Implementing AI-supported, educator-led simulation to support motivational interviewing practice in pharmacy students.

OBJECTIVE: Motivational interviewing is an important skill taught to student pharmacists, however providing adequate practice and quality feedback in the didactic curriculum is challenging. Generative artificial intelligence (AI)-supported simulations may address some of these issues. The purpose of this study was to evaluate the impact of an educator-led, AI-supported simulation (ATLAS) on pharmacy students' confidence in motivational interviewing (MI) skills, and to determine the optimal timing for its integration, either before or after peer-based MI practice, to maximize impact of MI learning interventions. METHODS: Students were divided into two groups: Pathway A used ATLAS after doing practice interviews with peers; Pathway B used ATLAS before doing practice interviews with peers. Surveys measured MI confidence and attitudes at baseline and at key timepoints before and after ATLAS use and peer interviews, enabling comparison of confidence trajectories across both pathways. RESULTS: Students in Pathway B, who completed ATLAS followed by a peer interview, demonstrated gains in MI confidence (P = 0.0063). In contrast, Pathway A who completed a peer interview followed by ATLAS, did not show robust improvements in confidence (P = 0.68). These findings suggest that using ATLAS prior to peer interaction may better support confidence in MI skill acquisition. Qualitative feedback revealed that among students who commented on practice preferences, most favored ATLAS over peer interviews, citing its low-stakes environment and flexibility. CONCLUSION: Educator-led, AI-supported simulation using ATLAS offers scalable, time-efficient and a pedagogically sound approach to enhancing MI training. Students who engaged with ATLAS prior to peer interviews reported the highest confidence gains, suggesting that early exposure to AI-supported practice may enhance learners' confidence and preparedness for MI conversations. These findings support the integration of structured, low-stakes simulation tools as valuable precursors to peer-based learning, enhancing confidence development during MI training in pharmacy students.