Researchers and industry teams reported advances in AI‑driven drug discovery, including open‑source models such as Boltz‑2 and physics‑informed large quantitative models (LQMs) that significantly speed binding‑affinity prediction. These developments promise to reduce reliance on resource‑intensive experimental screens and accelerate lead prioritization. Boltz‑2 demonstrated high accuracy and rapid prediction times, while LQMs couple physics with data to model molecular interactions. Sponsors and platform companies say these tools can cut early discovery timelines and lower screening costs, though real‑world validation and integration into existing pipelines remain ongoing tasks. Experts caution that AI models complement, not replace, experimental validation; they expect hybrid workflows combining computational predictions with targeted biophysical assays to be the near‑term path forward.