A team from the University of Cambridge has reported the first results of a clinical trial for a universal vaccine developed using artificial intelligence. According to the university, this marks the first time a vaccine whose active component was entirely computer-designed has been tested in humans. No significant side effects were detected in participants.
The study enrolled 39 healthy volunteers aged 18 to 50 and was conducted at medical facilities in Southampton and Cambridge. The vaccine targets the Sarbecovirus subgroup, which includes SARS-CoV-2 (the virus that caused the COVID-19 pandemic), SARS, and related bat viruses that could potentially cause future outbreaks.
After receiving the shot, volunteers developed a protective immune response against several viruses in this group simultaneously. Scientists believe this approach could help intercept threats before they emerge, rather than developing vaccines after an epidemic begins. In the future, similar super-antigens could be used against other dangerous infections, including influenza and Ebola.
To create the vaccine, researchers fed known genetic sequences of Sarbecoviruses from around the world into an AI model. Machine learning helped design an antigen that reflects the common features of the entire virus group, not just a single strain. This should allow the vaccine to remain effective even as new variants appear.
Professor Jonathan Heeney from the University of Cambridge's Laboratory of Viral Zoonotics said the team is trying to shift vaccine development from a reactive mode to one of preparation for future threats. He noted that traditional vaccines often protect against a limited set of variants, and the new approach could free medicine from the constant race to keep up with circulating mutations.
For now, this is a small initial trial, so it's too early to draw final conclusions. The next phase should include more participants with more diverse backgrounds to confirm safety and assess efficacy more broadly. However, the successful testing of a vaccine with a fully AI-designed active component already looks like an important step toward a new model of pandemic preparedness.