Computer science enrollment crashed for the first time since the dot-com bust, with graduate programs down 15% and the UC system falling 6% system-wide in 2025. The Computing Research Association reports 62% of computing programs experienced enrollment declines last fall. This isn’t a tech exodus—it’s a strategic migration. Students are abandoning generalist CS degrees for specialized programs in AI, cybersecurity, and computer engineering, driven by AI automation fears and a brutal entry-level job market where junior developer hiring has plunged 73% since 2022.
For two decades, a CS degree was the golden ticket to a $120k FAANG job. That promise is dead. Students don’t trust it anymore, even though CS grads still earn $81,535 starting salaries with 716% ROI. The issue isn’t economics—it’s psychology.
The Numbers Tell the Story: First Decline in 20 Years
The data is unambiguous. The Computing Research Association’s Fall 2025 survey shows 62% of computing programs reporting enrollment declines, with departments losing an average of 11-15% of students. Graduate programs took the worst hit at 15% down, while the UC system fell 6% after dropping 3% in 2024. This is the first sustained decline since the dot-com crash ended in 2003.
Only 13% of programs reported growth—almost exclusively in cybersecurity, AI, and computer engineering. Traditional computer science, software engineering, and information systems are hemorrhaging students. The 20-year streak of continuous growth is over.
Moreover, this isn’t a blip. The dot-com parallel is ominous: CS enrollment crashed from 2003 to 2008 even though tech jobs recovered by 2004. Students react to headlines and fear, not labor market data. If history repeats, we’re in year two of a five-year enrollment collapse.
The Migration to Specialization
Students aren’t abandoning tech careers—they’re fleeing the generalist CS degree for specialized programs that promise clearer value. AI majors are exploding. MIT’s “AI and decision-making” program is now the second-largest major on campus. UC San Diego was the only UC campus with CS enrollment growth because it launched a dedicated AI major in Fall 2025. Universities including USC, Columbia, Pace, and New Mexico State are all rushing to launch AI degrees.
The message is clear: “AI major” means I’m building AI, not being replaced by it. “Cybersecurity major” signals an automation-resistant career path. “CS major” raises the question: will this job exist when I graduate?
This is a vote against generalist education, not against technology itself. Students want specialized skills with clear return on investment. Specialization beats flexibility when the future feels uncertain.
Why They’re Running: AI Fears Meet Entry-Level Brutality
Two forces converge to drive students away. First, AI automation anxiety—64% of pessimistic CS majors cite generative AI as a factor in their career outlook. They’re watching AI write code and wondering if their future jobs will exist. Second, the entry-level market has become brutal. Junior developer hiring crashed 73% since 2022, while starting salaries for the lucky few who land roles rose 7% to $81,535.
CS graduates now face unemployment rates nearly double that of philosophy majors. Read that again. The “learn to code” generation is struggling to find work while liberal arts majors walk into jobs. The irony is crushing.
Parents who once pushed kids toward CS are now steering them away. “They’re reflexively directing children toward majors that seem more resistant to AI automation, including mechanical and electrical engineering,” reports Marketplace. The parental panic is real. Meanwhile, students ask: will AI take my job before I even graduate?
The 716% ROI Paradox Nobody Talks About
Here’s what makes this enrollment crash bizarre: CS degrees still deliver the highest return on investment in higher education. The numbers are extraordinary—716% ROI, $81,535 starting salary (up 7% from 2025), $145,080 median mid-career salary, and a six-year payback period faster than any other field. The Bureau of Labor Statistics projects 22% job growth for CS and IT occupations through 2033, far outpacing the average.
Yet enrollment is cratering. This isn’t an economic crisis—it’s a psychological one. The fundamentals are strong, but the perception is toxic. Students don’t believe the promise anymore. They see 73% fewer junior roles, AI writing code, and tech layoff headlines. Rational? Maybe not. Understandable? Absolutely.
This mirrors the dot-com crash exactly. Tech employment recovered by 2004, salaries stayed high, but enrollment kept falling until 2008. Student reactions lag market reality by three to five years. We’re likely in year two of that lag right now.
History Repeating: The Dot-Com Lag
We’ve seen this movie before. When the dot-com bubble burst in 2000-2001, CS enrollment didn’t immediately crash. It peaked in 2003, then plunged for five years despite the tech job market recovering by 2004 and employment surpassing pre-crash levels by 2005. Students and parents react to headlines—”tech bubble pops,” “AI takes jobs”—not labor market fundamentals.
The pattern suggests enrollment could continue falling through 2027-2029 even if the job market remains strong. This creates a paradox: strong demand for CS grads plus shrinking supply equals rising salaries and easier hiring for those who graduate. But “CS enrollment crash” headlines scare more students away, reinforcing the cycle.
For companies, this means talent shortages ahead. The students who should be entering CS programs in 2026 would graduate in 2030. That pipeline is drying up.
Not a Crisis, a Correction
The “crisis” narrative is wrong. This is a healthy market correction. CS enrollment grew 15% per year from 1997 to 2003, ballooning beyond sustainable levels. The 2010s “learn to code” boom pushed too many students into generic programs with promises of guaranteed wealth. Now the market is correcting to sustainable levels, and students are choosing specialized skills over generalist degrees.
Are students making a mistake? Maybe. Specialized degrees age poorly—today’s hot AI major could be tomorrow’s automated skill. CS fundamentals like algorithms, systems, and networks endure longer than trendy specializations. But the students choosing cybersecurity and AI aren’t irrational. They’re responding to market signals: specialization offers clearer value and lower anxiety in an uncertain future.
For the industry, this correction is healthy. It removes students who only entered CS for money, leaving those genuinely interested in technology. For parents panicking about AI: take a breath. CS degrees still deliver 716% ROI and $81,535 starting salaries. The fundamentals matter more than headlines. For current students: specialize early, build portfolios, and focus on skills that are hard to automate—security, infrastructure, AI research, not generic full-stack development.
Key Takeaways
- CS enrollment is experiencing its first sustained decline since the dot-com crash, with 62% of programs reporting drops and graduate programs down 15%
- Students aren’t leaving tech—they’re migrating to specialized AI, cybersecurity, and computer engineering programs for clearer career paths and lower AI automation anxiety
- The enrollment crash is driven by AI automation fears (64% of pessimistic majors cite AI) and a brutal entry-level market where junior hiring fell 73% since 2022
- CS degrees still deliver 716% ROI and $81,535 starting salaries, but students no longer trust the “learn to code, get rich” promise—this is psychology, not economics
- The dot-com pattern suggests enrollment could keep falling through 2027-2029 even if job markets stay strong, creating talent shortages and higher salaries for those who do graduate
The era of the generalist developer degree is ending. What replaces it: specialized, automation-resistant skills and programs with clearer return on investment. Students are voting with their feet, and the market is listening.
