |
| Own job at risk | 21% | 30% (+9pts) |
| AI is harmful | - | 55% |
Gen Z stands out. They're the most AI-fluent generation but also the most pessimistic about the job market — 81% expect AI to reduce jobs. Professor Triantoro noted: "AI fluency and optimism are moving in opposite directions".
65% of Americans oppose AI data centers in their communities, and 66% say businesses aren't transparent enough about AI use. The same 66% say government AI regulation is insufficient.
Americans are not rejecting AI outright, but they are sending a warning. Too little transparency, too little regulation, too few answers as the technology races ahead.
— Tamilla Triantoro, Quinnipiac University
How to Design for Trust
- Show your sources
Display where AI results come from. Perplexity's inline citations succeeded for this exact reason. - Display confidence scores
"This answer's confidence: 87%" lets users make their own judgment calls. - Offer human review
For important decisions, provide a "human verified" option. Remember Klarna's lesson of replacing 700 people then hiring back. - Market transparency
Proactively disclose AI usage. 66% cite insufficient transparency from businesses. - Default to opt-in
Instead of GitHub's opt-out, follow Anthropic's opt-in approach. Small choices make big trust differences.
For product builders
Users use AI but don't trust it. That's not a crisis — it's an opportunity. "Trustworthy AI" with source citations, confidence scores, and human review options is the competitive edge of 2026.
