Artificial Intelligence-Driven Drug Discovery: Transforming the Pharmaceutical Pipeline from Target Identification to Clinical Translation

Abstract

The pharmaceutical industry faces profound challenges in its quest to develop novel, safe, and efficacious therapeutics. Traditional drug discovery pipelines are characterised by astronomical costs, protracted timelines, and high attrition rates, with the average cost of bringing a new drug to market exceeding two billion US dollars. Artificial intelligence (AI), encompassing machine learning, deep learning, and natural language processing, has emerged as a transformative force capable of reshaping every stage of the drug discovery and development continuum. This narrative review critically examines the integration of AI technologies across the pharmaceutical pipeline, from target identification and structure-based drug design to de novo molecular generation, absorption–distribution–metabolism–excretion–toxicity (ADMET) property prediction, drug repurposing, and clinical trial optimisation. Landmark advances—including the revolutionary AlphaFold protein structure prediction system, deep learning-enabled antibiotic discovery, generative molecular design platforms, and AI-assisted synthesis planning—are discussed in depth. The review further explores explainability challenges, regulatory implications, and ethical considerations surrounding the deployment of AI in pharmaceutical research. Despite substantial progress, significant hurdles remain, including data quality and availability, model interpretability, and the validation gap between computational predictions and experimental outcomes. This review synthesises current knowledge to provide a comprehensive assessment of the state of the art, highlights critical limitations, and outlines promising future directions for AI-driven drug discovery.