TL;DR: The latest Quinnipiac University national poll reveals that 76% of Americans trust AI-generated information only some of the time or hardly ever, highlighting a severe trust deficit. Public concern is rising, with 55% now expecting AI to do more harm than good in their daily lives, up from 44% in April 2025. To succeed, B2B technology leaders must address these concerns by designing collaborative human-in-the-loop systems rather than pushing for fully automated solutions.

A new national poll by Quinnipiac University, conducted in partnership with their School of Computing & Engineering and School of Business, shows that public skepticism toward artificial intelligence is intensifying. Even though hands-on adoption is growing—with the share of Americans who have never used an AI tool dropping to 27% from 33% in April 2025—overall trust is stagnant. B2B enterprise leaders cannot ignore this gap between usage and trust. See our Full Guide to understand how these shifting perceptions affect commercial technology deployment.

Why is public trust in AI-generated information so low?

Public trust in AI-generated information is low because 76% of Americans report they can trust AI outputs only some of the time or hardly ever. This metric is virtually unchanged from the April 2025 Quinnipiac survey, showing that rapid technical deployments have failed to reassure the public. Only 21% of respondents trust AI most or almost all of the time, with a mere 3% expressing absolute confidence. This trust deficit creates friction for enterprise software vendors trying to sell AI-enabled solutions to risk-averse corporate buyers.

Excitement is giving way to widespread concern

The data shows that 80% of Americans are very or somewhat concerned about artificial intelligence, while only 35% express any level of excitement. This anxiety spans every age group. Furthermore, 51% of adults state that the pace of development is moving faster than they expected. This mismatch between engineering velocity and public comfort makes users wary of automated systems, particularly in sensitive enterprise deployments.

American workers increasingly fear job displacement from AI automation

Seventy percent of Americans believe that advancements in artificial intelligence are likely to cause a net decrease in available job opportunities. This figure represents a sharp increase from April 2025, when 56% of respondents anticipated job losses. Only 7% of Americans believe AI will increase job opportunities, down from 13% in the previous survey. This growing pessimism is not isolated to specific industries; it spans both white-collar and blue-collar sectors.

Bipartisan and cross-sector job anxiety

Among employed respondents, 71% of white-collar workers and 73% of blue-collar workers agree that AI advancements will shrink the job market. This uniform concern across job classifications means B2B buyers must manage employee resistance when introducing AI software. If software vendors pitch their tools solely as cost-saving automated replacements for human labor, they will encounter internal resistance from the client's workforce.

How do Americans want AI integrated into healthcare and daily workflows?

Americans want artificial intelligence integrated as a collaborative tool that supports, rather than replaces, human decision-making. The Quinnipiac poll illustrates this preference clearly in the healthcare sector, where 81% of respondents prefer a combination of both AI and a human physician to read medical scans. Even when presented with a scenario where an AI tool is proven to be more accurate than a human, only 3% of Americans want to rely solely on the AI. Conversely, 14% would prefer to rely solely on a human.

The necessity of human-in-the-loop system design

The message for product developers is clear: fully autonomous systems face severe adoption barriers. While 45% of Americans think AI will do more harm than good in healthcare, 43% believe it will do more good, making medicine the most balanced sector in the poll. This narrow margin suggests that presenting AI as an assistant to human experts is the only viable pathway to market acceptance.

Pessimism is rising in daily life and education sectors

Public sentiment regarding the impact of AI on education and personal life has degraded significantly over the last year. The poll reveals that 64% of Americans now believe AI does more harm than good in education, a ten-percentage-point increase from the 54% recorded in April 2025. Similarly, 55% of respondents believe AI does more harm than good in their day-to-day lives, up from 44% in the previous year's survey. These shifting numbers show that familiarity does not breed comfort. As Americans interact more with these technologies, their skepticism deepens rather than dissipates.

Implications for consumer-facing enterprise tools

This rise in pessimism affects how corporations deploy customer-facing AI. While 50% of Americans believe their lives are impacted "a lot" or "some" by AI, the negative perception surrounding this impact threatens user retention. Enterprise buyers will reject customer-support bots, automated grading tools, and personal productivity assistants if end-users perceive these deployments as inherently harmful.

Key Takeaways

  • Prioritise Collaborative UX: Build systems that position AI as a co-pilot or assistant, matching the 81% consumer preference for human-AI collaboration over pure automation.
  • Address Trust Directly: Enterprise vendors must offer clear, auditable trail tracking for AI decisions to counter the fact that 76% of users distrust AI-generated information.
  • Mitigate Job Anxiety: Frame software sales around employee empowerment and capability enhancement to offset the 70% of workers who expect AI to eliminate job opportunities.

Read More

For a comprehensive overview, check out our master guide: Read the Full Guide Here.