AI-driven discovery of herbal drugs: a new era in pharmacological intelligence

Herbal medicines represent a rich source of bioactive compounds with significant therapeutic potential. However, the discovery of herbal drugs is challenged by complex phytochemical compositions, variability in biological activity, limited standardization, and labor-intensive experimental screening. Conventional drug discovery approaches are costly, time-consuming, and often inadequate for efficiently translating traditional medicinal knowledge into scientifically validated therapies.This review aims to examine the role of Artificial Intelligence and its associated technologies, including Machine Learning, Deep Learning, and Natural Language Processing, in revolutionizing herbal drug discovery by facilitating phytochemical screening, target identification, pharmacological prediction, and formulation optimization.A comprehensive review of recent scientific literature was conducted to evaluate AI-driven methodologies employed in herbal medicine research. Key applications, including phytochemical database mining, molecular docking, virtual screening, network pharmacology, multi-omics integration, literature mining, and pharmacokinetic and toxicity prediction, were critically analyzed.AI-based platforms have substantially accelerated herbal drug discovery by enabling rapid identification of promising phytoconstituents, prediction of drug–target interactions, optimization of herbal formulations, and assessment of pharmacological efficacy and safety. The integration of computational intelligence with phytochemical and omics datasets has improved predictive accuracy, reduced experimental costs, and shortened preclinical development timelines. Nevertheless, challenges such as heterogeneous datasets, limited model interpretability, inadequate validation, and evolving regulatory frameworks continue to restrict broader implementation.AI is transforming herbal drug discovery by integrating traditional medicinal knowledge with advanced computational approaches. Future developments in explainable AI, personalized herbal medicine, network pharmacology, real-time ethnobotanical data mining, and multi-omics integration are expected to accelerate the development of safe, effective, and evidence-based natural product therapeutics.