Stanford researchers use AI to create viruses that kill bacteria

Scientists from Stanford University and the Arc Institute have successfully created and printed viruses with AI-designed DNA that can target and kill specific bacteria, marking the first time AI has generated functional genome-scale sequences. The breakthrough demonstrates AI’s potential for bioengineering applications while raising significant ethical concerns about the technology’s potential misuse for creating bioweapons.

How it works: Researchers used an AI model called Evo, specifically trained on millions of bacteriophage genomes, to design new virus sequences.
• The team chose phiX174, a well-studied virus that infects E. coli bacteria, as their starting point due to its simple structure of around 5,400 base pairs and 11 genes.
• Evo generated 302 different virus designs, which researchers then chemically assembled and tested against real E. coli strains.
• Of the 302 AI-designed viruses, 16 successfully infected their bacterial hosts, reproducing and killing the cells in the process.

Why this matters: This represents the first demonstration of AI creating entire functional genomes that work in real-world biological systems.
• “This is the first time AI systems are able to write coherent genome-scale sequences,” senior author Brian Hie, a Stanford computational biologist, told Nature. “The next step is AI-generated life.”
• The AI-designed viruses outperformed natural phiX174 in many cases, killing three different E. coli strains despite carrying major genome alterations that humans would be unlikely to design.

Key performance details: While 16 successes out of 302 attempts represents roughly a 5% success rate, experts consider this a remarkable achievement for AI-generated biological systems.
• According to Niko McCarty, a former bioengineer for Caltech and Imperial College London, “In many cases, they were more infectious than wildtype phiX174 despite carrying major genome alterations that a human would be unlikely to rationally design.”
• Some of the AI-written genomes are “so distinct from any known bacteriophage genome that they would technically be classified as their own species.”

Therapeutic potential: The research team sees significant promise for medical applications using AI-designed viruses to target harmful bacteria.
• “It was quite a surprising result that was really exciting for us because it shows that this method might potentially be very useful for therapeutics,” coauthor Samuel King told Nature.
• However, King cautioned that “a lot of experimental advances need to occur in order to design an entire living organism.”

Safety concerns: The breakthrough has sparked warnings from experts about potential dual-use risks and bioweapon applications.
• Craig Venter, founder of the J. Craig Venter Institute, urged extreme caution: “One area where I urge extreme caution is any viral enhancement research, especially when it’s random so you don’t know what you are getting.”
• “If someone did this with smallpox or anthrax, I would have grave concerns,” Venter told MIT Technology Review.
• The research could potentially be abused to create bioweapons or unintentionally produce dangerous, out-of-control viruses.

The big picture: While Venter noted that the AI approach is essentially “just a faster version of trial-and-error experiments,” the ability to generate functional genomes at scale represents a significant leap forward in synthetic biology capabilities.