Abstract
It takes ten to fifteen years for a compound to progress from its identification to regulatory approval as a drug. Drug discovery is complex and resource-intensive process in which more than 90% of compounds never make it from bench to bedside and eventually get rejected during the development process. Experimental drugs failures often occur due to poor target selection, inadequate preclinical models, unforeseen toxicity, lack of efficacy in human trials, and the complexity of disease mechanisms, which make it difficult to predict drug responses accurately. Additionally, drug discovery is slowed down by a lack of collaboration between academia and industry, limiting the timely exchange of knowledge and expertise. Artificial intelligence (AI) is becoming an important tool in drug discovery, offering new possibilities to overcome existing challenges. It can help researchers identify better drug targets, make the screening process more efficient, and optimize drug design, which could speed up development and improve success rates. However, use of AI is associated with certain drawbacks such as potential exacerbation of healthcare gaps, protection of sensitive patient data and a need for informed consent. This review aims to discuss key challenges that hinder drug development process and explore future directions to enhance the efficiency of drug discovery.