citybiz+ EvolutionaryScale Raises $142 Million to Advance GenAI Models for Biology

New York-based EvolutionaryScale has emerged from stealth, releasing a generative artificial intelligence model for the life sciences on Wednesday, and announcing $142 million in funding from a clutch of five investors, including chipmaker Nvidia’s NVentures and Amazon.

The funding round was led by former GitHub CEO Nat Friedman, Safe Superintelligence co-founder and former Apple executive Daniel Gross, and Lux Capital.

‘ChatGPT Moment for Biology’

Lux Capital cofounder and managing partner Josh Wolfe termed EvolutionaryScale’s launch of a generative AI model called ESM3 a “ChatGPT moment for biology,” Reuters reported. ESM3 can be used to create novel proteins and other biological systems.

EvolutionaryScale’s co-founder and Chief Scientist Alex Rives told Reuters he and his colleagues Tom Sercue and Salvatore Candido started developing ESM3 when working at Meta. The trio co-founded EvolutionaryScale last summer, after Meta shut down the unit they worked in.

Sercu was previously a research engineer at IBM’s T. J. Watson Research Center, and Candido was a distinguished software engineer at Google. Rives has co-founded multiple drug companies including Syros and Fate Therapeutics.

‘Code of Life’

“We’re excited to launch EvolutionaryScale today with ESM3 — a frontier generative language model for the life sciences that advances our ability to program the code of life,” the company said on LinkedIn. In the future, ESM3 and more advanced AI tools would “engineer biology from first principles in the same way we engineer structures, machines and microchips, and write computer programs,” the company said.

EvolutionaryScale said it already coaxed ESM to step “across 500 million years of evolution to create a new fluorescent protein.” It did not say how the new protein could be used.

citybiz+ Sponsors

On its website, EvolutionaryScale asserts that biology is fundamentally programmable. “Every living organism shares the same genetic code across the same 20 amino acids — life’s alphabet. ESM3 understands all of this biological data, translates it, and speaks it fluently to be used as a generative tool,” it added.

citybiz+ Cohorts

3 Versions of ESM3
The company is releasing three versions of ESM3. One is being open sourced for non-commercial research, while AWS and Nvidia are to build and commercialize the other two models. The models would have wide range of applications, from accelerating drug discovery to engineering microbes capable of breaking down plastic in the environment, Rives told Reuters.

EvolutionaryScale plans to use the funding to train its next generation of AI models as well as build out a team to partner with the biotech industry.