The rapid emergence of ChatGPT appears to align with a potential shift in one of the nation's most popular college majors.

Computer science course enrollment dropped in the 2025-26 school year for the first time in about twenty years, based on fresh findings from Jacob Light, an economist at Stanford University's Hoover Institution.

Although the analysis does not quantify the decrease, separate figures from the National Student Clearinghouse Research Center indicated that undergraduate enrollment in Computer and Information Sciences and Support Services at four-year institutions declined 8.1 percent from the prior year in fall 2025.

The period coincides with widespread use of generative AI, yet Light emphasized that the figures do not prove artificial intelligence triggered the slowdown.

Drawing on records of more than 50 million course sections since 1996 from 1,019 US colleges and universities, Light examined enrollment trends against advances in large language models.

"The standout recent change is the leveling off and reversal in the Computer Science enrollment share pattern," Light noted. "In 2025-26, Computer Science enrollment share fell for the first time in roughly two decades, ending a stretch of steady and unusually fast expansion."

'Descriptive' rather than 'causal' effect of AI

Computer science had seen strong growth for years as demand for software engineers and related roles increased sharply.

The count of bachelor's degrees in computer and information sciences more than doubled in the past ten years, climbing from roughly 56,000 in 2014 to about 122,000 in 2024, per National Science Foundation data.

Still, the rise of AI tools that can generate code, fix errors, and handle technical tasks has prompted discussion about whether the major's appeal is evolving.

Light outlined multiple factors that might explain the enrollment dip.

"The decline could stem from shifts in how coursework is completed due to generative AI, changes in student views on the value of Computer Science skills, reactions to job market conditions influenced by AI adoption, general softness in the tech sector unrelated to AI, or other changes in student preferences," he wrote.

He noted the data cannot separate these influences.

"As a result, I view the enrollment patterns as descriptive evidence rather than a measure of the causal impact of ChatGPT or AI exposure on student choices," Light stated.

The observation aligns with National Student Clearinghouse Research Center findings.

Its Fall 2025 report showed undergraduate enrollment in Computer and Information Sciences and Support Services at four-year colleges fell 8.1 percent year over year, from about 659,700 students in 2024 to 606,100 in 2025.

Even with the drop, enrollment stayed above 2022 levels of 574,333 students, indicating the recent change followed a period of strong growth.

Whether the slowdown is short-lived or signals a longer-term change in student interest in computer science is still unclear.

Light added that more clarity will come "as new groups of students select their courses and majors."