In a First, AI Designs Complete Genomes of Viable Bacteria-Infecting Viruses
Researchers in the United States say they have used artificial intelligence to design entirely new viruses that are fully functional and able to replicate in laboratory conditions. The 16 novel viruses, which infect bacteria and pose no known threat to humans, represent the first time a whole genome has been successfully designed by AI.
The work, led by Brian Hie, assistant professor at Stanford University, was described as a 'very significant turning point' in synthetic biology. It could open new avenues for treating diseases, particularly through engineered phages—viruses that target harmful bacteria—and other therapeutic applications. But independent experts have also cautioned that such AI-designed biological agents raise urgent safety and security questions that must be addressed before the technology advances further.
AI tools have already been used to design new antibiotics, a comparatively simpler task than designing a viable virus from scratch. The new breakthrough involves generative AI producing a complete viral genome, a molecule that contains all genetic instructions needed for the virus to reproduce inside a host cell.
'This is a next step in the complexity that's designable by generative AI. This is the first time generative AI has been used to design a complete genome, something that can replicate and have other functions inside cells. This was new territory for us,' Hie told the BBC.
The researchers designed the genomes using a model trained on known viral sequences. They then synthesised the resulting DNA and inserted it into bacteria to see if the AI-generated instructions would produce working viruses. The viruses successfully multiplied, confirming that the AI-generated genomes were biologically functional.
While the viruses in this experiment are harmless to humans, experts point to the growing capability of AI to design biological materials with high precision. That capability, they warn, could eventually be misused to create pathogens with pandemic potential. The risk is not immediate, but the pace of progress suggests governance frameworks need to evolve alongside the science.
Current biosecurity regulations focus on physical access to dangerous pathogens and DNA synthesis screening, but AI-generated designs can be shared digitally, potentially bypassing traditional oversight. Some scientists have called for stricter international rules requiring DNA synthesis companies to screen orders for sequences that match known pathogens or AI-generated threat signatures.
At the same time, proponents of the research emphasise its constructive potential. Phages engineered by AI could be used in agriculture to combat crop diseases, in medicine to treat antibiotic-resistant infections, and in environmental cleanup. The ability to design entire genomes could one day lead to custom-made organisms for producing fuels, materials, or vaccines.
The study has not yet been peer-reviewed and was published on a preprint server, but it has already sparked discussion among bioethicists and synthetic biologists. Hie and his team have stated that they are committed to responsible research and are in contact with safety review boards.
As the world watches AI's rapid expansion into biology, this experiment serves as both a demonstration of scientific creativity and a reminder that the same tools can be used for harm. The challenge for policymakers, researchers, and the public is to find a balance that encourages innovation while guarding against misuse.
