Artificial Intelligence Reshapes Drug Discovery While Facing Downstream Hurdles

China News Service reports that artificial intelligence pharmaceuticals are becoming an active hotspot across the current medical research and development field. Academician Chen Kaixian of the Chinese Academy of Sciences and Shanghai Institute of Materia Medica stated that artificial intelligence intervention primarily improves efficiency at the front end of new drug development, assisting tasks such as literature retrieval, data analysis, and molecule screening. At Westlake University, the development team used artificial intelligence technology during early research to perform virtual screening across a large molecular pool, filtering out false positives to select potential candidate compounds before using models to enhance safety and toxicity predictions. Sheng Ding, founding dean of Tsinghua University School of Pharmaceutical Sciences, observed that while artificial intelligence accelerates preclinical molecule generation, failure rates in later stages have not fallen, potentially generating more hypotheses requiring validation and greater resource expenditure. Shanghai Guotou Innovation Center general manager Zhang Xuan noted that stronger front-end generative ability elevates requirements for subsequent screening and validation, emphasizing that artificial intelligence cannot automatically rectify missing, biased, or noisy data.