Health Literacy and Acceptance of AI/XR-Enabled Telemedicine Among Romanian Medical Students: A Cross-Sectional Survey

Background and Objectives: AI- and extended reality (XR)-enabled telemedicine is increasingly relevant to clinical training, yet evidence from Central and Eastern Europe is limited. We assessed Romanian medical students’ acceptance of AI/XR-enabled telemedicine and examined whether health literacy moderates the association between AI/XR knowledge and acceptance. Methods: We conducted an anonymous cross-sectional online survey of 212 medical students (years 1–6) at a single Romanian university (March 2024–June 2025). Acceptance was measured using a study-specific Acceptance Index (mean of three 4-point items: trust in AI-assisted recommendations, perceived improvement in telemedicine quality with AI/XR, and willingness to participate in AI/XR-enabled teleconsultations; internal consistency acceptable, Cronbach’s α ≈ 0.8). Health literacy was assessed with the validated Romanian version of the European Health Literacy Survey Questionnaire (HLS-EU-Q16). We performed group comparisons, Spearman correlations, multivariable and hierarchical regression with a Knowledge × Health Literacy interaction, and k-means clustering. Results: Participants had a mean age of 22.5 ± 1.9 years; 66.0% were female. Overall acceptance was high (2.9 ± 0.6). Acceptance was higher in clinical vs. preclinical years (3.1 ± 0.6 vs. 2.8 ± 0.5; p < 0.001; Cohen’s d ≈ 0.55) and in prior AI/XR users vs. non-users (3.2 ± 0.5 vs. 2.7 ± 0.6; p < 0.001; d ≈ 0.89). Knowledge correlated strongly with acceptance (ρ = 0.68; p < 0.001). In multivariable models (R2 = 0.61), knowledge, perceived educational value, prior AI/XR use, and clinical stage independently predicted acceptance, whereas privacy concern and gender did not. Health literacy was sufficient in 64.2% and significantly moderated the knowledge–acceptance link (interaction p = 0.012). Conclusions: Romanian medical students report favorable acceptance of AI/XR-enabled telemedicine. Findings support curriculum integration that combines structured AI/XR teaching with literacy-sensitive scaffolding to ensure knowledge translates into informed, critical acceptance across student subgroups.