AI Creates Novel Viruses! Hope for Medicine, Fears for Safety?
AI Creates Novel Viruses! Hope for Medicine, Fears for Safety?
Stanford researchers used AI to generate never-before-seen viruses, offering new hope for battling superbugs but sparking urgent global biosecurity alarms. Dive into the groundbreaking, yet concerning, discovery!
In a groundbreaking, and somewhat unsettling, development, researchers have successfully utilized artificial intelligence to identify and create functioning viruses that have never been observed in nature. This remarkable achievement, published in the journal Science, has opened up new frontiers in synthetic genomics, as stated by its authors from Stanford University and the California-based Arc Institute.
Dr. Brian Hie, a chemical engineer at Stanford, explained that his team's work holds significant promise for developing future medicines. Specifically, it could aid in treating drug-resistant bacteria by leveraging bacteriophages, viruses that specifically infect and kill bacteria. "If the bacteria gain resistance to a single phage, it’s game over for the medication," Hie noted. "But if you have multiple genetically distinct phages in a mixture, it would be harder for the bacteria to develop resistance to the entire cocktail."
However, this scientific leap isn't without its serious concerns. Experts from the Center for Health Security at Johns Hopkins University, in a commentary accompanying the research, voiced alarm over the rapid advancement of such technologies. They highlighted that governments are currently ill-equipped to manage these powerful tools, which could potentially fall into the hands of malicious actors. "Although this is promising for life sciences applications, it also raises urgent biosafety and biosecurity questions," they wrote, emphasizing, "The ability to compose viral genomes using generative AI now exists; the governance to safely steer it does not."
To achieve this feat, the research team trained an open-source AI model called Evo on a vast dataset of genetic sequences from millions of animals, plants, microbes, and existing viruses, carefully excluding data from viruses known to infect humans. They then focused the AI on the virus ΦX174, a well-studied bacteriophage known to infect only E. coli bacteria, feeding it genomic data from this virus and similar ones. The AI model subsequently generated an astonishing 700,000 potential new viruses. From this pool, researchers converted 285 into DNA molecules, with 16 of them proving to be novel, functional viruses.
