Artificial intelligence generates innovative viruses with the potential to combat resistant bacteria
For the first time in history, researchers used artificial intelligence (AI) to develop a synthetic virus that did not exist in nature. The microorganism, detailed in an article in Science magazine, has the ability to replicate itself in the laboratory. According to the study authors, this achievement could transform the treatment of infections that are resistant to antibiotics, but the publication also opened an important debate about biosafety.
Viruses created under the supervision of researchers at Stanford University and the Arc Institute in the United States do not infect humans. They are known as bacteriophages, organisms that exclusively attack bacteria. By employing AI models trained on millions of genomes, scientists have engineered hundreds of never-before-seen versions of these microbes.
After the synthesis process in the laboratory, 16 of the viruses proved to be functional and efficient in destroying strains of *Escherichia coli*, a bacterium that can cause serious urinary and intestinal infections. A cocktail made with these artificial viruses has been shown to be capable of overcoming resistant bacteria, presenting superior performance compared to natural microorganisms with similar characteristics.
More on this story: Scientists reverse superbug resistance to vancomycin in groundbreaking discovery
Technological advancement in the creation of genomes by artificial intelligence
The achieved result represents a significant milestone for the field of synthetic biology, according to the authors. It proves that algorithms are capable of designing complete genomes, and not just isolated genes. Until now, this ability has been beyond the reach of artificial intelligence due to the complexity of genetic organization.
The enthusiasm, however, was accompanied by concern. In an article also published in Science, Tom Inglesby and Moritz Hanke, from the Center for Health Security at Johns Hopkins University, in the United States, issued a warning. “While this is promising for life science applications, it also raises urgent biosafety and biosecurity questions. The ability to assemble viral genomes using generative AI already exists; the governance to safely control it does not,” they wrote.
For biosafety experts, the research highlights a growing gap between the speed of technological advances and international control mechanisms. The fear does not focus on the study itself carried out with viruses that do not infect humans and under strict safety protocols, but on the possibility that similar techniques could, in the future, be applied to much more dangerous microorganisms. In the text written for Science, Inglesby and Hanke stated that the challenge is not just controlling artificial intelligence models, “but the entire ecosystem that allows digital sequences to be transformed into real organisms, including companies that synthesize DNA and laboratories capable of producing these microorganisms.”
See Also
Attract prosperity: check out the lucky numbers for each sign to bet on
Researchers identify doctor in Pompeii through his surgical instruments
Luigi Mangione admits to having shot CEO and pleads guilty at federal level
Learn more: Artificial intelligence recreates the face of a victim in Pompeii who tried to escape Vesuvius
The research authors themselves acknowledged, in the article, that any future application will depend on rigorous security mechanisms. They clarified, however, that the study focused on bacteriophages precisely because they are small viruses, well known to the scientific community and incapable of causing disease in humans.
Promising impact in the fight against antibiotic resistance
Scientists from Stanford and the Arc Institute also highlight that the potential benefits of the technique are notable, especially in a period when many bacteria have developed resistance to antibiotics. According to the World Health Organization (WHO), the loss of effectiveness of this category of medicines is one of the biggest threats to global public health, intensifying the search for new treatment solutions.
As AI has enabled the creation of hundreds of bacteriophages, the authors claim to have at their disposal an arsenal capable of challenging even the most resistant strains. “If the bacteria acquires resistance to one of the artificial viruses, that’s the end for the medicine. But if you have several genetically distinct viruses in a mixture, it will be more difficult for the bacteria to develop resistance to the entire cocktail”, explained biochemical engineer Brian Hie, one of the study’s authors, in a note.
Although the experiment described in Science is only a proof of concept, that is, the demonstration that an idea is technically viable, Hie believes that similar approaches could be used to develop viruses targeting other harmful bacteria. Among them are *Staphylococcus aureus*, which causes tuberculosis and is resistant to the antibiotic methicillin, and *Pseudomonas aeruginosa*, one of the main causes of serious hospital infections.
