AI-driven Drug Repurposing: A Comprehensive Review and FutureProspects

Abstract: Advanced artificial intelligence (AI) technology has significantly advanced pharmaceutical research, particularly drug repurposing, by providing cost-effective and efficient alternatives to traditional drug discovery pipelines in addition to cutting costs and time. The review explores AI-powered approaches for discovering new therapeutic uses of existing drugs, focusing on machine learning, deep learning, and network-based methods. It highlights the role of drugtarget interaction predictions, molecular docking, and precision medicine in enhancing AI-driven drug repurposing. Despite its benefits, AI-driven drug repositioning faces challenges such as information quality, model stability, and practicability in the clinical setting. There are also ethical and regulatory issues that make its real-world adaptation to healthcare difficult. This review examines existing computational models and explains their advantages and shortcomings, with a clear indication of the need for explanation, validation, and application of the AI methods in clinical practices. Furthermore, drug repurposing has a promising future regarding the crossdisciplinary collaboration of searching researchers, clinicians, and regulatory agencies, and the implementation of multimodal data integration and explainable AI. This review concludes that the possibility of repurposing made possible by AI has the potential to transform how new therapeutic applications are made available to patients.