How does artificial intelligence improve ophthalmology education outcomes?—The mediating role of learning motivation and self-efficacy

Background Artificial intelligence (AI) is increasingly being applied in medical education, yet evidence on how it translates into achievement it is most effective remains limited in ophthalmology. Methods This cross-sectional study examined undergraduate ophthalmology students at a university in Jilin Province, China. A total of 416 students were randomly selected at two time points—November 2024 and March 2025—corresponding to the autumn 2024 and spring 2025 course cohorts, respectively. Survey measures captured AI usage, learning motivation, self-efficacy, final examination scores, and AI literacy. A moderated parallel mediation model was tested using structural equation modeling. Results AI usage was positively associated with academic performance ( β = 0.247, p < 0.001). Indirect effects via learning motivation and self-efficacy were significant, with standardized effects of 0.081 and 0.079, with confidence intervals of 0.059 to 0.103 and 0.017–0.141, respectively. AI literacy strengthened the path from AI usage to learning motivation, interaction ( β = 0.183, p < 0.001). Conclusion AI usage is associated with higher ophthalmology academic performance partly through increased learning motivation and self-efficacy, and AI literacy strengthened the AI usage in the learning motivation path. Given the cross-sectional design, longitudinal validation is warranted.