Stanford Evo 2 AI model generates phages against E. coli
Stanford's Evo 2 generated 300 phage candidates in silico. 16 survived lab validation. One could change how we fight antibiotic-resistant bacteria.
Why it matters
Evo 2 demonstrates frontier generative AI moving beyond text/code into wet-lab biology—a model trained on genomic sequences can propose novel molecular structures that pass real-world validation. This is a capability inflection: practitioners in biotech and pharma now have a precedent for using frontier models in the design cycle.
The key facts
5 to knowStanford researchers generated ~300 phage candidates using Evo 2 AI model
16 phages showed strong E. coli-killing activity in laboratory testing
Work centers on bacteriophage ΦX174
Led by Brian Hie, assistant professor of chemical engineering
Published August 2026
Go to the source
AI Newsartificialintelligence-news.com
Publisher excerpt: Stanford researchers have synthesised nearly 300 phages from DNA sequences produced by the Evo 2 generative AI model. Laboratory testing narrowed the group to 16 phages that showed particularly strong E. coli-killing activity. The work centres on bacteriophage ΦX174, pronounced “FYE-ex-1-7-4”.…