This article is the fourth in a six-part series exploring AI's impact on medical research and treatments. It focuses on how AI is revolutionizing drug discovery, featuring the efforts of companies like Insilico Medicine, founded by Alex Zhavoronkov. Insilico has developed a new drug for idiopathic pulmonary fibrosis (IPF) using AI, which has shown promising results in small trials. Though the drug isn't approved yet, it's part of a new wave of AI-driven drug development, potentially reducing the time and cost of bringing new drugs to market. The article highlights that AI is increasingly employed in identifying therapeutic targets and designing drugs, traditionally the domain of medicinal chemists. Companies like Alphabet's Isomorphic Labs are also making strides in AI-driven drug discovery. AI can mine large databases to make connections between molecular biology and disease, potentially speeding up the drug discovery process. While BCG's analysis indicates that at least 75 AI-discovered molecules are in clinical trials, there's still significant human involvement in the process.

AI is mainly used in two areas: identifying therapeutic targets and designing molecules using generative AI. Insilico, with over $425 million in funding, has used AI to predict clinical trial success and has several molecules in trials, including one for IPF. The firm's AI-designed molecule to treat IPF targets a protein called TNIK, illustrating a faster and leaner drug discovery approach. Despite AI's potential, challenges remain, particularly regarding the limited availability of data to train AI systems, which might introduce biases. Recursion Pharmaceuticals addresses this by generating extensive data through automated experiments and using AI to uncover unexpected relationships. The company has a supercomputer for this purpose, although the real test is whether these AI-discovered drugs can successfully complete clinical trials and prove more successful than traditional methods. Recursion's molecule for treating cancer is now in early trials. Ultimately, the article suggests that if AI can reliably improve drug discovery outcomes, it will be seen as the future of the pharmaceutical industry.