AI-based clinical simulations, including virtual patients and conversational chatbots, offer significant potential for enhancing undergraduate medical students’ clinical reasoning and communication skills, according to a recent scoping review. Published on November 16, 2026, the review analyzed 30 empirical studies from 2015 to 2025. These AI technologies provide interactive, scalable platforms for repeated practice, allowing students to navigate complex decision-making scenarios without posing risks to real patients. As clinical errors remain a challenge, such simulations are a strategy for potentially enhancing decision-making training.
The research team followed the Joanna Briggs Institute guidance and the PRISMA-ScR checklist to ensure methodological rigor throughout the synthesis. Their comprehensive search spanned major databases including PubMed, Scopus, Web of Science, and BIREME, and was further bolstered by manual backward and forward citation tracking to ensure full coverage of the available evidence. Database searching identified 777 records, yielding 21 eligible studies, while citation tracking identified nine additional studies for a total sample of 30.
Among the technologies evaluated, AI-enabled virtual patients were frequently researched in the included studies. Conversational simulations, including chatbots and large language models, were also prominent in the literature. Fewer studies investigated the impacts of more immersive environments, such as virtual reality, or the use of social robotics.
The review noted that many included studies described favourable findings across evaluated learning outcomes. Included studies reported outcomes such as proficiency in history-taking and foundational communication skills. Frequently reported outcomes also included clinical reasoning, decision-making, and OSCE-type performance. Reported outcomes were described as favourable, particularly in interventions that incorporated deliberate practice and structured feedback.
Despite these findings, the authors emphasize that the current body of literature is marked by significant methodological heterogeneity. The design, duration, and metrics used to measure learning outcomes varied widely between the 30 included studies. This diversity in study design and outcome measurement limits claims regarding the universal efficacy of these tools. Most evaluations focus on short-term gains, leaving questions regarding the long-term impact on clinical performance.
Ultimately, the study suggests that while AI-based simulation is a valuable asset, its role in medical education warrants cautious optimism and critical oversight. Careful curricular integration and critical appraisal of implementation are warranted.
Subject of Research: Integration of artificial intelligence and clinical simulation in undergraduate medical education training.
Article Title: Artificial intelligence–based clinical simulation in undergraduate medical education: contributions of virtual patients, chatbots, and immersive environments – a scoping review
Article References: Hernandez Rincon, E. H., Pesantes Bonilla, M., Rincon Franco, S., Suarez Sanmiguel, J. A., & Orozco Rodriguez, L. Y. (2026). Artificial intelligence–based clinical simulation in undergraduate medical education: contributions of virtual patients, chatbots, and immersive environments – a scoping review. BMC Medical Education. https://doi.org/10.1186/s12909-026-10793-8
Image Credits: AI Generated
DOI: 10.1186/s12909-026-10793-8
Keywords: Artificial Intelligence, Clinical Simulation, Medical Education, Undergraduate Training, Virtual Patients, Chatbots, Natural Language Processing, Clinical Reasoning, Communication Skills, OSCE Performance, Deliberate Practice, Educational Technology, Scoping Review, Medical Training, Scalable Feedback, Educational Outcomes, Interactive Learning, Adaptive Learning, Curriculum Integration, Healthcare Technology
News Source: Courtney Benton. (November 16, 2026). AI-Based Clinical Simulation May Support Medical Training, Scoping Review Finds. Scienmag.



