Generative AI Use, Perceived Learning, and Perceived Academic Performance Among University Students
Main Article Content
Abstract
Generative AI tools are increasingly embedded in university study practices, yet institution-level evidence remains limited, particularly in underrepresented regions. This study examined the frequency and patterns of AI use among undergraduates at Maaref University of Applied Sciences in northern Syria and explored its relationship with perceived understanding, academic performance, attitudes, and concerns. An online questionnaire was completed by 58 students from seven faculties, and data were analyzed using descriptive statistics, correlations, t-tests, ANOVA, and multiple regression. Most respondents (81.0%) used AI daily or several times weekly, with ChatGPT the most commonly reported tool. AI use was positively associated with perceived understanding (r = .677, p < .001) and academic performance (r = .765, p < .001). The regression model explained 72.1% of the variance in perceived academic performance. Although causal conclusions cannot be drawn, the findings support structured AI-literacy training, verification practices, and clear institutional guidance for responsible academic use.