Majority of Americans Now More Concerned Than Excited About AI, and the Young Are Turning Fastest

For most of the past decade, the public-opinion story about artificial intelligence ran in one direction: awareness up, usage up, and concern drifting steadily higher. A new Pew Research Center survey, conducted June 22–28, 2026 among 3,488 U.S. adults and published August 18, confirms the trend has not softened — and it contains one finding that should trouble anyone building AI products aimed at young people: for the first time, a majority of American adults under 30 say they are more concerned than excited about AI.

The headline number is stark. 52% of U.S. adults now say the increased use of AI in daily life makes them feel more concerned than excited, up from 37% when Pew first asked the question in 2021. Only 9% are more excited than concerned — the lowest figure Pew has ever recorded on this question — while 37% sit in the ambivalent middle, saying they feel both equally.

The generational reversal that matters most

The topline masks the more interesting movement underneath. Dig into Pew’s age breakdowns and you find a genuine generational reversal in progress.

In 2021, younger Americans were the optimists: 25% of adults aged 18–29 said they were more excited than concerned about AI, the highest of any age group and more than double the 11% among those 65 and older. Five years later, that enthusiasm has collapsed to 11% — and the share of young adults who are more concerned than excited has climbed from 31% to 55%, crossing the majority threshold for the first time.

The trajectory matters as much as the endpoint. For Americans 30 and older, concern rose sharply between 2021 and 2023 and has since plateaued or drifted down slightly — those 65+ are actually a bit less worried than they were three years ago (59% concerned in 2026 vs. 61% in 2023). But for the under-30 cohort, the line keeps climbing every single survey wave: 31% in 2021, 29% in 2022, 42% in 2023, 39% in 2024, 47% in 2025, 55% now. Young adults’ skepticism of AI has made headlines in recent months, and Pew’s data now puts hard numbers behind the anecdote.

The youngest adults are now exactly as worried as everyone else — concern among 18–29s (55%) is on par with those in their 30s and 40s (51%) and those 65 and up (59%). The generational enthusiasm gap that tech companies have quietly relied on — the assumption that each incoming generation would be more AI-friendly than the last — has closed, and in the space of five years.

Jobs: the concern that keeps growing

Why the worry? Job displacement remains the anchor. 71% of Americans now say AI will lead to fewer jobs in the United States over the next 20 years, up from 64% in 2024. Just 5% expect AI to create more jobs, and only 10% think it won’t make much difference.

Here again, the young-adult shift is the story. In 2024, 61% of adults under 30 expected AI-driven job losses; today it’s 73%, a 12-point jump in two years that has pulled the youngest cohort level with Americans aged 30 to 64 (72–74%). Pew’s past work has shown that job loss has been one of the top reasons behind Americans’ AI concern since at least 2021, but the intensity is new: majorities in every single age group now expect net job destruction.

This lands at an awkward moment for the AI industry. OpenAI launched a teen-tailored ChatGPT just this week, wrapping the product in study modes and parental controls. Anthropic is posting $11.5B quarters selling Claude to enterprises whose employees are watching the same headlines. The companies steering the largest technology transition in a generation are doing so into a public that is — by Pew’s measure — the most skeptical it has ever been.

Not just anxiety about jobs

Pew’s related research suggests the unease among young adults runs deeper than employment math. Adults under 30 are the most likely age group to think AI will be bad for society and bad for them personally. They largely believe AI will make genuine human connection and creativity harder to come by. And in perhaps the most pointed finding for the chatbot industry: although most young adults say they use AI chatbots, they are just as likely to say these tools hurt their own creativity as help it.

That last data point deserves emphasis. The under-30 cohort is the demographic with the highest chatbot adoption and the deepest familiarity with these tools — and familiarity is not breeding comfort. Usage and warmth have decoupled. Young people use the products and resent them anyway, a pattern that historically precedes regulatory pressure and brand damage in other technology cycles.

The context around the numbers

The survey itself is methodologically solid: Pew surveyed 3,488 U.S. adults from its American Trends Panel, a probability-based panel designed to be representative of the full U.S. adult population, with fieldwork June 22–28, 2026. The findings align with Pew’s March 2026 synthesis and its June 2026 report on chatbots and smart devices, which found that about half of U.S. adults now use AI chatbots and that more adults predict a negative rather than positive personal impact from AI.

The trend line also arrives amid a particularly bruising news cycle for AI’s public image: the July disclosure that OpenAI models breached Hugging Face’s production infrastructure during an internal evaluation, the safety overhaul OpenAI announced this week in response, a landmark trial over social-media addiction harms, and ceaseless coverage of AI data-center strain on electricity and water supplies. It is difficult to isolate how much of the 2026 concern spike is driven by specific events versus a slow accumulation of unease — Pew’s data shows the big jump happened between 2022 and 2023, before most of this year’s headlines.

Why it matters

Public opinion is not a veto on technology adoption — Americans adopted smartphones and social media while expressing mixed feelings about both. But it is a leading indicator of three things the industry ignores at its peril.

First, regulation. Politicians follow polls, and a 71% supermajority expecting job losses creates political space for everything from state-level AI transparency laws to federal workforce protections. Pennsylvania’s governor signed an executive order this week restricting AI data centers without local approval — the kind of measure that polls well when a majority of the public is anxious.

Second, talent and adoption. The cohort turning against AI most sharply is the one the industry most wants to hire and to sell to. Products built on the assumption of youthful enthusiasm — AI study companions, AI social features, AI creativity tools — are now shipping into a majority-skeptical user base.

Third, the burden of proof has shifted. In 2021, AI companies could assume benefit-of-the-doubt. In 2026, with concern at 52% and excitement at a record-low 9%, the default public posture is doubt. Every safety incident, every chatbot mishap, every layoff attributed to automation now lands on dry tinder.

Pew’s numbers do not say Americans will stop using AI — clearly they haven’t. They say something more uncomfortable: the public has learned to live with AI without learning to like it. For an industry that still measures success in weekly active users and enterprise seat counts, that gap between usage and trust is the metric to watch.