The Application of AI Patients in Medical Education: A scoping review

Objective: This scoping review will systematically map the extent and nature of existing evidence on the application of AI patients in medical education. The aim is to clarify the application scenarios, implementation modes, core effects, influencing factors and existing challenges of AI patients in medical education and identify the research gaps in this field. Introduction: Standardized Patients (SPs) are a cornerstone of medical education for training clinical skills and communication. However, human SPs are resource-intensive and limited by availability. The integration of artificial intelligence (AI) into medical education, particularly through "AI patients" (including virtual standardized patients, conversational agents, and generative AI-driven patient simulators), represents a transformative shift. These technologies offer scalable, consistent, and safe clinical training environments, which effectively make up for the shortages of human SPs. However, the existing research on the application of AI patients in medical education is scattered, covering multiple fields such as clinical skill training, interactive learning, and assessment, with inconsistent research focuses and no systematic scoping review to sort out the application status, core effects, existing challenges and research gaps. This makes it difficult for medical educators, researchers and relevant institutions to comprehensively grasp the overall situation of this field, which restricts the standardized and in-depth application of AI patients in medical education. Methods: This scoping review will follow the Joanna Briggs Institute (JBI) methodology for scoping reviews and be reported according to the PRISMA extension for Scoping Reviews (PRISMA-ScR).We will search electronic databases including PubMed, Scopus, Web of Science, CNKI, WanFang Data and CQVIP, from inception to 2026, supplemented by grey literature searches and reference list screening. The PCC (Population-Concept-Context) framework will guide eligibility criteria: Population—medical students and healthcare professionals; Concept—AI patients, virtual patients, conversational agents, chatbots, or AI-driven simulations used for educational purposes; Context—medical education and training settings. Studies that do not involve AI technology (only traditional virtual patient systems without AI empowerment), studies that focus on AI applications in clinical practice rather than medical education, and non-English and non-Chinese studies will be excluded. Two reviewers will independently screen titles/abstracts and full texts, with conflicts resolved through discussion or a third reviewer. Data will be extracted and managed in Zotero. A summary of results will be presented in tabular and diagrammatic form with a narrative synthesis. Expected Outcomes: This scoping review will clarify the main application scenarios of AI patients in medical education, summarize their core effects, analyze the main challenges, and identify the research gaps in this field. Ethics and Dissemination: Ethical approval is not required as this study synthesizes published data. Findings will be disseminated through a peer-reviewed journal publication and presentations at relevant academic conferences.