A billionaire walks into a stadium full of students who just spent four years and tens of thousands of dollars preparing for careers. He tells them the technology most likely to erase those careers is actually exciting. They boo.
That is what happened when former Google CEO Eric Schmidt took the stage at the University of Arizona's spring commencement ceremony. The moment he pivoted to artificial intelligence - comparing it to previous technological transformations and urging graduates to get on board - sections of the crowd erupted. He tried to acknowledge the fear. "There is a fear in your generation that the future has already been written, that the machines are coming, that the jobs are evaporating," he told the crowd. The boos continued. His follow-up - "when someone offers you a seat on the rocket ship, you do not ask which seat, you just get on" - did not help. You do not tell people the ship is leaving without them and then describe it as an opportunity.
The incident became a minor viral moment. But behind the footage of a man being drowned out by his own audience is a very specific economic story. These were not students who had vague anxieties about technology. They were people entering a labor market that has been measurably, statistically worse for them than for every graduating class before them - and getting worse faster.
The background
To understand why a room full of young adults booed a speech about artificial intelligence, it helps to understand what entry-level work actually is - and why it has historically mattered.
In most professional industries, the career ladder works like this: a company hires someone out of university, pays them relatively little, and puts them to work on routine tasks. Data entry. First drafts. Basic analysis. Research. Customer calls. These jobs are not glamorous. They are, however, how someone learns the actual mechanics of a profession - the unglamorous groundwork that eventually becomes expertise. Junior developers write simple code. Junior lawyers read contracts. Junior analysts build spreadsheets. They do it badly at first, then better, then well enough to be trusted with harder things.
AI, for those coming to this for the first time, refers to software systems trained on enormous amounts of data that can now perform many of these same tasks - drafting documents, writing code, analysing data, summarizing information - at a fraction of the cost and in a fraction of the time.
The economic logic here is not complicated. If a junior employee and an AI tool can both produce a first draft, and the AI costs almost nothing compared to a salary, a company's incentive to hire the junior employee collapses. The career ladder does not disappear - it just loses its bottom rungs. The question that nobody in tech leadership has convincingly answered is: how do people get to the middle of the ladder if there is no bottom?
That tension has been building for years. It exploded into a commencement speech this month.
What is actually happening
The booing of Schmidt was not an isolated incident. According to NBC News, at least one other commencement speaker this season was met with similar hostility after calling AI "the next industrial revolution." Steve Wozniak, the Apple co-founder who spoke at Grand Valley State University and told the audience "you all have AI - actual intelligence," received a standing ovation for the contrast he drew.
The data behind the sentiment is stark. According to research compiled by Goldman Sachs, employment among workers aged 22 to 25 in AI-exposed roles fell 16% between late 2022 and mid-2025, while experienced workers in the same fields remained essentially stable. Among young software developers specifically, the drop was closer to 20%. Entry-level job postings overall declined approximately 35% between January 2023 and September 2025, according to labor research firm Revelio Labs.
As Fortune reported, over 10,000 U.S. job cuts in 2025 were directly linked to AI automation, with entry-level roles taking the heaviest share. Shopify CEO Tobi Lutke told staff the company would no longer hire where AI could do the job. Salesforce announced it would stop hiring software engineers entirely. McKinsey deployed thousands of AI agents to handle tasks previously done by junior workers. These are not rumours about the future. They are stated corporate policy, announced publicly, in the present tense.
At the same time, Meta - the company behind Facebook and Instagram - announced it was laying off roughly 8,000 workers, about 10% of its total staff, while simultaneously committing between $125 billion and $145 billion to AI infrastructure spending in 2026 - roughly double what it spent the year before. The message encoded in that arithmetic is not subtle: the company is replacing human labour with AI capacity, and it is spending more on the replacement than the original workforce ever cost.
According to IntuitionLabs analysis, the unemployment rate for young college graduates aged 20 to 24 rose to 9.5% by September 2025 - nearly double the general adult rate, and the highest in years outside the 2020 pandemic spike. The broader economy was growing. The gains were simply not reaching the people just entering it.
The money trail
Here is the economic logic that the people doing the booing have, perhaps instinctively, figured out.
When a technology company replaces a junior employee with an AI tool, it does not reduce its total costs - not in the short run, anyway. It redirects them. The salary that would have gone to a 24-year-old developer instead goes to GPU clusters, data centre construction, and the salaries of the senior engineers and executives who oversee those systems. The money does not disappear. It flows upward and inward - away from the broad base of newly credentialed workers and toward a smaller number of people who were already well-compensated.
Anthropic CEO Dario Amodei predicted in 2025 that roughly half of all white-collar entry-level positions could disappear within five years. Most other tech executives disagreed with the timeline while broadly agreeing on the direction. The debate is about pace, not destination.
The hosts of The WAN Show, a tech-focused YouTube programme that discussed the Schmidt incident today, put one version of this dynamic plainly: AI development, they argued, represents "the revenge of the idea man." For decades, the people with product ideas depended on the developers who could execute them. The power, and much of the salary, sat with technical talent. AI shifts that balance. If any sufficiently motivated non-developer can instruct a system to build something, the leverage moves away from the people who could previously do it and toward the people who control the systems and fund the infrastructure.
That shift has winners. It has losers. The winners tend to be already-wealthy technology investors and senior executives. The losers tend to be younger workers who planned careers around skills that are now partially automated before they had a chance to build the experience that makes those skills irreplaceable.
The WAN Show discussion also touched on a harder moral question: what about the engineers who build these systems? One participant noted that individual contributors at leading AI companies - people with no management responsibilities but significant technical roles - can earn close to or above $700,000 per year. They are not figureheads. They are not Sam Altman. They are also not people being forced to work by lack of options. When someone at that salary level says they morally oppose their company's direction but cannot afford to leave, the conversation, the hosts suggested, stops being about survival and starts being about something else entirely.
The comparison they reached for - though loosely - was Jean Valjean, the protagonist of Victor Hugo's "Les Miserables." Valjean steals bread because a family member is starving. That is necessity. Staying at a seven-figure job because the alternative is a pay cut, while describing the work as socially harmful, is a different calculation.
What people are doing about it
The responses are fragmented and, so far, mostly symbolic.
Students are booing commencement speeches. That is not a policy. It is a signal - a measurable expression of the gap between the optimistic framing that tech executives deploy in public and the material conditions that graduates are actually entering.
California Governor Gavin Newsom signed an executive order this month directing state agencies to study AI's impact on the workforce and develop worker protection measures, responding directly to the wave of tech layoffs. The Alphabet Workers Union - a rare labour organisation inside a major tech company - issued a statement after Meta's layoffs calling for industry-wide accountability.
Some workers are making quieter choices. In the Floatplane community chat referenced during the WAN Show discussion, one person described turning down a significant salary offer from Palantir - a data analytics company that works extensively with government surveillance and military contracts - on the grounds that they could afford to say no. That is an individual moral decision with no structural impact. But it represents the other end of the spectrum from the high-six-figure interviewee who wanted full compensation for work they claimed to find socially destructive.
Universities, for their part, have not visibly adjusted their response. The University of Arizona defended its decision to invite Schmidt, citing his "extraordinary leadership and global contributions in technology." No mention of the job market its graduates were walking into.
According to a Pew Research Center survey cited by Fox Business, a majority of Americans remain more concerned than excited about AI's expanding role in daily life. That concern is sharpest among younger workers - the people with the most direct financial exposure to AI's effect on entry-level hiring.
The bottom line
A former CEO of the company most responsible for normalizing AI in daily life went to tell graduating students to embrace the technology most likely to cost them their first jobs - and they booed him. The gap between that reaction and the continued wave of corporate AI investment is not primarily a communication problem, as tech executives tend to frame it. It is a distribution problem. The wealth created by automating entry-level work accumulates at the top of the industry. The cost - delayed careers, weaker labour market leverage, eroded rungs on the professional ladder - is absorbed by the people just starting out. The boos were not irrational. They were accurate.
Timeline
- 2022 (late): Goldman Sachs data later shows this is the peak employment moment for 22 to 25-year-old workers in AI-exposed roles, before a sustained decline begins.
- 2023 (January): Entry-level job postings begin declining; Revelio Labs later documents a 35% fall from this point to September 2025.
- 2024: Big Tech companies reduce new graduate hiring by 25% compared to 2023, according to SignalFire research.
- 2025 (February): Meta conducts a layoff round targeting roughly 5% of staff; Salesforce announces it will stop hiring software engineers.
- 2025 (August): Over 10,000 U.S. job cuts directly linked to AI automation are recorded, per Fortune reporting.
- 2025 (September): Unemployment among college graduates aged 20 to 24 reaches 9.5%, nearly double the general adult rate.
- 2025: Anthropic CEO Dario Amodei publicly predicts AI could eliminate roughly half of all white-collar entry-level positions within five years.
- 2026 (May, early): Meta announces layoffs of roughly 8,000 workers - about 10% of its workforce - while committing up to $145 billion in AI infrastructure spending for the year.
- 2026 (May 22): Eric Schmidt is booed at University of Arizona commencement after urging graduates to embrace AI and comparing it to previous technological revolutions.
- 2026 (May 22): California Governor Newsom signs an executive order directing agencies to study AI's workforce impact following the Meta layoff announcement.
- 2026 (May 27): The WAN Show discusses the Schmidt incident and the broader ethical questions facing engineers working on AI tools, including the moral calculus of high-compensation roles in AI development.
Summary
Who: Graduating students at the University of Arizona; former Google CEO Eric Schmidt; young workers in AI-exposed industries broadly.
What: Schmidt was booed during a commencement address after urging graduates to embrace AI, reflecting wider documented anxiety among young workers about AI's effect on entry-level employment.
When: The commencement ceremony took place on May 22, 2026. The broader labour market deterioration it reflects has been building since late 2022.
Where: University of Arizona in Tucson; the pattern of job losses and hiring slowdowns spans the United States, with concentrated impact in technology and white-collar professional services.
Why: Entry-level job postings have fallen significantly since 2023, young workers in AI-exposed roles have seen measurable employment declines, and major technology companies are explicitly redirecting spending from human labour to AI infrastructure - all while corporate messaging continues to frame AI as an opportunity rather than a cost borne disproportionately by those entering the workforce.