Growing AI adoption in the US is met with declining trust in results

Artificial intelligence is becoming a regular part of everyday life in the United States, with more people using it for research, writing, work-related tasks, and data analysis. However, even as usage increases, confidence in the technology remains low, reflecting a growing disconnect between adoption and trust.

A new survey released by Quinnipiac University highlights this divide. Among nearly 1,400 respondents, more than three-quarters expressed limited trust in AI systems. Around 76% said they trust AI either rarely or only sometimes, while just 21% reported trusting AI-generated outputs most or nearly all of the time.

At the same time, the number of Americans who have never used AI tools continues to decline. Only 27% said they have not used AI at all, down from 33% in April 2025, showing that adoption is steadily growing despite scepticism.

Chetan Jaiswal, a computer science professor at Quinnipiac University, pointed out the contrast between usage and confidence. He noted that over half of respondents now use AI for research, with many also relying on it for writing, professional work, and information analysis. Yet trust levels remain significantly lower, indicating that people are engaging with the technology cautiously rather than confidently.

Part of this hesitation appears to be driven by concerns about AI’s broader impact. The survey found that only 6% of respondents described themselves as “very excited” about AI, while 62% said they were either not very excited or not excited at all. When it comes to concern, the numbers are even more striking: 80% reported being somewhat or very worried about AI, with millennials and baby boomers among the most concerned groups, followed closely by Gen Z.

A majority of respondents — about 55% — believe AI will do more harm than good in their daily lives, while only one-third think the benefits will outweigh the risks. Compared with previous surveys, negative sentiment toward AI has increased, which researchers suggest may be influenced by factors such as widespread layoffs in the tech sector, high-profile AI-related incidents, and concerns about the environmental impact of large-scale data centres.

Public opposition to AI infrastructure is also evident. Around 65% of respondents said they would not want an AI data centre built in their local community, citing concerns about electricity consumption and water usage.

Concerns about employment are particularly strong. About 70% of respondents believe AI advancements will reduce job opportunities, while only 7% think the technology will create more jobs. This marks a shift from the previous year, when fewer people expected job losses. Among younger respondents, pessimism is especially pronounced: 81% of Gen Z participants expect a decline in job opportunities.

These concerns are not entirely unfounded. Entry-level job postings in the U.S. have dropped significantly since 2023, and leaders in the AI industry — including Dario Amodei of Anthropic — have warned that automation could eliminate a large portion of entry-level roles in the coming years.

Tamilla Triantoro, a professor of business analytics and information systems at Quinnipiac University, observed that younger Americans, despite being the most familiar with AI tools, are also among the least optimistic about the future job market. She noted that familiarity with AI does not necessarily translate into confidence about its long-term effects.

Interestingly, while many respondents are worried about the overall job market, fewer believe their own roles are immediately at risk. Among employed individuals, 30% said they are concerned that AI could make their jobs obsolete — an increase from 21% the previous year, but still lower than broader concerns about the labour market.

Another major factor contributing to low trust is scepticism toward institutions. Around two-thirds of respondents believe that companies developing AI are not being transparent enough about how the technology is used. A similar proportion feel that the government is not doing enough to regulate AI effectively.

This sentiment comes at a time when debates over AI regulation are intensifying, with states seeking to maintain authority over rules. At the same time, federal policymakers and industry leaders push for lighter regulatory approaches.

Overall, the findings suggest that while Americans are not rejecting AI outright, they remain cautious. Concerns about job security, transparency, and regulation continue to shape public perception, even as the technology becomes more deeply integrated into everyday life.