A team at the University of Cambridge has completed the first human trial of a vaccine with an active ingredient designed entirely by artificial intelligence. The experimental vaccine showed no significant side effects in 39 healthy volunteers and triggered immune responses against multiple coronavirus strains.
This represents a major shift in how vaccines could be developed. Instead of racing to catch up with new virus variants after they spread, scientists used AI to create a ‘universal’ vaccine that could protect against diseases that don’t even exist yet.
How the AI vaccine works
The researchers fed their AI model genetic sequence data from Sarbeco coronaviruses – a family that includes SARS-CoV-2, the virus behind COVID-19. The machine learning system then designed an antigen containing features common across the entire virus group.
This approach produced what the team calls a ‘super-antigen’ that triggered protective immune responses against:
- SARS-CoV-2 (COVID-19)
- SARS
- Related bat viruses that could cause future pandemics
The trial involved 39 volunteers aged 18-50 at medical facilities in Southampton and Cambridge. All participants showed immune responses without serious side effects.
Breaking the vaccine update cycle
Traditional vaccines face a constant challenge: viruses mutate faster than scientists can update vaccines. This creates an endless cycle of playing catch-up with new variants.
‘We’ve converted vaccine development from being reactive to being future proof,’ said Professor Jonathan Heeney from Cambridge’s Department of Veterinary Medicine, who led the research. ‘Our vaccines will continue to provide protection against viruses even as they mutate into new strains.’
The approach could extend beyond coronaviruses to other diseases that jump between humans, including flu and Ebola.
What comes next
The small trial size means more testing is needed. The next phase will include a larger, more diverse group of participants to better assess the vaccine’s effectiveness across different populations.
If successful, this AI-driven approach could fundamentally change pandemic preparedness. Rather than waiting for the next outbreak to develop vaccines, scientists could create broad-spectrum protection against entire virus families before new threats emerge.