AI-created viruses could help beat antibiotic resistance, study finds
Artificial intelligence has created 16 entirely new viruses in a breakthrough researchers say could one day help tackle antibiotic-resistant infections.
It’s been hailed as a hopeful step forward in medicine, but experts say it also raises concerns around the potential for misuse.
Artificial intelligence has created 16 entirely new viruses in a breakthrough researchers say could one day help tackle antibiotic-resistant infections. BSIP/Universal Images Group via
Scientists from Stanford University and the Arc Institute trained an AI model called Evo – similar to ChatGPT – using genetic information from millions of living organisms.
They then asked it to design thousands of new viral genomes capable of infecting E. coli bacteria.
Researchers built and tested about 300 of those designs in the laboratory, with 16 proving viable as viruses.
The new viruses are bacteriophages – which target bacteria – and researchers hope they could eventually be used to treat bacterial infections that no longer respond to antibiotics.
“The breakthrough achieved is significant,” Jordi García Ojalvo, professor at Pompeu Fabra University in Barcelona, said.
However, the study - published on Thursday in the journal Science - has also stirred debate around biosecurity and how generative AI should be regulated.
In a commentary published alongside the study, experts from the Johns Hopkins Centre for Health Security warned that AI capabilities are outpacing governance.
Of the thousands of genomes that the AI generated, scientists built and tested about 300 in the lab and found that 16 were viable viruses. AP
“Although this is promising for life sciences applications, it also raises urgent biosafety and biosecurity questions,” they said.
“The ability to compose viral genomes using generative AI now exists; the governance to safely steer it does not.”
While the Stanford and Arc Institute teams stressed their research was strictly limited to harmless bacteriophages, the Johns Hopkins scholars warned against extending the technology to human, animal, or plant pathogens.
“Such genomes might encode new pathogens that can infect humans, animals, or plants in ways that cannot be contained by existing countermeasures,” they said.
Ojalvo said the immediate risk remains relatively low because AI-generated virus designs still need to be created and tested in a lab before scientists know if they will work.
“It is difficult to imagine these models automatically generating viable genomes ‘out-of-the-box,’” he said.
Reported with CNN
Share a tip-off, video or photo with us
