The AI jobs knowledge inside Stanford HAI’s 2026 AI Index, launched Monday, confirms what many entry-level employees have been experiencing: employment for software program builders aged 22 to 25 has fallen practically 20 % since late 2022, the precise second generative AI instruments entered mainstream use.
Abstract
- The decline is particular to younger employees in AI-exposed roles: builders aged 30 and older in the identical corporations noticed employment develop 6 to 12 % over the identical interval, whereas name middle hiring dropped 15 % and comparable age-based divergence appeared in accounting, advertising and marketing, and customer support.
- The sample doesn’t seem in occupations with low AI publicity: well being aides, manufacturing supervisors, and guide laborers noticed regular or rising employment throughout all age teams, confirming the impact is concentrated in roles the place AI can replicate the textbook information that early-career employees depend on most closely.
- Agency surveys cited within the Stanford report point out executives count on the development to speed up, with deliberate headcount reductions in AI-exposed roles anticipated to outpace current cuts, which means the 20 % decline in younger developer employment could also be nearer to a place to begin than a peak.
As MIT Know-how Evaluate famous in its protection of the index, “the job market is struggling to keep up” with AI improvement. The Stanford research underpinning the discovering used ADP payroll data monitoring thousands and thousands of employees at tens of hundreds of corporations from 2021 by 2025, one of many largest labor datasets utilized to the AI employment query. Researchers, led by Erik Brynjolfsson, had been in a position to rule out different explanations together with distant work patterns, COVID-era hiring, and broader macroeconomic shifts, leaving the correlation with AI publicity as essentially the most constant clarification for the divergence between younger and older employees in the identical roles.
The productiveness good points AI is delivering in software program improvement are displaying up in the identical fields the place younger employment is contracting. AI can now code for hours at a time and deal with fundamental programming quicker and with fewer errors than it may when ChatGPT launched in late 2022.
The mechanism is structural. Younger builders enter the workforce with textbook information, the coding syntax and fundamental algorithms taught in laptop science packages. That’s exactly what AI instruments are finest at replicating. Skilled builders carry tacit information, system considering, and organizational context that AI can not replicate from a immediate. The result’s that AI is just not changing builders usually; it’s changing the entry-level layer that has traditionally served because the apprenticeship mannequin for the career. Stanford laptop science professor Jan Liphardt described it plainly: graduates are “struggling to find entry-level jobs” in “a dramatic reversal from three years ago.”
What the Sample Appears to be like Like Throughout Different Professions
The divergence is just not restricted to software program. Customer support representatives, accountants, and administrative assistants all confirmed the identical age-based break up, with employees aged 22 to 25 dropping floor whereas skilled employees in the identical corporations held regular. Employment for nursing aides and manufacturing supervisors, occupations the place AI augments slightly than replaces human judgment and bodily presence, grew quicker for younger employees than for older ones over the identical interval.
What This Means for the Labor Market By way of 2026
The Stanford index additionally discovered that AI is boosting productiveness by 14 % in customer support and 26 % in software program improvement, and {that a} third of organizations count on AI to shrink their workforce within the coming 12 months, notably in service and software program. These sectors are the identical ones the place the younger employee employment decline is already documented, making a suggestions loop between rising AI functionality, documented productiveness good points, and declining entry-level hiring that the 2026 knowledge exhibits is already underway slightly than nonetheless projected.
