Until recently, drug discovery has been limited by fundamental practical constraints such as the range of proteins that exist in nature or the availability of samples, time, and funding for experiments. Now, drug hunters are circumventing these limitations by using generative AI models to explore a vast chemical design space that extends far beyond the bounds of natural protein diversity, or to simulate complex biological experiments entirely in silico. This session will examine how leading drug developers are integrating AI-driven protein and gene-expression models with high-throughput experimental systems to navigate biological complexity. These approaches are increasingly demonstrating their ability to accelerate preclinical discovery timelines, uncover drug candidates that might be overlooked by traditional methods, improve the oral bioavailability of biologics, and design first-in-class therapies with no natural analogues. Drawing on the experience of biotech’s at the forefront of AI-enabled drug design and clinical development, speakers will chart how the convergence of computation and experimentation is paving the way for a new era of molecular innovation. Specific discussion topics will include strategies for designing biologically meaningful assays, building differentiated models through proprietary, high-quality training data, and optimizing feedback loops between computation and experimentation to create more predictive, productive discovery engines. Drawing on the experience of biotechs at the forefront of AI-enabled drug design and clinical development, this session will chart how the convergence of computation and experimentation is paving the way for a new era of molecular innovation.