Last Updated: 15 June 2026
Radical Numerics, an AI research lab, has launched with $50 million in seed funding and a specific technical goal: build a single AI system capable of reading, writing, and reasoning across the full range of biology, covering DNA, RNA, proteins, and more. The company was founded by the team behind Evo and Evo 2, the models credited with creating the field of generative genomics. Eric Nguyen, who holds a PhD in Bioengineering and AI from Stanford, serves as CEO.
The market they are entering is growing fast. According to Grand View Research, the global AI in genomics market was valued at $1.26 billion in 2025 and is forecast to reach $18.82 billion by 2033, at a compound annual growth rate of 40.3% between 2026 and 2033. Demand is coming from drug discovery, cancer diagnostics, and the broader push to analyze biological data at scale. Most existing AI tools in biology are built for one task at a time. They cannot reason across different types of biological information in a single system, and that constraint is the opening Radical Numerics is trying to fill.
The $50 million seed round was led by Emergence Capital, a San Francisco venture firm with early investments in Salesforce, Veeva Systems, and Zoom. Other participants include Obvious Ventures, Triatomic Capital, Factory, and First Spark Ventures. Patrick Collison, co-founder and CEO of Stripe, was among the pre-seed backers.
Standard AI models in biology are trained to do one thing well. A model built to predict protein shapes cannot also analyze gene regulation or scan for disease-causing DNA variants. Each task requires its own separate tool. Radical Numerics is training its models on multiple types of biological data at the same time, across what researchers call molecular layers: DNA, RNA, proteins, and beyond. The company calls this a multimodal approach. In plain terms, one model handles many types of biological information at once, rather than handing off between specialized tools.
That capability also brings serious risks. Major AI labs have warned that commercial AI models could soon lower the barriers to planning biological attacks, and that more specialized biological design tools could help bad actors produce harmful agents. Scientists and policymakers are actively debating how to regulate access to this kind of software. Radical Numerics says it holds a dual mandate: advance biological design for human health, and build the detection systems needed to protect against its misuse. It is partnering with a US national laboratory on pathogen detection and biosurveillance, using the same models that power its medical work.
"Most labs bolt safety on at the end. Radical Numerics built it into the foundation," said Gordon Ritter, Founder and General Partner at Emergence Capital. "They've paired frontier-model capability with real biosecurity expertise to open a scientific field that didn't exist before. That combination is rare, and it's why we led this round."
Gordon Ritter, Founder and General Partner at Emergence Capital
The funding will go toward three priorities: scaling the next generation of models, growing the research team, and building the computing infrastructure required to train at this level. A data center is already under construction, equipped with NVIDIA Blackwell chips. Hiring is underway for AI researchers, systems engineers, and computational biologists from institutions including Stanford, MIT, and Google DeepMind.
Alongside the funding announcement, the company is previewing Omnii, its next-generation genomic language model. Early tests show it can identify functional genetic variants linked to Alzheimer's disease without having been specifically trained on that condition. This is called zero-shot transfer: the model applies what it learned during training to a domain it was never directly shown. The same model is also being used to detect AI-generated or AI-modified pathogens, an early application of the company's biosecurity work.
On the medical side, Radical Numerics is working with a cancer diagnostics company to apply Omnii to pancreatic and multi-cancer detection. The approach combines multiple molecular signals into one diagnostic, targeting cases that existing single-signal tools may miss.
The four co-founders all came from Liquid AI and hold advanced degrees from Stanford. Eric Nguyen is CEO, with a PhD in Bioengineering and AI. Michael Poli, Chief AI Scientist, also holds a Stanford PhD. Stefano Massaroli, President, did a postdoctoral fellowship with Yoshua Bengio, one of the most influential researchers in modern deep learning. Armin Thomas, CTO, completed a postdoc at Stanford under Chris Ré.
Before starting Radical Numerics, the team built Evo and Evo 2, the largest openly available AI models trained on biological DNA. Both models were featured on the covers of Science and Nature respectively, and Eric Nguyen presented the work at TED2025. External scientists using the open-source models later produced the first complete genome designed by AI, a bacteriophage, which is a virus that infects bacteria and poses no risk to humans.
"Evo showed that AI can generate DNA and whole genomes, the next generation of models will go further with the ability to control function, and eventually, create entirely new forms of life," said Eric Nguyen, CEO of Radical Numerics. "Our multimodal models are already far more capable, and we understand the responsibility that comes with that. The same models that can help cure disease may also lower the barrier to designing harmful biology. These forces are inseparable. Biology will be the most consequential application of AI."
Eric Nguyen, CEO of Radical Numerics
Scientific advisors include Eric Horvitz, Chief Scientific Officer at Microsoft; Chris Ré at Stanford; George Church at Harvard; and Andrew Weber, former US Assistant Secretary of Defense for Nuclear, Chemical, and Biological Defense Programs.
Emergence Capital, which led the round, is a San Francisco-based firm founded in 2003 with a focus on early-stage enterprise technology. The rest of the round included Obvious Ventures, Triatomic Capital, Factory, and First Spark Ventures, with Patrick Collison participating at the pre-seed stage.
