Evaluating Perceptions of AI-Simulated Patients as an Adjunct to Peer-Based Clinical Case Practice in Undergraduate Medical Education: A Mixed Methods Survey Study

AI-simulated patients are an emerging educational tool that may provide learners with accessible opportunities to practise clinical consultations independently.This study explored participants' perceptions of AI-simulated patients compared with traditional peer-based practice, focusing on perceived effectiveness, acceptability, usability, and self-reported confidence.A questionnaire-based study was conducted involving final-year medical students and Foundation Year 1 (FY1) doctors.Participants used AI-simulated patient cases and completed a questionnaire comprising quantitative ratings and qualitative free-text responses about their experiences.Quantitative responses were analysed descriptively, while qualitative responses were examined to identify key themes in participants' perceptions.Participants reported their experiences of the usability and perceived educational value of AI-simulated patients, including their perceived benefits and limitations compared with peer-based practice.Qualitative feedback provided further insight into the perceived advantages of independent practice and immediate feedback, alongside limitations in human interaction and communication.These findings reflect participants' self-reported perceptions rather than objectively measured changes in knowledge or clinical performance.AI-simulated patients may offer a useful adjunct to traditional peer-based practice by providing an accessible and flexible means of practising clinical consultations.Although participants' perceptions suggest potential benefits for self-directed learning and confidence, these outcomes were not objectively assessed.Further research incorporating objective measures of knowledge, clinical performance, and communication skills is required to establish the educational effectiveness of AI-simulated patients.