Scientists have harnessed artificial intelligence to design novel viruses that were able to destroy cells in laboratory tests.
The breakthrough is being described as the first successful use of the technology to produce complete genomes — the full genetic blueprint required to create a functioning organism.
Advocates say the research could open the door to powerful new medical treatments, while critics argue it also brings “urgent” questions over biosafety and security.
The study, carried out by researchers at Stanford University in California, used AI to design the genome of a virus that targets bacteria.
After the system proposed thousands of possible genomes, the team synthesized 302 of them in the laboratory and tested them against bacteria.
Of those AI-designed candidates, 16 proved capable of killing E. coli.
The engineered viruses were bacteriophages, a class of viruses that infect only bacteria and cannot infect human, animal or plant cells.
Announcing the findings, Dr Brian Hie, a chemical engineer involved in the work, said: “In this case, we wanted the model to generate the entire genome end-to-end in a single left-to-right pass. We didn’t add anything.”
Scientists have used AI to design a new virus that was capable of infecting other cells
The research was published alongside an accompanying article warning over the potential risks of the advance.
Written by Johns Hopkins experts Dr Thomas Inglesby and Dr Maurice Hanke, it warned: ‘Although this is promising for life sciences applications, it also raises urgent biosafety and biosecurity questions.
‘The ability to compose viral genomes using generative AI now exists; the governance to safely steer it does not.’
For the research, published in Science, scientists used the AI tools Evo1 and Evo2.
These work in a similar way to large language model chatbots ChatGPT and Grok, except they have been trained on genetic codes rather than written text.
Researchers trained the models on two million genomes for bacteriophages, and then tasked them to create new potential genomes.
Scientists synthesized the AI-made genomes in the lab and put them into petri dishes containing E.coli, prompting the bacteria to start making copies of the viruses.
Petri dishes were then monitored to establish whether the bacteriophages had started to attack and kill the bacteria.
Samuel King, a PhD student in the lab, told the BBC: ‘We were starting to see these clear spots and it was just extremely exciting.’
They wrote in the paper: ‘This work provides a blueprint for the design of diverse synthetic bacteriophages and useful biological systems at the genome scale.’
Scientists said bacteriophages have one of the smallest genomes known, making them much easier to create.
They said, however, that the work was a step towards AI being used for more advanced research.
Dr Patrick Cai, a researcher at the University of Manchester in the UK, said: ‘While these are relatively small bacteriophage genomes, the significance extends far beyond phages.
‘It suggests that genome language models are beginning to learn the design principles encoded by evolution, opening the door to AI-assisted genome writing.’
Tom Ellis, a professor of synthetic genome engineering at Imperial College London, called the work impressive but noted it highlights the challenges of creating larger, more complex genomes.
He told The Guardian: ‘This is literally the smallest and easiest genome to make.’
He added that an AI trained on dangerous pathogens could theoretically be used to design harmful viruses, but controlling access to genetic data and restricting the synthesis of risky genomes would help mitigate that risk – and governments are already working on these measures.
Still, he cautioned against overblowing the threat: ‘The threat from full AI design and writing of a genome of a virus or bacteria is very overblown.
‘Just taking existing pathogens and making gain-of-function changes to their genomes is so much easier and much more likely to be a real pathogenic threat.’
Gain-of-function research is essentially the scientific practice of genetically altering a pathogen to study how it might evolve; enhancing traits like transmissibility, virulence or host range to better understand and prepare for future pandemic threats.
But the term became a lightning rod during the Covid pandemic, fueling fierce debate over whether such experiments at the Wuhan Institute of Virology – some of which were funded by US taxpayer dollars – played a role in the virus’s origins.
