The Evidence That AI Is Destroying Jobs For Young People Just Got Stronger Author: Derek Thompson Date: August 27, 2025 --- Overview This article discusses the complex and evolving debate on whether artificial intelligence (AI), particularly generative AI like ChatGPT, is negatively impacting employment opportunities for young people, especially recent college graduates. It synthesizes diverse viewpoints and new empirical evidence suggesting AI's influence on the labor market. --- The Debate So Far: Four Answers Possibly! Early in 2025, The New York Federal Reserve noted declining work opportunities for recent graduates. Hypotheses included economic factors and political chaos, but also the possibility that companies were using AI for tasks once done by young grads. Harvard economist David Deming suggested AI could replace typical entry-level white-collar work. Definitely Yes! Major media outlets like The New York Times and Axios reported AI causing a job crisis for recent grads. Anthropic CEO predicted up to half of entry-level white-collar jobs could disappear within five years due to AI. Almost Certainly No! Several analyses challenged the previous claims, showing little evidence AI caused unemployment. The Economic Innovation Group and researchers found most firms reported no net employment impact from AI. Entry-level tech hiring appeared to have rebounded recently. Plausibly Yes! (Latest Stanford Paper) New research by Stanford economists analyzing ADP payroll data (covering millions of workers) found a 13% employment decline for young workers (aged 22–25) in highly AI-exposed jobs like software development and customer service since ChatGPT's widespread use. Employment was steady or grew for older workers and in AI-low-exposure jobs like home health aides. This suggests a specific, age- and occupation-linked employment effect. --- Key Insights from Stanford Economists Erik Brynjolfsson and Bharat Chandar Data Richness: ADP’s huge monthly dataset allows precise tracking by age and occupation, overcoming sample size limits in traditional surveys like CPS. Employment Patterns: Declines are concentrated among young workers in AI-automatable jobs (software engineering, auditing, accounting). Jobs where AI augments human work (more complex or managerial roles) show no corresponding employment drop. Within-Firm Effects: Even inside the same company, more AI-exposed jobs see relative employment decline compared to less exposed jobs, reinforcing the AI linkage beyond economy-wide shocks. Nature of AI Impact: AI excels at tasks using codified knowledge accessible in texts and online, overlapping with what young workers typically do. Senior workers rely more on tacit knowledge, which AI cannot easily replicate. Shorter time-horizon, observable tasks are more susceptible to AI substitution. --- Implications for Education and the Future Workforce Young workers and students should learn how to effectively use AI tools as part of their training. Universities have yet to adapt curricula for this AI shift. Physical and tacit knowledge-heavy jobs may grow in importance. --- Broader Context and Conclusion AI's impact is not just future speculation; it is already affecting GDP growth, education integrity, mental health, and crucially, the young labor market. Predicting the future is impossible, but careful observation and analysis of current trends is essential. The debate over AI and employment will continue to evolve, and evidence now points toward AI already reshaping young people's job prospects. --- Community Discussion Highlights Some commentators wonder if declines reflect actual displacement or premature executive fears of AI's capabilities. AI currently helps with a subset of tasks