Nvidia unveils powerful AI system for genetic research

The new AI system, called Evo 2, can read and design genetic code across all forms of life.

Scientists have high hopes that such AI technology will dramatically accelerate research by spotting patterns in vast amounts of data that would normally take years to analyze by hand.

The system learned from nearly 9 trillion pieces of genetic information taken from over 128,000 different organisms, including bacteria, plants, and humans.

In early tests, it accurately identified 90 percent of potentially harmful mutations in BRCA1, a gene linked to breast cancer.

Scientists say this could help develop more precise treatments, including gene therapies that only target specific cells.

The model was built using 2,000 Nvidia H100 processors on Amazon's cloud infrastructure.

Developed with the Arc Institute and Stanford University, Evo 2 is now freely available to scientists worldwide through Nvidia's BioNeMo research platform.

"Designing new biology has traditionally been a laborious, unpredictable and artisanal process," said Brian Hie, assistant professor at Stanford University.

"With Evo 2, we make biological design of complex systems more accessible to researchers."

Beyond medicine, researchers believe the technology could help create crops that better withstand climate change and develop new ways to break down pollution.

The project brings together Nvidia's computing power with the Arc Institute, a nonprofit research center founded in 2021 with $650 million in funding.

The institute works closely with Stanford, UC Berkeley, and UC San Francisco to tackle long-term scientific challenges.