AI is quietly destroying entry-level jobs, Stanford and Anthropic studies find

A November 2025 Stanford Digital Economy Lab working paper found that workers aged 22 to 25 in AI-exposed occupations experienced a 16% relative employment decline after generative AI spread. An Anthropic report from March 2026 reached similar conclusions. More experienced workers in the same occupations did not suffer the same decline, nor did entry-level jobs with low AI exposure. The Federal Reserve Bank of New York reported that in Q4 2025, unemployment for recent college graduates rose to 5.6%, and underemployment reached 42.5%, the highest since the pandemic. Georgios Petropoulos, assistant professor at USC Marshall School of Business, argues that firms are using AI to substitute for junior tasks traditionally providing career footholds, particularly in software development, customer service, programming, and information systems management. He calls for educational institutions to embed AI literacy, data literacy, and domain judgment into degrees; governments to create tax credits and subsidies for early-career AI-augmented hiring; and firms to treat entry-level hiring as an investment in future judgment, not a short-run cost. The article warns that the old advice to 'learn to code' no longer holds, as AI now handles routine coding. Instead, supervising AI systems and combining AI fluency with domain expertise will be the scarce, valuable skill set.