Young Americans' AI optimism collapses as job fears take hold

Pew's headline number, that 73% of adults under 30 now expect AI to shrink the U.S. job market over the next two decades, up from 61% just two years earlier ^[1], obscures a more specific fear once you look past the topline stat. Discussion around the survey converged less on whether AI eliminates jobs and more on how. The dominant read wasn't senior employees losing their positions to automation; it was entry-level 'grunt work' disappearing first, the unglamorous tasks that have historically doubled as an apprenticeship for turning a new graduate into someone trusted with judgment calls.

That distinction matters because it reframes AI anxiety as a pipeline problem rather than a pure substitution problem. If the tasks that once taught a 24-year-old how to do the job are the first ones automated away, the anxiety isn't really about a machine doing your job today, it's about never accumulating the experience needed to do a bigger job tomorrow. That is a structurally different, and arguably harder to fix, kind of labor-market fear than the generic robots-will-take-my-job narrative the industry is used to rebutting.

One of the more counterintuitive threads in the data is that the demographic most worried about AI is also the demographic using it the most. Under-30s report the heaviest day-to-day AI usage of any age group, yet they are also the group most likely to say AI hurts their creativity, with one estimate putting the share of under-30 chatbot users who say the tools damage their creative output at 20%, the highest of any age bracket, alongside 61% of young adults who worry AI is broadly worsening people's capacity for creative thinking ^[2].

The intuitive assumption would be that daily exposure breeds comfort. The data suggests the opposite: proximity to the tools is what makes the risk feel concrete rather than abstract. Adults early in their careers, whose day-to-day work most closely resembles what AI is being trained to automate, are precisely the ones with the clearest view of what could be lost. Worry here reads less like technophobia and more like an informed risk assessment from people who understand the tool well enough to see themselves being replaced by it.

Pew's parallel survey of AI experts found 47% are more excited than concerned about the technology's growing role in daily life ^[4]. Among the general public and young adults, the equivalent figure is just 9-11%, a 36-to-38-percentage-point gap between how the people building AI feel about it and how the people living with its rollout feel about it.

The size of that gap has consequences beyond public opinion polling. Policy, product design, and corporate deployment decisions about AI are disproportionately shaped by the optimistic minority, researchers, engineers, and executives closest to the technology, while the group bearing the most direct labor-market exposure, workers under 30, has the least representation in that optimism. That disconnect shows up in the wider online conversation too: commentary tracking this same Pew data has framed it as part of a broader collapse in institutional trust, not just a jobs story, pointing to figures like roughly two-thirds of people saying they don't trust the companies building AI and Gen Z anger toward AI climbing from 22% to 31%. Those numbers come from social and video commentary rather than Pew's own published materials, so they read as amplification of the mood the Pew data documents rather than as additional hard statistics. The direction is consistent regardless: the expert-public split isn't an isolated finding, it's one symptom of a wider trust deficit around who is steering the technology and why.

The shift did not happen overnight. In 2021, only 31% of adults under 30 said they were more concerned than excited about AI ^[1]. By 2024, the story had already started turning toward jobs specifically: 61% of under-30s and 64% of all adults said they expected AI to reduce the number of available jobs ^[1]. By the June 2026 survey, those job-loss figures had climbed further to 73% and 71% respectively, while overall concern crossed the 50% threshold among young adults for the first time, hitting 55% ^[1].

What makes this trajectory notable is that older adults' concern has largely leveled off over the same period, while younger adults' has kept climbing ^[3]. Even outside Pew's own tracking, the same reversal shows up against a completely different starting point: media coverage of the release has noted the all-adult concern figure is up from roughly 38% in late 2022, just after ChatGPT's public debut, a shorter and more recent baseline than Pew's 2021 anchor, but one that lands on the identical conclusion. However you slice the timeline, concern has been climbing steadily rather than plateauing. Early AI anxiety was diffuse, touching on job loss, creativity, misinformation, and human connection in roughly equal measure. Five years later, for the under-30 cohort specifically, job displacement has pulled ahead of every other fear as the dominant driver, and it is the one input still pulling the topline number upward rather than leveling off.