Nvidia announced a three‑year collaboration with Sheba Medical Center’s ARC Innovation and Mount Sinai to develop a Genomic Foundation Model (gFM) that links genetic variation — especially noncoding regions — to disease mechanisms and therapeutic response. The partnership will combine institutional genomic datasets, clinical expertise, and Nvidia’s AI stack to accelerate interpretation of regulatory sequences and to surface potential drug targets. Teams intend to use large language model techniques adapted for genomics to explain regulatory mechanisms and to power downstream discovery workflows. The collaboration taps Mount Sinai’s Million Health Discoveries cohort and Sheba’s clinical-genomic assets. If successful, the gFM could shorten target validation timelines by generating mechanistic hypotheses for noncoding variants, but its impact will depend on access to high-quality labeled data and robust clinical validation strategies.