Medical AI Use Intention Among Medical Students and Faculty and Its Associations With AI Literacy, Perceived Benefits, and Risks: Cross-Sectional Survey Study.

BACKGROUND: AI is being increasingly integrated into health care and medical education. Although AI literacy is considered an essential competency for future physicians and educators, limited evidence exists regarding how perceived benefits, perceived risks, and AI literacy jointly influence AI use intention among different learner groups. OBJECTIVE: This study examined the factors associated with AI use intention among medical students and faculty members and investigated whether the relationships between AI literacy and AI use intention differed according to learner group. METHODS: A cross-sectional self-report survey was conducted among 141 medical students and 94 faculty members (N=235) at a single South Korean medical school. Group-specific linear regression models were estimated for descriptive purposes. To directly test whether associations differed between groups, a pooled regression model included group-by-predictor interaction terms for perceived benefits, perceived risks, AI literacy, and AI use frequency. HC3 heteroscedasticity-robust SEs were used for primary inference, and ordinal logistic regression was conducted as a sensitivity analysis because the 5-point intention outcome showed a marked ceiling effect. RESULTS: Perceived benefits were positively associated with AI use intention in both groups. In the pooled HC3-robust model, the group-by-AI literacy interaction was significant (B=0.361, 95% CI 0.083-0.638; P=.01), whereas the interactions for perceived benefits, perceived risks, and AI use frequency were not significant. The group-by-AI literacy interaction remained significant in ordinal logistic regression (odds ratio 2.48, 95% CI 1.24-4.98; P=.01). In subgroup models, the negative adjusted AI literacy coefficient among students was model dependent, and the positive perceived risk coefficient was not robust to HC3 inference. Among faculty members, AI use frequency remained positively associated with use intention using both HC3 linear regression (B=0.137, 95% CI 0.025-0.249; P=.02) and ordinal logistic regression (P=.02). CONCLUSIONS: Perceived benefits were consistently associated with stronger AI use intention. The adjusted association between AI literacy and use intention differed between students and faculty, although the subgroup-specific student coefficient was sensitive to model specification. These findings provide exploratory evidence that learner context may shape the relationship between AI literacy and broad self-reported AI use intention; they do not establish causal effects or the effectiveness of specific curricular strategies.