Is AI Triggering the Third Major Decline in Computer Science Majors?

Venture capitalist Deedy Das recently posted on X that the United States is entering the "third major decline" of computer science (CS) majors, echoing a Washington Post deep‑dive titled "The Hottest College Majors Hit a Wall. What’s Happening?".

Historical enrollment trends

From 2008 to 2024, four‑year U.S. universities awarded CS degrees about five times more than in 2008, a growth rate more than double that of the second‑most popular major, sports science.

In the fall of 2025, enrollment in four‑year CS programs fell by 8.1%, the largest single‑year drop recorded since 2020, pushing CS from fourth to sixth place in the national major rankings.

First and second declines

First decline (1984‑1994): The introduction of the Apple II and IBM PC in the late 1970s spurred a surge in CS enrollment that peaked around 1984, then fell 42% by 1994 because universities lacked sufficient faculty to accommodate the influx, forcing enrollment caps.

Second decline (2001‑2007): The dot‑com bubble drove a 15% annual increase in CS graduates from 1997 to 2003. After the 2001 burst, enrollment slipped for several years, even though the tech industry recovered by 2004. Concerns then existed that software jobs might be outsourced to India—a fear structurally similar to today’s AI‑driven programmer‑job anxiety.

The current, AI‑driven decline

National Student Clearinghouse data show an 8.1% drop in CS enrollment for the 2025‑26 academic year. A survey of 133 universities by the Computing Research Association reports that 62% of respondents observed declining CS admissions, and the University of California system recorded its first post‑bubble enrollment decline.

Federal Reserve Bank of New York data for 2025 indicate a 6.1% unemployment rate among CS undergraduates, following massive tech layoffs of over 150,000 in 2024 and another 100,000 in 2025, creating an oversupply where experienced developers compete directly with new graduates.

Washington Post interviews a freshman, Gavin O’Malley, who chose mechanical engineering over CS because of perceived competition, illustrating how peer pressure and viral memes deter applicants.

Nevertheless, many “lost” CS students are shifting to related fields such as data science, AI, robotics, and cybersecurity. The University of California, San Diego— the only UC campus with a growing enrollment—is also the sole UC school offering an undergraduate AI major. Nationwide there are now 193 AI bachelor programs and 310 AI master’s programs.

The ACM Communications review "The Outlook for Computer Science Education" attributes the crisis to three factors: AI being used to justify cuts in entry‑level hiring, large‑scale tech layoffs amid economic uncertainty, and a 2022‑2023 enrollment surge that flooded the market just as demand contracted.

An unnamed CS professor warned that a typical CS degree from a regular university now offers a much lower chance of landing a software job than two or three years ago, reserving opportunities for the top‑performing students.

Is this a decline or a transformation?

Historical patterns show CS enrollment reacting strongly to labor‑market sentiment. This cycle differs because AI’s impact on programming jobs is concrete, not speculative.

The core question remains whether AI will eliminate demand for software engineers or merely reshape the role’s content and level.

Carnegie Mellon University’s undergraduate dean Tom Cortina, while acknowledging AI’s impact, remains optimistic, calling the situation a temporary dip. In contrast, University of Washington’s dean Magdalena Balazinska worries that students are scared off by headlines of layoffs and AI‑driven job loss, despite genuine interest in CS.

Gavin O’Malley ultimately chose mechanical engineering not out of AI fear but because the "competition pressure was too high," underscoring that CS’s past success has made the field overcrowded, with AI pushing the pressure to a tipping point.

Engineers who can collaborate with AI rather than be replaced by it will continue to be highly sought after, while CS enrollment numbers will serve as a market‑sentiment barometer rather than a definitive judgment of the discipline’s value.