How researchers are using AI to speed up drug discovery and development: Q&A

University of Washington researchers said a new institute is using artificial intelligence to address bottlenecks in drug development after the initial discovery of promising molecules. The Institute for Innovations in Drug Delivery and Disposition opened in July 2026 and is led by three faculty co-directors. Co-director Gaurav Bhardwaj said traditional drug discovery relies on trial and error with low success rates. He said AI methods have accelerated finding molecules that bind to target proteins, but the bottleneck has shifted to adding other drug-like properties such as stability and delivery. The institute aims to build models that create molecules successful through the full development pipeline. Co-director Marco Pravetoni said AI allows his lab to design thousands of antibody candidates instead of just 10, accumulating data to reduce risks before clinical trials. Co-director Nina Isoherranen said researchers are building digital twins to predict how drugs behave in individual patients, including pregnant women who are often excluded from trials. She said the goal is to predict drug disposition in humans before human testing begins.