“`html
Addressing the Looming Crisis in Entry-Level Work
Artificial intelligence has not yet produced a clear narrative of widespread unemployment. Aggregate employment in developed countries remains largely stable, and recent assessments have found limited evidence that AI has shifted headline numbers. However, there may be a hidden change beneath the surface: a decline in early-career hiring.
The most concerning evidence is emerging where we expect it first—among workers aged 22 to 25 in occupations heavily exposed to generative AI. A working paper released in November 2025 by the Stanford Digital Economy Lab found that these workers experienced a 16% relative decline in employment after the spread of generative AI, even after controlling for other factors. An anthropic.com/research/labor-market-impacts”>Anthropic report from March 2026 provides similar evidence.
This is not a minor signal. It suggests that firms may be substituting junior tasks traditionally handled by entry-level workers with AI, specifically in occupations where generative AI is used extensively—like software development, customer service, programming, and information systems management. More experienced workers in these same occupations did not suffer the same decline.
This change matters because it affects those just entering the workforce. Educational institutions need to reorient their curricula for an era of an AI-augmented workforce. Governments must incentivize businesses to hire and train early-career workers. Businesses themselves need to recognize the importance of developing a long-term, AI-savvy workforce, starting with entry-level workers. And students should take responsibility for becoming not just AI fluent but also adept in applying that knowledge across various fields.
It is especially true because recent graduates are facing a softening labor market. The Federal Reserve Bank of New York reported that the unemployment rate for recent college graduates rose to 5.6% in Q4 2025, with underemployment reaching its highest level since the pandemic at 42.5%. Although no single statistic proves AI as the sole cause, hiring has generally declined post-pandemic, and young people are particularly vulnerable.
Recent graduates today often submit hundreds of applications before receiving a single offer, and surveys consistently find elevated rates of anxiety, financial precarity, and burnout among young workers in extended job searches. If AI quietly closes the door on typical early jobs, people will pay the price in delayed independence, postponed family formation, and the sense that their first serious professional efforts have been refused.
Entry-level jobs are part of the economy’s training system. Junior analysts learn which numbers to trust; young software developers learn how systems fail; new marketers learn how customers behave outside neat language; early-career legal and financial staff learn how rules, judgment, deadlines, and human relationships interact. If AI absorbs more of these tasks that once helped train entry-level workers, firms may become more efficient in the short run while society becomes less capable in the longer run.
The right way to improve young workers’ skills is not to tell them simply “Learn to code.” This advice, shaped by decade-long federal initiatives and university expansions, rested on the premise that coding was a stable, scalable skill anyone could learn and apply to a middle-class job. The premise no longer holds true. The layer of work AI handles well—translating specifications into routine code, reproducing standard patterns, debugging predictable errors—is precisely what “learn to code” programs were built around.
Understanding the outputs produced by AI systems will become crucial instead.
To help people develop these skills, we should require universities, community colleges, and professional programs to embed AI literacy, data literacy, prompt-based workflow skills, verification skills, and domain judgment into ordinary degrees. Every graduate should know how to use AI tools, verify their output, understand its limitations, and combine them with human expertise. This matters even for graduates entering occupations that appear relatively safe from AI, such as those in healthcare.
Almost every job contains tasks—drafting, summarizing, scheduling, research, basic data work, routine communication—that now have substantial productivity tools provided by AI. The competition most young workers will face is not human versus machine but colleague versus an AI-augmented colleague. For most young workers, the realistic path to making themselves valuable is not to avoid AI but to become fluent in the technology and combine that with domain judgment, contextual reasoning, and human relationship skills.
To this end, schools should emphasize paid co-ops, apprenticeships, and employer-linked projects so students build judgment in real workplaces before they graduate. Governments should also create targeted tax credits, wage subsidies, and training grants for employers hiring early-career workers into structured AI-augmented roles. The architecture for these conditional, behavior-linked subsidies already exists within US tax policy.
Firms must stop making hiring decisions solely based on short-term cost savings from AI. Young workers are valuable not just for the tasks they perform this quarter but for their future learning, skill formation, institutional memory, and productivity. Entry-level hiring is an investment in the future stock of judgment inside the firm. The most effective AI-augmented senior workforce of the late 2030s will be drawn overwhelmingly from the junior cohort today. Firms that automate away the learning stage may improve their immediate margins but find themselves lacking anyone who understands how their own AI-driven workflows actually behave in a decade.
Students graduating this spring and next face a tough labor market in transition. AI fluency is becoming a commodity, while domain expertise without AI fluency is being outpaced. The combination of the two—AI proficiency with domain knowledge—is genuinely scarce. The mechanical engineer with knowledge of manufacturing alongside AI proficiency; the software programmer with financial services expertise and also an AI whiz—are types of people in high demand.
Georgios Petropoulos, an assistant professor at the USC Marshall School of Business, focuses on how information technologies affect innovation, competition policy, and labor markets.
Key takeaways
- Educational institutions need to reorient their curricula for an AI-augmented workforce.
- Governments should incentivize businesses to hire and train early-career workers.
- Schools should emphasize paid co-ops, apprenticeships, and employer-linked projects.
“`
Stay ahead of AI. Get the most important stories delivered to your inbox — no spam, no noise.
