Mount Sinai’s Icahn School of Medicine has announced a landmark three-year partnership with NVIDIA and Sheba Medical Center’s ARC Innovation to apply advanced AI specifically large language models to unexplored regions of the human genome.

Despite the majority of genomics research focusing on just ~2% of the genome, the remaining 98% is now believed to hold critical regulatory and functional elements. The collaboration aims to decode this “dark matter” using generative AI and powerful computation.

Under the agreement, Sheba and Mount Sinai will provide rich, clinically annotated genomic datasets and phenotypic data, while NVIDIA brings its AI infrastructure including its high-performance computing and AI platforms to build a genomic foundation model (gFM).

This model is expected to identify regulatory mechanisms, genetic variations, and disease-linked patterns that have historically eluded conventional sequencing approaches.

At Mount Sinai, this effort is rooted in its Million Health Discoveries Program, a flagship initiative in its Charles Bronfman Institute for Personalized Medicine and the Windreich Department of AI and Human Health.

The program aims to translate AI-driven genomic insights into clinical impact enabling deeper risk stratification, earlier disease detection, and ultimately, more personalized therapies.

The leadership from both institutions underlines the high stakes: “AI has the power to unlock the secrets of the human genome,” said NVIDIA’s AI team.

Meanwhile, Mount Sinai’s Dean, Eric J. Nestler, called the initiative a “global collaboration” that brings together expertise in AI, genetics, and clinical medicine to forge new pathways to improve health outcomes.