AI used to create antibacterial viruses but experts warn of biosafety risk

AI used to create antibacterial viruses but experts warn of biosafety risk Scientists hope work could lead to new treatments for infections resistant to antibiotics - Bookmark - CommentsGo to comments Scientists have created the first fully functional viruses designed by artificial intelligence – a breakthrough they believe could pave the way for new treatments for infections that no longer respond to antibiotics. However, some experts have raised questions about biosafety and regulation of that type of work. The viruses, known as bacteriophages or phages, infect only bacteria and pose no threat to humans. In laboratory tests, a mixture of the AI-designed phages killed E.coli bacteria that had become resistant to naturally occurring phages. Some types of E.coli infection cause severe food poisoning, stomach cramps, bloody diarrhoea and urinary tract infections. Researchers at Stanford University in the US created 16 novel viruses capable of infecting bacteria – the first time AI has been used to design entire viral genomes that can function and replicate in the laboratory. Experts hailed the work a milestone that could lead to a new era of disease treatment. Researchers led by Brian Hie, a chemical engineer and assistant professor at Stanford, used genome language models, genetic equivalents of the large language models that power AI chatbots, to design the bacteriophage genomes. The viruses are already used in some parts of the world to treat persistent bacterial infections. According to the study, published in the journal Science, the ability to tailor viral genomes to target specific bacteria could also broaden the range of tools available to researchers. “This is a next step in the complexity that's designable by generative AI,” Dr Hie told the BBC. He says he believes the technology could “massively improve human health” through the development of new therapies and drugs. But some experts warned the technology could be misused. In an accompanying article, Prof Tom Inglesby and Dr Moritz Hanke, of the Johns Hopkins Center for Health Security, said the work raised “urgent biosafety and biosecurity questions”. “The ability to compose viral genomes using generative AI now exists; the governance to safely steer it does not,” they wrote. “Such genomes might encode new pathogens that … cannot be contained by existing countermeasures,” they wrote. Join our commenting forum Join thought-provoking conversations, follow other Independent readers and see their replies Comments