Artificial Intelligence in Gastrointestinal Precision Medicine: A Critical Appraisal Through the Lens of Inflammatory Bowel Disease

Artificial intelligence (AI) now supports many facets of gastrointestinal and hepatic care, from lesion recognition during endoscopy to the interpretation of large molecular datasets. This review first outlines AI's broader role across gastroenterology and hepatology, and then uses inflammatory bowel disease (IBD) as a case study for AI-assisted multiomics integration. The review applies a consistent evidentiary framework that distinguishes exploratory, retrospective, and prospective study designs. Retrospective genomic and image-based studies report promising discriminative performance for risk classification and lesion detection, but this evidence remains largely exploratory or retrospective, whereas prospective randomized evidence is currently strongest for AI-assisted endoscopic polyp detection. Reviewing the biochemistry linking genomics, transcriptomics, proteomics, and metabolomics to host-microbiota interactions clarifies where this supporting evidence is strong and where it is weak, and highlights the standardization, validation, and translational barriers that continue to separate model performance from routine clinical use.