11. AI-Designed Antibiotics
Machine learning models trained on libraries of millions of molecules have begun predicting candidates with potent antibacterial properties. In 2020, an MIT team used a deep neural network to discover halicin, an antibiotic effective against drug-resistant pathogens (MIT News).
This AI-driven approach reduces reliance on slow trial-and-error screening and accelerates drug candidate selection. Startups such as Insitro and Atomwise combine predictive models with rapid lab validation to propose novel antibiotic scaffolds. As described in Cell, machine-led discovery can cut development times drastically, offering hope against rising bacterial resistance.