by Coinstrooper Crypto Articles
Quick Take
  • Goldman Sachs has found the first clear evidence that college students are steering away from majors exposed to AI.
  • Enrollment in computer science and programming each fell by more than 10% in the 2025-2026 academic year.
  • The retreat is part of a wider repricing of higher education.
  • Students, employers, and even business schools are now judging degrees by how well they survive automation.

What Happened

Goldman Sachs has found the first clear evidence that college students are steering away from majors exposed to AI. Enrollment in computer science and programming each fell by more than 10% in the 2025-2026 academic year.

The retreat is part of a wider repricing of higher education. Students, employers, and even business schools are now judging degrees by how well they survive automation.

Students are Dropping AI-Exposed College Majors

Market Context

These students are reading a market where Goldman estimates AI keeps cutting US jobs each month. Mei expects the response to come faster than in past technological shifts.

Applications keep falling 20% to 30% this cycle, yet none of the cutting schools sit in the top 20. Elite programs still command their price, while the discounters are those that sell technical skills AI now supplies cheaply.

A new $100,000 cap on graduate borrowing takes effect in July. The frustration is visible on campus, where Stanford graduates walked out on Google’s chief executive this month.

Not everyone reads the shift as a clean escape. Many economists argue that AI will transform most jobs rather than delete whole professions.

Why It Matters

Enrollment in fields tied to low AI displacement risk grew about 3% on average, led by healthcare and engineering.

He mapped more than 180 majors to their jobs using Census survey data from 2022 to 2024. He then scored each occupation for automation risk.

Computer science, statistics, and quantitative business majors carried the highest risk. Pharmacy, nursing, and education ranked among the safest.

“Historically, such adjustments have taken a few years… But the current adjustment may be unfolding more quickly, given the heightened salience of AI disruption,” Fortune reported, citing Mei.

ServiceNow CEO Bill McDermott sells the AI agents doing the automating. He warns that new-graduate unemployment could climb into the mid-30s as those agents absorb early-career work.

AI May Reshape Work, Not Erase It

Computer science has cycled before, sliding after the dot-com crash before rebounding to record highs. The World Economic Forum projects AI could create 170 million new roles by 2030.

The bigger divide may be skill, not major. NVIDIA CEO Jensen Huang made the point at the Milken Institute.

Details

The fall reverses years of booming computer science enrollment, a pattern absent from the data before the 2024-2025 year.

The drop shows up on individual campuses. At Arizona State University, bachelor-level computer science enrollment fell about 14% between fall 2024 and fall 2025.

At Washington University in St. Louis, the share of computer science majors dropped 16% over two years.

Rather than poll students on their fears, Goldman Sachs economist Pierfrancesco Mei studied where graduates actually work.

The Squeeze Runs from Entry Level to Business School

The anxiety has a basis. The Federal Reserve Bank of New York put recent-graduate unemployment near 5.7% at the end of 2025. Underemployment hit 42.5%, the highest since 2020.

Employers are automating the entry-level rungs that once required trained graduates. AI now claims entry-level roles across tech, and Block alone cut about 4,000 jobs while tying the decision to automation.

The repricing has reached graduate school. Mid-tier programs are slashing MBA tuition by up to 50%, with Purdue cutting its online MBA from $60,000 to $36,000.

Goldman itself frames the data as adaptation, noting young workers have adjusted to past technology waves faster than older ones.