The explosive rise of ChatGPT may coincide with a turning point for one of America’s fastest-growing college subjects.
Computer science enrollment declined during the 2025-26 academic year for the first time in roughly two decades, according to new research by Jacob Light, an economist at Stanford University’s Hoover Institution.
While the study doesn’t estimate the size of the drop, separate National Student Clearinghouse Research Center data showed undergraduate enrollment in Computer and Information Sciences and Support Services at four-year colleges fell 8.1% year over year in the fall of 2025.
The timing overlaps with the rapid adoption of generative AI, but Light cautioned that the data does not show artificial intelligence caused the decline.
Using a dataset covering more than 50 million course sections offered since 1996 across 1,019 US colleges and universities, Light tracked long-term enrollment patterns alongside the growing capabilities of large language models.
“The most notable recent development is the flattening and reversal in the Computer Science enrollment share trend,” Light wrote. “In 2025-26, Computer Science enrollment share declined for the first time in roughly two decades, interrupting a period of sustained and unusually rapid growth.”
‘Descriptive’ rather than ‘causal’ effect of AI
Computer science had enjoyed years of booming popularity as demand for software engineers and other tech workers surged.
The number of bachelor’s degrees awarded in computer and information sciences more than doubled over the past decade, rising from about 56,000 in 2014 to about 122,000 in 2024, according to data from the National Science Foundation’s National Center for Science and Engineering Statistics.
But the arrival of AI assistants capable of writing code, debugging programs, and completing technical assignments has sparked debate about whether the field’s value proposition is changing.
Light said several explanations could account for the enrollment drop.
“The decline could reflect changes in the education production function as generative AI alters how students complete coursework, changes in students’ expectations about the returns to Computer Science skills, responses to labor market conditions that themselves partly reflect AI adoption, broader weakness in the technology labor market unrelated to AI, or other contemporaneous shifts in student demand,” he wrote.
He added that the data cannot distinguish among those possibilities.
“Therefore, I treat the enrollment patterns as descriptive evidence rather than as an estimate of the causal effect of ChatGPT or AI exposure on student demand,” Light wrote.
The finding is also consistent with data from the National Student Clearinghouse Research Center.
Its Fall 2025 enrollment report found that undergraduate enrollment in Computer and Information Sciences and Support Services at four-year colleges fell 8.1% year over year, from about 659,700 students in 2024 to 606,100 in 2025.
Even so, enrollment remained well above 2022 levels of 574,333 students, suggesting the recent decline followed years of rapid growth.
Whether the slowdown proves temporary or marks a lasting shift in students’ attitudes toward computer science remains an open question.
Light wrote that it will become clearer “as additional cohorts make course and major choices.”
