Abstract
Artificial intelligence (AI) is changing the pharmaceutical industry at every stage of drug development. This review examines how AI is currently used and the recent progress in areas such as manufacturing, drug discovery, formulation, clinical trials, and regulatory work. It covers topics such as predictive maintenance enabled by the Internet of Things (IoT) and machine learning, AI-powered virtual screening and drug design, smart drug delivery systems enhanced by deep learning, and quality-by-design strategies using process analytical technology and digital twins. The review also discusses how AI helps predict drug interactions, optimize clinical trials, and address the ethical, regulatory, and organizational challenges that come with these changes. The available evidence suggests that AI has the potential to reduce costs and accelerate certain phases of pharmaceutical development, particularly in the early stages of discovery. However, these benefits are markedly uneven across the lifecycle. While studies document significant reductions in computational time and costs in early discovery, the evidence is limited and prospective validation is scarce in preclinical phases. Clinical trials still provide no robust evidence that computational acceleration translates into reduced attrition or faster regulatory progress. Important challenges remain, including poor data quality, a lack of transparency in AI algorithms, inconsistent regulations, and insufficiently trained staff. The review suggests that to fully benefit from AI, the industry needs stronger regulations, high-quality, interoperable data systems, and teams with a mix of skills.