AI for TCM: From target identification to drug design

Traditional Chinese medicine (TCM) achieves therapeutic effects through multi-component and multi-target regulation, but its complexity makes mechanistic interpretation and rational prescription optimization difficult. The rapid growth of chemical structure data, multi-omics profiles, and large-scale clinical records has facilitated data-driven TCM research, although it also presents significant analytical challenges. Artificial intelligence (AI) offers a solution to this data-to-knowledge gap. This review covers recent progress in AI-driven TCM research, from active ingredient and target discovery to mechanism elucidation and intelligent prescription recommendations. We highlight AI-assisted compound screening and structure-based target prediction, integration of multi-omics data and knowledge graphs (KGs) for network inference, and AI-based prediction of synergistic drug combinations and formula optimization. Finally, we discuss key challenges related to data quality, interpretability, robustness, and clinical translation, along with future directions in AI-driven TCM research.