← The Wire

Top 5 · 2026-04-14 · source-backed

CS Enrollment Drops 11.2%. Students Aren't Leaving Tech. They're Leaving CS.

Confidence
source-backed
Sources
2
Redaction
redacted

Story

Computer science enrollment fell 11.2% in the 2025-2026 school year, the steepest decline of any major, according to the National Student Clearinghouse via the Washington Post. A CRA survey found 62% of computing departments reporting declines. But here's the thing everyone's missing: students aren't fleeing tech. They're fleeing to AI.

MIT's "AI and Decision-Making" major is now the second-largest on campus. USF enrolled 3,000+ students in a new AI/cybersecurity college. The migration pattern is clear. Students look at entry-level coding jobs and see a future where agents handle the work. They look at AI-specialist roles and see growth. They're making a rational economic decision.

This connects to the Google adoption story in a way that should worry every engineering manager. The incoming talent pipeline is optimizing for AI skills, not traditional CS fundamentals. In three years, your junior hire pool will know how to prompt-engineer and fine-tune but might struggle with data structures and systems programming. Whether that's a problem depends entirely on how much of your stack is agent-orchestrated by then.

The Stanford AI Index adds a sentiment dimension: Gen Z excitement about AI collapsed from 36% to 22% year-over-year, while anger rose from 22% to 31%. They're not excited about AI. They're scared of it. And they're adapting by pivoting their education toward it. Fear as a career driver.

What to do about it: If you're hiring, update your job descriptions. "Computer Science degree required" is going to filter out a growing pool of AI-native candidates who chose different programs. If you're a developer, the skills that matter most are the ones AI can't replicate well: systems thinking, architecture decisions, and taste. The 11.2% drop is the market telling you where the puck is going.


Related stories

Each link below shares sources, entities, or timing with this story.

  1. Shared entity: Stanford AI Index / Same source / What happened next / Tension

    Stanford AI Index 2026: Agent Task Success 12% to 66%.

    Both cover Stanford AI Index; cite the same source (Stanford AI Index); picks up the Stanford AI Index thread on 2026-04-21.

  2. Shared entities / Same source

    Google Has the Same AI Adoption Curve as a Tractor Company

    Both cover Google, Stanford AI Index; cite the same source (Stanford AI Index).

  3. Shared entities / Same source domain / Earlier coverage

    Anthropic-Pentagon-OpenAI Triangle Dominates the News Cycle

    Both cover Google, Washington Post; reported by the same outlet (washingtonpost.com); earlier Google coverage from 2026-02-28.

  4. Shared entities / What happened next / Tension

    An Open-Source Model Just Beat GPT-5.4 and Opus 4.6 on the Hardest Coding Benchmark

    Both cover Google, MIT; picks up the Google thread on 2026-04-20; pushes against this story (against).

  5. Shared entities / What happened next

    Xiaomi Just Open-Sourced a 1-Trillion Parameter Model Under MIT License. It Built a Compiler in 4 Hours.

    Both cover Google, MIT; picks up the Google thread on 2026-05-01.

  6. Shared entities / Earlier coverage

    Cisco Open-Sources A2A Protocol Scanner

    Both cover Google, MIT; earlier Google coverage from 2026-02-23.

  7. Shared entity: MIT / Shared topic / What happened next

    Microsoft SkillOpt Trains Markdown Skills for Frozen LLMs. 52/52 Benchmark Wins. Your CLAUDE.md Can Be Optimized Like a Neural Network.

    Both cover MIT; overlapping topics (aren, skill); picks up the MIT thread on 2026-06-01.

  8. Shared entity: Google / What happened next / Tension

    Altman Pitches US-Led Forum to Set AI Standards

    Both cover Google; picks up the Google thread on 2026-07-02; pushes against this story (against).

Source trail

Entities

Provenance

AI generated
no
Story unit
2026-04-14-cs-enrollment-drops-11-2-students-aren-t-leaving-tech-they-re-leaving-cs
Labels
source-backed, canonical briefing excerpt