Large language models (LLMs), such as ChatGPT, Gemini, and Claude, are increasingly being used in medical education. One emerging application is the generation of multiple-choice questions (MCQs). This perspective offers a practical approach for medical educators to use LLMs in assessment design. It describes how LLMs can assist in drafting questions, suggesting distractors, and providing language variation. It also explains where human judgment is essential, such as ensuring content accuracy, curriculum alignment, and proper validation. The article highlights the need for structured prompts and offers strategies for responsible use of LLMs in MCQ development.
Mondal et al. (Thu,) studied this question.