Policy—not PR—will determine Gen Z’s trust in AI

This spring, college graduates routinely booed commencement speakers who praised artificial intelligence (AI). A real estate executive drew jeers at the University of Central Florida for calling the rise of AI the “next industrial revolution”; a record-label CEO told booing Middle Tennessee State University graduates to “deal with it”; and former Google CEO Eric Schmidt was heckled at the University of Arizona for praising a world dominated by AI. 

These reactions reflect a growing body of survey evidence that Gen Zers—those born between 1997 and 2012—are increasingly skeptical of AI even as their own use of the technology has risen. About half of Gen Z respondents report in surveys that they use generative AI at least weekly, even as they express concerns over AI’s impact on human connection, creativity, the environment, and more. In fact, a vocal subset is turning away from the technology altogether, calling use of AI embarrassing or a “red flag.” Tech leaders appear to have sensed this reaction and softened their pitch, walking back predictions of mass layoffs in favor of a story about AI augmenting work rather than replacing it. This public relations shift on their part misdiagnoses the root cause of these frustrations and does not address the more fundamental problems swaths of Gen Z have with AI. 

As Brookings scholar Darrell West has argued, every transformative technology produces winners and losers—and it is near-term policy choices, not public communications, that determine which way the balance tips. At present, survey after survey suggests much of Gen Z has concluded it is on the losing side. 

Younger adults are the most exposed to AI and, increasingly, the most wary of it. In a September 2025 Pew Research survey, 62% of adults under 30 said they had heard a lot about AI—the highest of any age group, yet this same group was also the most pessimistic about its effects: 61% said it would worsen people’s ability to think creatively, and 58% said it would erode their capacity to form meaningful relationships, both roughly 20 points higher than respondents 65 and older. A February 2026 Pew survey found just 14% of adults under 30 expect AI’s effect on society to be positive—two points below the average across age groups.

These concerns are elevated in four areas where the public has historically expected protection.

Cognitive offloading

A large majority of Gen Zers believe AI will make future learning harder and people lazier; in one workplace survey, 46% of Gen Z workers said AI is making them “dumber,” against 39% of workers overall.

The supporting evidence is early: A June 2025 not-yet-peer-reviewed pre-print from MIT Media Lab found weaker neural connectivity and poorer recall among participants who wrote an essay with a large language model (LLM). Those assigned an LLM partner struggled to accurately quote their own work and exhibited under-engagement, suggesting that use of AI may pose some cognitive costs to neural connectivity.

Researchers at Microsoft Research and Carnegie Mellon University surveyed 319 knowledge workers across 936 real-world AI-assisted tasks and found that higher confidence in generative AI was associated with less critical thinking and that the technology shifted cognitive effort away from analysis and toward information verification and task stewardship. A separate 2025 study by Michael Gerlich of SBS Swiss Business School found that higher AI use is correlated with greater cognitive offloading and lower critical thinking scores, especially among younger participants.

Job displacement

One of AI’s biggest uncertainties is the potential for mass disruption in the labor market. Research shows that, at least at present, mass labor disruptions are not happening. Instead, companies using AI as an excuse for layoffs driven by other economic motivators (or “AI-washing”) may be more common. Nonetheless, many are convinced that the technology will lead to massive changes, if not, a complete upheaval of the job market. With very early labor market data, and most experts mixed or uncertain, it is difficult to affirm or negate these fears.

These concerns are common among Gen Z, many of whom are recent graduates inheriting a less-than-ideal job market. This is reflected in several surveys: 48% of employed Gen Z workers say AI risks outweigh benefits in the workforce, up from 37% last year. Similarly, 69% say they trust work completed by only humans more than that assisted by AI. And just 3% have the most trust in fully AI-generated work.

Human relationships

Gen Z is also worried that use of generative AI could deteriorate or replace existing human social relationships. A July 2025 Common Sense Media report found that 72% of U.S. teens have tried AI companions and 52% use them regularly; one-third reported telling a companion important or serious matters instead of another person. There is a significant amount of research demonstrating that teens today report having fewer friends than in previous decades and that a “loneliness epidemic” is largely due to social media. In a New York Times interview, Abigail Hackett, 27, said that she does not use AI in her personal life because “she does not want her social muscles to atrophy.”

Extending from this are perceptions of AI’s impact on creativity and authenticity, where a social stigma seems to be arising. Some Gen Zers have described AI-generated content as “fake and deeply uncool.” One college student said others’ AI use was a “red flag” and would cause peers to “think less” of someone.

Environmental impact

Data center backlash appears to have reached a fever pitch, with disputes surfacing in state and municipal elections and likely to intensify ahead of the midterms. Communities across the country are pushing back on these facilities in their neighborhoods, and states and localities are increasingly considering bans. Maine’s legislature passed a moratorium on large new data centers, though it was ultimately vetoed by the governor, and members of Congress like Sen. Bernie Sanders (I-Vt.) and Rep. Alexandria Ocasio-Cortez (D-N.Y.) have proposed one at the federal level.

More than half of adults under 30 (54%) say data centers have a mostly negative effect on the environment, and 49% of Gen Z says they are “extremely/very concerned” about data center environmental impacts versus 37% of all American adults. The perception of these costs indicates a generation that sees AI benefits contributing to the success of the concentrated technology sector, while the costs flow to everyone else.

These are the concerns that surveys most consistently identify as Gen Z priorities, though the list not exhaustive. Others include confabulations and accuracy, data privacy, and the technology’s effect on democracy.

Yet, the companies building AI are largely familiar to Americans, as, in most cases, they are the same tech companies that consolidated power over the past two decades and are wielding it to lead in AI. Granted, there are some newer players, like OpenAI and Anthropic, but the reaction to these companies largely tracks established sentiments about previous tech giants. Take, for instance, the user backlash when OpenAI’s late 2025 test surfaced in-chat app suggestions that many users mistook for ads, a feature the company disabled within days. Despite the reaction, the company announced a separate monetized advertising pilot in February 2026, placing labeled ads beneath ChatGPT’s outputs.

Over the past decade, Americans’ confidence in tech firms has dropped even more significantly than it has for other institutions. A 2023 Brookings analysis of the American Institutional Confidence poll found the youngest cohort lost confidence in major tech firms faster than any other group and warned that honeymoon periods for new technologies like LLMs were likely to shrink.

Later, in 2024, Americans broadly reported concerns over Big Tech’s influence and voiced support for increased regulation of tech companies. This is especially apparent as people became disillusioned with social media: Per Pew, “Americans are far more likely to say social media has a negative rather than positive impact on the country,” with 64% saying it has had “a mostly negative effect on the way things are going in the country today.” These negative sentiments may rise this year with major AI firms spending millions to influence the 2026 midterm elections.

Given that many of these same companies are now touting AI as a technology that will change, if not largely “save,” the world, it’s unsurprising that Americans, and especially young people who grew up alongside the previous innovations, are skeptical of the latest offering.

In many ways, AI threatens to consolidate power further as it is embedded in government and supported by the Trump administration; beyond this, there are fears about how it may exacerbate inequality amid a deregulatory agenda in the United States. AI use varies across income levels: Only 9% of workers earning under $30,000 use the technology, while usage rises to 34% of workers making $100,000 or more. One survey suggests that the Americans most excited about AI are also those in higher income brackets, which largely excludes younger generations already facing staggering income inequality and wages lagging the rate of inflation. As Nathaniel Persily told the Atlantic recently, “Disruption has winners and losers. For many Americans, they’re not convinced they’re going to be the winners, and they base that conclusion on the history of technology over the last 20 years.”

Taken together, the evidence suggests that Gen Z’s reservations about AI are more than superficial or based on public narratives. And given the negative sentiments brewing, companies will have to re-earn the trust of many different groups—not just young people but also marginalized communities.

Building back meaningful trust would include targeted governance such as:

• Labor protections: Policies could include portable benefits, wage insurance for displaced workers, lifelong learning accounts, disclosure standards around AI-attributed layoffs, and targeted support for an entry-level cohort bearing early costs.
• Creativity protections: There is currently active litigation on whose creative work trained AI systems and whether those creators are compensated. A framework for consent and compensation would protect creative labor as a category.
• Environmental protections: Policymakers can implement improved mutualistic data center siting standards, mandatory energy and water disclosures, and ratepayer protections and affordability assistance so households do not subsidize private compute through higher utility bills.
• Basic consumer and safety protections: Adding privacy- and safety-by-design requirements, anti-discrimination enforcement, and heightened protections for minors can increase trust.

Tech CEOs have begun to sense the shift—pledging to cover the electricity costs of their data centers and walking back earlier predictions of mass layoffs. But these voluntary gestures are unlikely to address major public risks and distributional problems. While literacy programs are worth developing, Gen Z knows AI well, and negative sentiments indicate a desire for greater protection of important values. Whether AI tilts toward more winners or losers will be determined by decisionmakers, including federal and state policymakers and firms willing to accept binding obligations. Gen Z is issuing an early warning that AI’s social contract has yet to be finally written, and much more is needed to address the generation’s fundamental concerns.