Agentic artificial intelligence in inflammatory bowel disease: toward autonomous and adaptive care

Inflammatory bowel disease (IBD) is a chronic, heterogeneous condition requiring ongoing monitoring and iterative therapeutic adjustment. Current management relies on multiple clinical, biochemical, imaging, and molecular data sources that are collected intermittently and rarely integrated into a unified representation of disease activity, leading to reactive clinical decision-making. Although artificial intelligence (AI) has demonstrated promise in IBD, most applications remain task-specific and operate on data collected at isolated time points, failing to capture the longitudinal nature of disease. Agentic AI offers a shift toward continuous and adaptive care by integrating diverse data streams into an evolving representation of disease state. This enables early detection of change and supports proactive intervention through a closed-loop system linking monitoring, interpretation, and action. Despite its potential, challenges related to data quality, interpretability, and clinical integration must be addressed for implementation.