Become A Member

Background: Artificial intelligence (AI) and machine learning (ML) are transforming pharmaceutical research. In traditional medicine, most applications so far concern diagnosis and knowledge digitisation, while applications to the manufacture and quality of medicines remain limited, particularly for Ayurveda. Objective: To review current applications of AI and ML relevant to Ayurvedic pharmaceutics, drawing on experience in traditional Chinese medicine (TCM) and pharmaceutical sciences, and to propose priority use cases and requirements. Methods: A narrative review of reviews and primary studies on AI in traditional medicine, AI-based quality evaluation of herbal materials, computational toxicology, formulation modelling and Indian phytochemical databases. Results: A 2025 review of AI in traditional medicine found only two Ayurveda-specific AI studies, both outside pharmaceutics. In TCM, AI-based spectroscopic and imaging methods have reached high accuracy in quality tasks, for example raising adulterant identification from 77.1% to 97.6% with FT-NIR and machine learning, and achieving 99.35% accuracy in species identification with convolutional neural networks. ML models have flagged potentially hepatotoxic constituents in medicinal plants. Artificial neural networks have long been used for formulation optimisation in conventional pharmaceutics. India has curated resources, such as a database of 1,742 medicinal plants, 9,596 phytochemicals and 974 traditional formulations. Barriers include small and heterogeneous datasets, inconsistent terminology, limited external validation and unclear regulation. Conclusion: AI could strengthen raw material authentication, fingerprint interpretation, process control and safety screening in Ayurvedic pharmaceutics, but only with standardised, well-annotated datasets and transparent validation.