Researchers Created Functional Phages Using Artificial Intelligence

Scientists generated synthetic viral genomes that successfully infected bacteria resistant to natural phages.

Updated on Sept. 18, 2026 in Life Sciences

A close-up view of complex, synthetic 3D viral structures illuminated by sharp, cool light against a black background. — Researchers from Stanford and the Broad Institute successfully used artificial intelligence to design synthetic bacteriophages capable of infecting antibiotic-resistant E. coli. AI Illustration. Upload story photo >

Live Poll

Do you believe using artificial intelligence to create new biological organisms is a positive development?

Researchers from Stanford University and the Broad Institute used an artificial intelligence model to design new bacteriophage genomes. The team successfully produced 16 functional phages that can infect E. coli strains previously resistant to natural viral counterparts.

Why it matters

This breakthrough demonstrates a new method for designing phage cocktails capable of overcoming bacterial resistance. It offers a promising path forward for developing advanced antimicrobial therapies and biotechnology tools.

Researchers tested 302 AI-generated genomes, chemically synthesizing 285 to confirm viability. The 16 functional phages retained 93% to 98% sequence identity to natural genomes, with some variants differing by over 5% from known relatives.

The players

Stanford University

This major research institution based in California facilitated the primary study on synthetic viral genomes.

Broad Institute

This biomedical and genomic research center collaborated on the development of the artificial intelligence model used in the study.

The details

Using the Evo AI model trained on E. coli-infecting phage sequences, scientists generated viral templates inspired by the natural phage ΦX174. These synthetic phages were computationally screened for size and gene organization before physical synthesis, allowing them to bypass defense mechanisms in resistant bacterial strains.

1915: Frederick Twort observed clear patches in bacterial growth.
1917: Félix d'Hérelle identified the agent as a bacteriophage.
Early 1970s: DNA sequencing technology emerged for phage genomes.
September 18, 2026: Researchers published their study on AI-generated phages.

Biotech
Sept. 23
Anthropic AI Discovered New Enzyme System
Life Sciences
Sept. 23
Google DeepMind Released AlphaGenome Atlas Database
Life Sciences
Sept. 20
Researchers Developed AI to Map Genetic Mutations

The Big Picture

This achievement signals a paradigm shift where AI models like the Evo AI genomic model move beyond simply reading genetic code to actively authoring it. This capability bridges the gap between theoretical synthetic biology and the creation of targeted, lab-grown solutions for complex medical challenges.

This development could eventually lead to more effective treatments for bacterial infections that are currently resistant to traditional antibiotics. Future applications may result in highly personalized phage therapies designed to target specific bacterial pathogens in clinical settings.

The takeaway

The success of these AI-designed phages proves that artificial intelligence can effectively navigate complex genetic variations to solve biological problems. This milestone suggests that future medical treatments may be custom-engineered to bypass the evolutionary defenses developed by harmful bacteria.