Scientists have successfully created the first viruses designed by artificial intelligence. These viruses, known as bacteriophages, specifically infect bacteria and are used globally to treat persistent infections.
In laboratory tests, a combination of these AI-designed viruses effectively killed E. coli bacteria that were resistant to natural bacteriophages. Dr Brian Hie, a chemical engineer at Stanford University, used genome language models to design functioning bacteriophage genomes, which were then created in the lab and tested against E. coli.
While this development could transform phage therapy and expand biotechnological tools, the scientists involved have also raised concerns regarding biosafety, biocontainment, and biosecurity. Prof Tom Inglesby and Dr Moritz Hanke from Johns Hopkins University noted that while the ability to compose viral genomes using generative AI now exists, the necessary governance to steer it safely does not.
The AI models, Evo1 and Evo2, were trained using genetic data from 2 million bacteriophages, deliberately excluding genetic code for viruses that infect plants, humans, or other animals to mitigate risks. Although the process of creating viable bacteriophages was not highly efficient, the successful cocktail demonstrated its ability to overcome resistance in two different E. coli strains.