JMIR Publications

ABSTRACT

Large Language models, such as ChatGPT, are artificial intelligence systems with advanced natural language processing capabilities that have recently been popularized amongst students and educators in the healthcare field due to their ability to provide real-time access and feedback to a vast amount of medical knowledge. This study is a pilot test for integrating ChatGPT into the clinical and teaching workflow of an inpatient academic medical service. ChatGPT assisted the team in addressing knowledge gaps, expanding differential diagnoses, challenging medical axioms, supporting acute care decision-making, and facilitating complex care management through improved specialty consultations. Overall, ChatGPT usage enhanced team learning through an engaged discussion of responses and the generation of further areas of exploration and inquiry. Although several ethical concerns regarding biases, misinformation, and equity in healthcare highlight the need for guidelines and regulations of LLMs, ChatGPT was well-received in the clinical learning environment, adding to the growing literature exploring its various uses as a promising addition to the future of medical education and clinical medicine.