The utility of conversational artificial intelligence with large language models for patient information on cardiac electrophysiology procedures

The advent of conversational artificial intelligence (AI) through large language models (LLMs) provides a new and evolving avenue for patient education. These models, trained on vast textual datasets, hold significant potential in digital health and education by generating multidisciplinary content, answering complex questions, and speeding up information delivery(1). Increasingly, patients will utilize these services to address their questions, akin to the use of web-based search engines(2). We aimed to evaluate the accuracy and interpretability of existing conversational AI models for patient education on various cardiac electrophysiology (EP) procedures.