Artificial Intelligence in Undergraduate Medical Education: A Cross-Sectional Study of Utilization Patterns and Perceptions Among Medical Students

Background Artificial intelligence (AI), particularly tools such as ChatGPT, has quickly become part of how medical students study and revise. In day-to-day academic settings, many students are already using AI to clarify concepts and prepare for exams. However, there are limited data from Indian medical institutions regarding how frequently these tools are used and how students perceive them. Objective To assess awareness, patterns of use, perceived usefulness, reliability, and concerns related to AI tools among undergraduate medical students at All India Institute of Medical Sciences (AIIMS) Rishikesh. Methods A cross-sectional questionnaire-based study was conducted among 297 undergraduate MBBS students at AIIMS, Rishikesh. The study was designed and carried out by faculty from the Department of Anatomy. A structured survey collected information on the frequency of AI use, preferred platforms, academic applications, trust in AI-generated information, and perceived risks. Data were analyzed using descriptive statistics, and chi-square tests were applied to assess associations between selected variables. Results Most students (91.6%) reported using AI tools for academic purposes, with ChatGPT (OpenAI, San Francisco, California, US) being the most commonly used platform (96%). The majority used AI to better understand difficult topics (88.2%). Although 70.7% considered AI outputs to be good or reliable, concerns were common, particularly regarding accuracy (72.1%) and data privacy (50.2%). Students who used AI more frequently were significantly more likely to support formal integration of AI into the medical curriculum (χ² = 16.82, p = 0.001). Overall, 69.4% favored structured incorporation of AI training. Conclusion AI tools are already widely used by undergraduate medical students and are largely viewed as helpful supplementary learning resources. At the same time, students remain cautious about accuracy and ethical implications. These findings suggest that rather than ignoring AI use, medical institutions should consider structured guidance and training to ensure responsible and effective integration.