Our Take
College enrollment data is a lagging indicator of labor demand, not a demand signal—students vote after the market has already shifted, so this reflects 2024 hiring weakness, not future scarcity.
Why it matters
If the pipeline of new AI talent is contracting while hiring remains weak, the downstream effect hits hardest at companies that need domain specialists, not generalists. This matters now because hiring managers are still budgeting for 2025 as if 2023 talent economics still apply.
Do this week
Recruiting teams: audit your AI/ML job descriptions for inflated seniority requirements before the next posting cycle, because candidate scarcity may be temporary but your hiring bar is fixed.
Goldman data reveals enrollment decline in AI programs
Goldman Sachs published enrollment data showing college students are reducing enrollment in artificial intelligence and machine learning degree programs. The firm tracked this trend across U.S. universities and found the shift happened in the last academic year, consistent with broader post-hype contraction in tech hiring (per Goldman Sachs analysis).
The decline aligns with hiring freezes and reduced spending that began in late 2023 and accelerated through 2024. Companies that aggressively recruited AI specialists during the ChatGPT boom have since scaled back headcount, making job placement less certain for graduates.
Talent pipelines move slowly; hiring markets move fast
Enrollment decisions lag job market conditions by 18 to 24 months. Students who started AI programs in 2022 and 2023 did so during peak hype and hiring. Graduation cohorts hitting the job market now face materially different conditions: lower starting salaries, stricter hiring filters, and fewer "AI engineer" roles that don't also require production systems expertise.
The risk is not talent scarcity in 2025 or 2026. The risk is mismatch. Companies still recruiting for "machine learning engineer" roles as if specialized AI training is the primary requirement will struggle. The talent that exists is either already employed or has deprioritized AI as a career focus, choosing software engineering, data engineering, or adjacent fields instead.
For hiring teams, this means the window to poach AI talent from competitors is narrowing. New graduates will be fewer and less motivated. The bar for internal retooling and cross-functional hiring goes up.
Check your hiring assumptions before 2025 planning
If your team assumes you can wait out the market and snap up AI talent in Q2 2025 because the pipeline is still warm: you're wrong. Enrollment is cooling now. Resumes are tightening. Begin recruiting and retention conversations this month. For teams that cannot afford to compete on salary, shift focus to roles where AI is a tool, not the title. For companies building agent systems and production ML platforms, start internal upskilling programs now rather than betting on external hires to close skill gaps.