Teaching Artificial Intelligence and Language Models in Medical Education

Abstract

The rapid evolution of Artificial Intelligence (AI), especially large language models (LLMs), is transforming medicine and poses new challenges for medical education. This theoretical-reflective article discusses the integration of AI and LLM teaching in the undergraduate medical curriculum, considering ethical, methodological, and practical aspects. Initially, it contextualizes the growing role of AI in healthcare and the need for AI literacy among future physicians, given the incorporation of tools like ChatGPT into clinical practice. We then review recent literature (2024-2025) and relevant guidelines, including the World Health Organization and Brazilian Ministry of Education, to identify educational benefits (such as virtual patient simulations, automated assessment and personalized feedback) and risks (algorithmic bias, response hallucinations, privacy and academic integrity issues). Methodologically, the study is based on a literature review and the author’s teaching experience implementing AI content for 3rd-year medical students. In the results and discussion, we present active teaching strategies, such as the use of prompting in simulated clinical cases and ethical debates, evaluating their perceived effectiveness in developing digital competencies. We conclude by emphasizing the importance of a structured AI curriculum in medicine, including faculty training, interdisciplinary collaboration, and adherence to ethical principles, to train physicians who are capable of harnessing new technologies in a critical and humanistic manner.

Ähnliche Arbeiten