AI and Employment: How Generative Technology is Reshaping the Labor Market

Stanford University Study: AI Impacts Entry-Level Jobs for Young Americans

Analyzing employment records from ADP, the largest US payroll provider, from late 2022 to July of this year, Stanford University researchers found that the AI ​​revolution is impacting the US labor market, particularly entry-level workers. The study showed a significant decline in employment rates for young workers aged 22-25 in highly AI-exposed occupations (such as software development and customer service representatives). Software developer employment plummeted nearly 20% from its peak in late 2022, while older workers were unaffected.

The study found that job losses for young people in highly AI-exposed occupations were so severe that they dragged down employment growth for the entire 22-25 age group. In contrast, employment for young workers in occupations with low AI exposure, such as maintenance and nursing, actually increased by 6-13%. After controlling for factors such as firm-specific shocks, the researchers found that the job losses were due to the jobs themselves, not to any difficulties faced by the firms themselves.

Introduction

The rapid advancement of generative artificial intelligence (AI) has sparked intense debate about its impact on jobs. While some view AI as a productivity booster, others fear widespread displacement of workers. This article examines recent research findings on AI’s employment effects, focusing on six key trends observed in the U.S. labor market.

Study Background

Our analysis draws from a comprehensive study using high-frequency payroll data from ADP, America’s largest payroll processor. The research tracked employment changes across 3.5-5 million workers monthly from 2021-2025, linking job data to AI exposure metrics developed by leading researchers.

Six Key Findings About AI’s Labor Market Impact

1. Young Workers Face Disproportionate Impact

Key Observation:
Workers aged 22-25 in AI-exposed occupations experienced significant employment declines after 2022, while other age groups maintained stable growth.

Example Occupations:

Why This Matters:
Early-career workers typically rely more on codified knowledge that AI systems can replicate, while experienced workers leverage tacit knowledge less susceptible to automation.

2. Overall Employment Growth Masks Sectoral Shifts

The Big Picture:
While national employment remains robust, young workers in AI-exposed fields show stagnant growth compared to other demographics.

Employment Growth Comparison (2022-2025):

Key Insight:
The divergence suggests structural changes rather than broad economic shifts.

3. Automation vs. Augmentation Matters

Critical Distinction:
AI applications that automate tasks (e.g., basic coding, routine customer queries) show employment declines, while augmentative uses (e.g., complex problem-solving support) show neutral or positive effects.

Occupational Examples:

4. Firm-Level Shocks Don’t Explain Trends

Robustness Check:
Controlling for company-specific factors (interest rate changes, industry disruptions) still shows 12% relative employment decline for young workers in high-exposure occupations.

5. Wages Show Less Dramatic Shifts

Compensation Trends:
Unlike employment numbers, salary data shows minimal variation across age groups and exposure levels, suggesting short-term wage rigidity.

Annual Compensation Change (2022-2025):

• High-exposure occupations: +2.1%
• Low-exposure occupations: +2.3%

6. Findings Hold Across Multiple Scenarios

Consistency Checks:

• Results persist when excluding tech industry workers
• Remote-work capable vs. non-remote occupations show similar patterns
• Pre-2022 data shows no predictive pattern for current trends

Historical Context and Future Outlook

The authors note parallels to past technological transitions, such as the IT revolution, which initially displaced workers but ultimately created new job categories. However, the current transition may unfold differently due to AI’s unique capabilities.

Key Considerations:

• Education systems may need to emphasize tacit knowledge development
• Experience-based career paths may gain new importance
• Continuous learning will be critical for workforce adaptation

Frequently Asked Questions

❓ Which occupations are most affected?

Software development, customer service, and marketing roles show highest sensitivity to AI exposure.

❓ Is this just a temporary trend?

The study notes similarities to past technological transitions but emphasizes the need for ongoing monitoring.

❓ What can workers do to adapt?

Focus on developing skills that complement AI capabilities:

• Complex problem-solving
• Creative thinking
• Interpersonal communication
• Industry-specific expertise

❓ How should companies respond?

Invest in:

• Employee reskilling programs
• Human-AI collaboration frameworks
• Career path diversification

Practical Recommendations

For Workers:

1. Develop hybrid skills combining technical and soft skills
2. Seek roles emphasizing human-AI collaboration
3. Prioritize continuous learning opportunities

For Employers:

1. Audit job roles for automation vulnerability
2. Create internal mobility programs
3. Invest in AI literacy training

For Policymakers:

1. Expand access to adult education programs
2. Strengthen safety nets for displaced workers
3. Promote public-private partnerships for workforce development

Conclusion

The relationship between AI and employment remains complex and evolving. While early data shows concerning trends for young workers in certain sectors, historical patterns suggest eventual labor market adaptation. The key lies in proactive adaptation strategies at individual, organizational, and societal levels.