AI‐Powered Virtual Patients in Health Professions Education: Learner Engagement and Knowledge Application

PURPOSE/OBJECTIVES: Immersive educational technologies, including applications of artificial intelligence in education (AIED), are increasingly used to develop clinical skills in health professions education. This study evaluated the feasibility and perceived value of integrating an artificial intelligence (AI)-enabled virtual patient platform into a required first-year dental medicine course and explored AI-assisted analysis of student interview transcripts. METHODS: Seventy-eight first-year Doctor of Dental Medicine (D.M.D.) students completed three faculty-developed virtual patient interviews aligned with course objectives. Students then completed a voluntary survey assessing usability, learning experience, and engagement/motivation on a 5-point Likert scale, along with preferences for interaction modality and avatar style. Exploratory qualitative analyses of interview transcripts were conducted across six domains: empathy, professionalism, thoroughness, attention to behavioral health, interview skill, and potential bias by race/sex. Descriptive statistics summarized responses. RESULTS: Mean ratings were high for usability (M = 4.22, SD = 0.24), learning experience (M = 4.47, SD = 0.09), and engagement/motivation (M = 4.42, SD = 0.04). For avatar preference, 33% of students preferred realistic avatars, 22% preferred cartoon avatars, and 45% had no preference; most (89%) preferred text-based over voice-based interaction. Exploratory analysis of 217 transcripts suggested generally strong performance with identifiable areas for improvement, including subtle differences in rapport by patient race and sex. CONCLUSIONS: These findings support the feasibility and perceived value of AI-enabled virtual patient simulation in early dental education. Controlled studies are needed to evaluate effects on learning outcomes and skill development relative to existing instructional approaches.