The booing at commencement ceremonies this spring was not about technology. It was about data. The class of 2026 had already read the numbers before the speakers took the podium, and the numbers were not reassuring.
The Next Web reported on May 21 that the scene repeated at two commencements. Eric Schmidt at the University of Arizona told graduates the AI moment would be larger, faster, and more consequential than anything they had lived through, and was booed. A real-estate executive at the University of Central Florida used the phrase “the next industrial revolution,” and was booed. The press framing reached for generational confusion. A better frame is labor-market accuracy.
The data the graduating cohort had already seen is specific. Goldman Sachs research this spring put AI-related US job losses at roughly 16,000 per month, with entry-level workers absorbing a disproportionate share. The Dallas Federal Reserve found that the unemployment-rate gap between entry-level and experienced workers widened sharply after the pandemic, concentrated in occupations exposed to AI substitution. ServiceNow CEO Bill McDermott told a conference that new-graduate unemployment could climb sharply within two years. Standard Chartered announced that it would cut a large share of back-office roles in HR, risk, and compliance by 2030, describing the trade as replacing lower-value human capital with AI. Those are the roles new graduates take in years one through three at a bank.
The thesis that emerges from these data points is that AI is removing the bottom rung of white-collar career ladders rather than reshuffling the rungs. The Dallas Fed finding is precise: the gap is between entry-level and experienced workers, not between technologists and non-technologists. The skill that protects a worker in this particular wave is not knowing how the models work. It is years of contextual judgment on a workflow that a model can now run in seconds. Experienced workers have that judgment. New graduates do not.
Structural skepticism is required before accepting the AI-causation thesis wholesale. Post-pandemic hiring patterns created an artificial demand surge for junior white-collar roles between 2021 and 2023, followed by a correction that would have produced widening unemployment gaps regardless of AI adoption. Interest-rate effects pushed companies toward headcount efficiency in 2024 and 2025 in ways that historically front-load cuts at the junior level. Offshoring of entry-level back-office work, accelerated by remote-work normalization, is a separate compressor running in parallel. What would actually confirm the AI-causation story is not the aggregate job-loss number but a documented shift in the ratio of junior-to-senior hires within individual firms that cite AI productivity as the specific reason, tracked across a full hiring cycle.
The evidence is not yet that clean. What we have is correlation with a plausible mechanism, not confirmed causation. That distinction matters for how founders and operators should read the signal.
The corporate narrative has also been running ahead of corporate action. The boardroom response to the displacement story is that AI will produce more interesting work for the cohort that survives the transition, with training programs converting employees into AI-operator roles. The problem is the asymmetry. When a company redeploys staff into AI-focused positions while cutting a comparable number in the same week, the redeployment applies to the headcount being kept, not the headcount being released. That asymmetry is what the booing was about.
The aggregate picture, as The Next Web reported, is capital absorbing the productivity gain rather than labor capturing it. The largest US technology companies have committed a combined sum measured in the hundreds of billions of dollars to AI infrastructure in 2026. Many of those announcements land in the same news cycle as a layoff announcement from the same balance sheet. The spending rises. The employment falls.
For founders and operators, the relevant question is not whether AI substitution is happening at some macroeconomic scale. The relevant question is whether your own junior-hiring decisions over the next 18 months are building the talent pipeline that will produce your senior hires in 2031, or whether you are optimizing the wage line today at the cost of organizational depth later. If eliminating entry-level roles removes the pipeline that produces experienced workers, the firms cutting deepest now are not running an efficiency play. They are running a deferred talent-debt trade, and the bill arrives when AI-assisted workflows hit the limits of contextual judgment that only tenure produces.
Reported by The Next Web on 2026-05-21.
