HARNESSING CHATGPT FOR ASSESSMENT IN MEDICAL EDUCATION

Abstract

Medical education is undergoing a major shift. Traditional methods that emphasized memorization and high-stakes final exams are being replaced by competency-based medical education (CBME), which focuses on the holistic development of knowledge, skills, attitudes, and values. At the heart of CBME is assessment—not merely as a tool for grading, but as a dynamic, ongoing process that encourages, guides, and evaluates meaningful learning. In this evolving framework, assessments are designed to be authentic, continuous, and aligned with real-world clinical competencies.

Despite progress in medical education, educators still encounter numerous hurdles in applying effective assessment strategies. Challenges include developing well-constructed questions, ensuring fairness and consistency, providing timely feedback, and customizing assessments to meet the needs of diverse learners. These tasks are often labor-intensive and require advanced teaching expertise, especially when dealing with large student cohorts and an expanding curriculum. Additionally, the post-pandemic era has heightened the demand for flexible, digital, and technology-enabled tools to support both educators and students.

Artificial Intelligence (AI), particularly language-based tools like ChatGPT by OpenAI, is increasingly being recognized as a useful aid in education(Ahmed, 2023; Masters et al., 2025). Trained on vast amounts of text, ChatGPT can understand prompts, generate human-like responses, and assist with various academic tasks. While it is widely used for tasks such as content development and brainstorming, its potential in the areas of assessment design, execution, and feedback delivery in medical education is only beginning to be explored.

From my own experience as a pharmacology educator, I have found ChatGPT to be an effective, creative, and surprisingly insightful tool. It has been helpful not only in generating complex, integrated essay questions but also in simulating viva voce scenarios, developing OSCE structures, and offering feedback on reflective writing. In this personal review, I intend to present a reflective and optimistic view of ChatGPT’s role in medical education assessments—highlighting its potential, sharing practical examples, discussing current limitations, and imagining its future applications.