AI Designs Bacteria-Killing Viruses, Sparking Debate on Artificial Life

In mid-September 2025, a team of Stanford University scientists published a paper describing their use of artificial intelligence (AI) to design new viruses capable of killing bacteria. This development has sparked a debate within the scientific community about the implications of AI's ability to create novel life forms.

The experiment involved training an AI model called "Evo" on the genomes of living organisms, allowing it to propose new DNA sequences for a simple type of virus known as a phage. The scientists then synthesized and tested around 300 of these AI-generated phage genomes, with 16 proving to be functional and able to infect and kill E. coli bacteria.

Reactions to this achievement have been mixed. Some experts, such as genome pioneer J. Craig Venter, see it as a natural progression of existing techniques, while others, like bioethicist Gregory Kaebnick, are more unsettled by the prospect of machines "rethinking what it is to be human, what it is to be alive." The debate centers on whether this represents a significant milestone in the creation of artificial life or simply a powerful new tool that builds on previous scientific advancements.

Regardless of the interpretation, the potential risks of this technology are clear. As AI becomes more adept at designing novel organisms, there are concerns about the possibility of creating dangerous pathogens or toxins that could threaten human and environmental health. Scientists are already working to address these biosecurity challenges, but the long-term implications of this technology remain uncertain.