Generative artificial intelligence has successfully created 16 functional viruses that attack bacteria. The breakthrough opens the door to new therapies, but also raises urgent biosecurity questions.
In a revolutionary scientific feat, researchers at Stanford University have used generative artificial intelligence to design entirely new, functional viruses that can replicate in the lab. This is the first time AI has successfully designed complete genomes, and the resulting 16 new viruses were created to infect bacteria and pose no threat to humans.
"This is the next step in the complexity of what is possible to design with generative AI, and this is the first time generative AI has been used to design a complete genome, something that can replicate and have other functions within cells... this was new territory for us," Brian Hie, an assistant professor at Stanford University, told the BBC.
The research, published in the journal Science on August 6, 2026, represents a key milestone in synthetic biology, but it has also sparked serious warnings from security experts.
The technology works similarly to large language models like ChatGPT, which predict sequences of text. In this case, AI models called Evo1 and Evo2 predict the language of life instead of words. The models were trained on the genetic codes of viruses, bacteria, plants, and humans, and then fine-tuned to produce a type of virus known as a bacteriophage, which infects only specific species of bacteria.
The Stanford researchers selected the 302 most promising AI designs and synthesized them in the lab. Of that number, 16 proved effective at killing the bacterium E. coli. The genetic code of these phages is about 5,400 base pairs long. For comparison, the smallest genome of living cells has about 500,000 base pairs, while the human genome contains three billion.
Samuel King, a doctoral student in the lab, described the moment they realized the phages were working. It was in the early morning hours while they were observing petri dishes with bacteria. "We started noticing these clear spots, and it was just incredibly exciting," King said.
When the results were shared with the broader team, "the room spontaneously burst into applause," Hie recalls. The development of new phages could lead to new ways of treating infections that have become resistant to antibiotics, one of the pressing public health problems of our time.
Hie argues that this technology has the potential to "massively improve human health" by developing new drugs and therapies. Professor Marc Güell from the synthetic biology lab at Pompeu Fabra University in Spain called the study "a very significant milestone" because for the "first time in history, we are beginning to design biology on the computer." He added that it "allows us to dream of exciting possibilities for solving humanity's greatest challenges."
However, the same breakthrough also opens a darker side. In a commentary published alongside the study, Dr. Thomas Inglesby and Dr. Moritz Hanke from the Johns Hopkins Center for Health Security wrote that the discovery raises "urgent biosecurity and biosafety questions." They warned that it is no longer a question of "whether generative design of viral genomes will exist," but whether it can be used without "enabling serious harm." They explicitly stated that new viruses with the potential to cause disease "should not be developed."
The researchers themselves have taken steps to increase safety. They excluded viruses that can infect complex organisms from the training database, conducted the research on phages rather than viruses that attack humans, and the entire process took place in a secure laboratory. Hie argues that existing safeguards go a long way toward "ensuring the technology is used for good."
Professor Patrick Cai, chair of synthetic genomics at the Manchester Institute of Biotechnology, called the study "an important milestone." "The significance extends far beyond phages, it suggests that genome language models are beginning to learn the design principles encoded by evolution, opening the door to AI-assisted genome writing," Cai said.
Although viruses are not alive and it would take another significant leap for AI to generate living organisms, Hie says that attempting to create some simple organisms "would probably not be impossible, but would require a lot of work" and that they are "definitely interested in working on it."
