ChatGPT in medical education: usage, perceptions, and the role of AI literacy among healthcare profession students

Abstract Background and Aims ChatGPT is a large language model using generative artificial intelligence (AI) technology to generate human-like, context-aware responses. While the role of AI in education has been well studied, ChatGPT remains a fresh frontier, especially in the UAE, where its usage among medical students is unexamined. This study aims to assess the prevalence of ChatGPT use among medical and health profession students, their perspectives on its benefits and risks, and its correlation with AI literacy and academic stress. Methods This cross-sectional study enrolled 304 students (78.9% females), representing the colleges of health sciences, medicine, pharmacy, and dentistry. Data were collected via a self-administered questionnaire incorporating validated tools. TAME-ChatGPT measured perceptions of benefits and risks, whereas the Meta AI Literacy Score assessed AI literacy. Data were coded and analyzed via SPSS 22. Descriptive statistics included frequency, relative frequency, median, and interquartile range. Bivariate analyses included correlation and nonparametric tests. The significance level was set at 5%. Results Among 304 participants, 67.4% reported using ChatGPT in the past 3 months. Nonuse was mainly due to lack of usefulness (49.4%) or lack of knowledge (40%). ChatGPT use was significantly associated with positive attitudes toward curriculum integration ( p < 0.001), higher GPA ( p = 0.003), and college degree ( p = 0.006). Importantly, students who used ChatGPT for practice tests, assignment feedback, and written work reported lower academic stress ( p < 0.01). However, most students used ChatGPT primarily as an information source (64.1%). AI literacy scores were moderate (6.06/10, IQR 2.77–7.89) and strongly linked to ChatGPT use ( p < 0.001). Higher AI literacy correlated with reduced anxiety, greater perceived usefulness, and more positive behavioral attitudes toward ChatGPT. Conclusion Widespread informal adoption of ChatGPT among health profession students highlights its potential as a supportive learning tool. However, the most impactful uses, such as feedback and practice testing, remain underutilized. Low AI literacy exposes an equity gap: students with limited skills risk being left behind in AI-driven education. Integrating structured AI literacy programs into curricula can empower all learners to use ChatGPT more effectively and safely, ensuring equitable access to its benefits while mitigating risks and facilitating smoother integration of AI into medical education.