Computer-aided drug discovery: historical foundations, practical AI tools, and emerging ethical considerations

Computer-aided drug discovery (CADD) has become an integral component of modern drug development, supporting hit identification, lead optimisation, and candidate refinement across both academia and industry. Over 4 decades of methodological progress have contributed to the discovery or optimisation of several approved therapeutics, establishing CADD as a crucial element of drug research. In parallel, recent advances in artificial intelligence (AI) and machine learning (ML) have introduced a new generation of practical tools that offer improved predictive performance, accessible software implementations, and increasing integration into everyday drug discovery and pharmacology workflows. This review provides a concise historical overview of CADD, with an updated account of approved drugs and clinical-stage candidates whose discovery or optimisation has involved computational methods and highlights a curated set of contemporary AI and ML tools that are readily usable by non-specialists. In addition, we compile and analyse two comprehensive, practice-oriented resources: a collection of cloud-based virtual-screening platforms, and an extensive suite of freely accessible ADMET prediction tools, offering medicinal chemists a consolidated guide to open-source computational workflows. Finally, we discuss emerging ethical considerations, including data bias, transparency, computational costs, and the environmental impact of large-scale models, outlining responsible paths for the continued adoption of AI in drug discovery.