AI Is Killing Entry-Level Jobs, Stanford Study Finds

The spread of GenAI has fueled global debate over its near- and long-term impacts on work. Now, new research from Stanford’s Digital Economy Lab provides the clearest evidence yet: AI is reshaping the job market — and young, entry-level workers are taking the biggest hit.

The study, "Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of Artificial Intelligence" by Erik Brynjolfsson, Bharat Chandar and Ruyu Chen, uses payroll data from millions of workers across tens of thousands of US firms. This granular view — based on records through July 2025 — offers a near real-time picture of how AI is altering employment dynamics.

Entry-Level Jobs in Decline

The researchers’ most striking finding: early-career workers (ages 22–25) in occupations most exposed to AI, such as software developers and customer service representatives, have experienced a 13% relative decline in employment since late 2022. By contrast, employment for more experienced workers in the same jobs has remained stable — or even grown.

In fact, while overall US employment continues to expand, job growth for younger workers has been stagnant for nearly three years. The sharpest declines are in occupations where AI primarily automates tasks, rather than augments them. For example, young software developers and customer support agents are seeing steep drops, while marketing managers and sales leaders have experienced only modest declines. Fields less exposed to AI, such as production supervision, stock clerks and health aides, show no meaningful impact and continue to grow.

Why Younger Workers Are Hit Harder

The authors suggest that AI disproportionately replaces codified knowledge — the formal training and “book learning” that recent graduates bring to the table. By contrast, it is less capable of displacing tacit knowledge — the practical judgment, intuition and experience that accumulate over time.

This helps explain why older workers are holding their ground while younger employees are losing traction. Anyone who has trained sales development reps or telesales teams can relate: it often takes months before new hires can apply classroom instruction effectively. AI tools are increasingly filling that gap, providing a faster and cheaper substitute.

Containment Before Deployment

Dresner Advisory Services research shows that 15% of organizations already have GenAI systems in production and 6.5% have agentic AI deployed. Laggards may be holding back on entry-level hiring as they anticipate AI adoption, “containing” employment before fully rolling out automation.

The Bigger Picture

The study highlights six key takeaways:

Entry-level exposure: Young workers in AI-heavy occupations face sharp employment declines.
Overall resilience: US employment remains robust, masking stagnation for workers under 25.
Automation vs. augmentation: Jobs where AI automates are shrinking; jobs where AI augments are still growing.
Persistent effects: The declines hold even after controlling for firm-level conditions.
Pay is steady: Impacts are showing up in employment, not compensation — at least so far.
Broad validity: Findings remain consistent across industries, job categories and remote vs. non-remote roles.

A New Kind of Transition

Historically, technological shifts — from the IT revolution to industrial automation — have displaced some workers while creating new opportunities elsewhere, ultimately driving productivity and wage growth. AI may follow a similar long-run trajectory, but in the short term, it is clear that entry-level workers are bearing the brunt.

As Brynjolfsson and his co-authors warn, early-career workers are the “canaries in the coal mine.” Their struggles signal that AI’s disruption is already here — and organizations, policymakers and educators must adapt before an entire generation of workers falls behind.

fa-solid fa-hand-paper Learn how you can join our contributor community.