Is AI Capable of Real-World Drug Discovery? Lessons From My Career

AI has transformed many areas of science, but its impact on real-world small molecule drug discovery remains nuanced. Many discovery programs operate in low-data regimes where small structural changes have large effects—conditions that challenge current AI methods.

Using case studies from orexin antagonists and PDE10 inhibitors, this talk explores how close integration of computational and medicinal chemistry, combined with deep understanding of molecular shape and protein–ligand interactions, enabled targeted molecular design and outsized gains in efficacy and selectivity. The talk highlights both the limitations of today’s AI approaches and where AI can meaningfully complement human expertise.

👉 Join NYAGIM for our 2026 Q1 event to hear a grounded, experience-based perspective on the real role of AI in drug discovery and connect with the local computational chemistry and informatics community.