Scientists Use AI to Invent 16 Viruses That Have Never Existed on Earth

A Scientific First With Enormous Stakes

US scientists have used artificial intelligence to create functional viruses that do not exist in nature for the first time — a capability that researchers say could advance medicine but also pose serious biosecurity risks. The findings, published on August 7 in the journal Science, sent shockwaves through the scientific community and reignited urgent conversations about whether the world is prepared for what AI can now do in a biology lab.

Researchers at Stanford University and the Broad Institute of MIT and Harvard used AI to design thousands of complete viral genomes. Using two large language models trained on the genomes of more than two million bacteriophages, the scientists generated new, not-found-in-nature viruses, using the actual genome of a bacteriophage called ΦX174 as a template. The team used AI to recognize natural DNA patterns before applying that knowledge to create viruses not found in nature.

From Thousands of Designs to 16 Living Viruses

The AI models generated thousands of bacteriophage genomes, nearly 300 of which the researchers chemically synthesized. Of these, 16 proved viable and able to kill certain strains of Escherichia coli — including strains that had mutated to be resistant to the original reference genome. That last detail is particularly significant. In lab tests, the AI-engineered phages successfully killed resistant bacterial strains that had evolved immunity to naturally occurring viruses.

The impetus for this work, according to the researchers, is that drug-resistant bacteria are rife and growing. More than 2.8 million antimicrobial-resistant infections occur in the U.S. each year, killing more than 35,000 people on average, according to the Centers for Disease Control and Prevention. AI had previously been used extensively to design individual genes and proteins, but the researchers said this was the first demonstration that generative models could design an entire functional viral genome. That distinction matters — it's a leap from editing nature's blueprints to writing entirely new ones.

Medical Promise Meets Serious Concern

Writing an accompanying commentary in Science, biosecurity leaders Dr. Thomas Inglesby and Dr. Moritz Hanke of the Johns Hopkins Center for Health Security warned the Stanford research proves AI is now capable of inventing dangerous bioweapons. They called for strict laws regarding future research and said it should be illegal to use similar generative techniques on pathogens affecting humans, animals, or agricultural crops.

Fatemeh Vafaee, a professor at the UNSW School of Biotechnology and Biomolecular Sciences, said that while the study itself posed no tangible risk to humans — as phages can only infect bacteria — the methods involved could conceivably be used for harm. "It's less 'should we worry about this virus' and more 'AI can now do this at all,'" she said, calling for stronger biosecurity oversight as a forward-looking precaution.

Regulation Is Already Playing Catch-Up

Biotechnology is regulated through a patchwork of laws and agencies, with breakthroughs frequently outpacing regulators. The Trump administration issued a policy in June to regulate high-risk life sciences research, including experiments that could make harmful biological agents more dangerous. But critics note that policy was designed with natural pathogens in mind — not AI-invented ones that have never existed before.

Hsu Li Yang, director of the Asia Centre for Health Security in Singapore, cautioned that the implications of the study should not be overstated. "It is certainly not the case that anyone with some scientific and laboratory background can now make life-saving or dangerous viruses in their garage," he said, noting that "the downstream wet laboratory capability for the steps post-design is still substantial and has not changed." Still, one of the researchers told Nature that the next step is "AI-generated life" — a prospect that makes both the promise and the peril of this moment impossible to ignore.