Zimbabwe University Study: About One in Three Students Showed AI Dependency Patterns

Published: September 22, 2025

A paper published on August 25, 2025 reports that about one in three students at one Zimbabwean university showed what the author calls addictive patterns of generative AI use, and that more heavily affected students had lower grades. The paper is Generative AI dependency: the emerging academic crisis and its impact on student performance: a case study of a university in Zimbabwe, by Zvinodashe Revesai, in the journal Cogent Education (volume 12, article 2549787). This post summarizes what its abstract says. It is not a review of the full paper, which we could not open. Please read the original for details.

What the study did

The author used a mix of methods at a single university: surveys, interviews, usage tracking and academic records. The sample was 248 undergraduate students. It is a case study of one institution, so the results cannot be applied to all students in Zimbabwe, or anywhere else, without more research.

What it found

According to the abstract:

• About 32.7% of the students showed addictive patterns of generative AI use. The paper uses the word “addiction” for this. That is the author’s label. It is not a medical diagnosis.
• Students in that group averaged about 18.3 AI interactions a day, and 65.8% of them reported failed attempts to cut back.
• Heavily affected students kept using AI even though they recognized academic harm.
• Students with more severe patterns had lower academic results. The mean grade point average of the most heavily affected group was 0.41 points lower than that of students without these patterns.
• The author’s statistical analysis pointed to three possible routes: weaker critical thinking, weaker development of writing skills, and less learning of course content.
• Interviews pointed to local conditions. Unreliable infrastructure produced bursts of intense use, money pressures made AI a stand-in for costly resources, and limited institutional resources made AI more appealing.

What it does not show

This is a correlation. The study found that heavier AI use and lower grades go together. It does not show that AI use caused the lower grades. Students who were already struggling may also have turned to AI more. Self-reported answers can also be unreliable.

The 33% figure is also a count of students who met this paper’s own criteria for “addictive patterns.” Other studies use other definitions and find other rates, so it is not safe to compare them directly. There is no agreed clinical definition of AI addiction or dependence.

If this sounds like you

If you are a student who feels you cannot do your work without AI, or you feel anxious when you cannot use it, you are not alone. It may help to try some of your work first without it, then use AI to check or improve what you did. Your school’s counseling or writing center can also help. If you want to look at your own use, our free AI addiction assessment is a short self-check for adults. It is not a diagnosis. The check-in is free and private to you.

If you are feeling overwhelmed, you can call or text 988 in the US, or text HOME to 741741 to reach Crisis Text Line.

Did a university study find 33% of students are dependent on AI?

A single-university study in Zimbabwe reported that about 32.7% of 248 undergraduates showed addictive patterns of generative AI use, by the author’s own measure. It cannot be generalized to students elsewhere.

Does AI use cause lower grades?

The study found that heavier AI use and lower grades went together, with a mean GPA 0.41 points lower for the most heavily affected group. That is a correlation. It does not prove that AI use caused the lower grades.