How AI is shortening drug discovery timelines in China

Details

Insilico Medicine has compressed early-stage drug discovery to roughly one year by pairing generative AI with laboratory validation in China, according to CEO Alex Zhavoronkov. The company's fastest program reached candidate nomination in nine months; its typical timeline sits at about 13 months. Conventional drug discovery averages around four and a half years to reach the same milestone. The workflow uses generative AI to identify biological targets, design candidate molecules, and rank compounds for lab testing. Researchers then synthesize and validate between 60 and 200 compounds per program before nominating a preclinical candidate. Insilico has not published a controlled comparison against equivalent programs developed without AI, so the relative contribution of AI versus China's research environment is not independently verified. Insilico has generated 31 preclinical candidates since 2021 and received 13 investigational new drug clearances, allowing those programs to advance toward human studies. AI model development happens in Montreal and Abu Dhabi; biological testing and compound screening take place in Shanghai, where the company has automated parts of its laboratory workflow. Zhavoronkov attributed part of the speedup to China's research infrastructure and regulatory environment, estimating that pharma companies operating labs there can cut traditional timelines by about two years. The broader context: China is increasingly central to new drug development, not just generic manufacturing — a shift that several international pharma firms are leveraging through local partnerships and CRO relationships.