Key points
- A Stanford study reveals a 13% decline in employment for those aged 22-25 in AI-exposed fields since 2022.
- The impact disproportionately affects entry-level workers in customer service, accounting, and software development.
- Employment for older workers in the same fields and younger workers in less AI-exposed jobs remains stable or has grown.
- AI’s ability to replace “codified knowledge” makes young workers particularly vulnerable.
A new study by Stanford University researchers, analyzing ADP payroll data covering millions of American workers, suggests a significant impact of generative AI on the job prospects of young workers. The research indicates a concerning 13% decline in employment among Americans aged 22-25 in occupations heavily exposed to AI, including customer service, accounting, and software development, since 2022.
This decline stands in contrast to the employment trends for older workers in these same fields, and for younger workers in less AI-affected sectors, where employment has remained steady or even increased. For instance, employment among young health aides rose at a faster rate than their older counterparts.
The study’s findings highlight the disproportionate impact of AI on entry-level positions. The researchers attribute this vulnerability to AI’s ability to replace “codified knowledge readily,” the book-learning acquired through formal education, which is prevalent in many entry-level roles.
Conversely, AI appears less capable of substituting for the experience-based knowledge accumulated by older, more seasoned workers. This disparity explains the stagnant employment growth for young workers, a contrast to the overall employment resilience observed since the pandemic.
The research carefully controlled for other factors that might skew the data, such as education levels, remote work trends, outsourcing, and broader economic shifts. Importantly, the study also notes that AI’s impact isn’t uniformly negative.
In occupations where AI serves as a complementary tool, boosting efficiency rather than replacing human workers, employment rates have remained relatively stable. This suggests that the manner in which AI is implemented has a significant influence on its impact on employment.
While this study—which is yet to undergo peer review—presents compelling evidence, it is crucial to remember that the full impact of AI on the job market is still unfolding. Most companies are yet to fully integrate AI into their daily operations, meaning the extent of its influence on employment remains to be seen.
The study’s findings, however, provide a crucial early insight into a complex and evolving situation. It supports earlier observations by Goldman Sachs economists that highlighted the already visible changes in the American labor market, particularly within the technology sector and among younger employees.
