Researchers at the University of Cambridge have developed what they describe as a fundamentally new type of vaccine using artificial intelligence. The vaccine’s key component was designed entirely by AI and has now been tested in people for the first time.
Goal
A single vaccine targeting all known human coronavirus variants and related bat sarbecoviruses that could jump to humans and cause future pandemics — moving beyond strain-specific immunity that requires annual flu shots and repeated Covid updates.
AI approach
By analyzing genetic data from thousands of related viruses, AI identifies parts that stay the same across strains and are unlikely to mutate. The Cambridge team scanned the sarbecovirus family (SARS, Covid, animal coronaviruses) for shared evolutionary-stable features used as vaccine targets.
DNA vaccine platform
Unlike mRNA pandemic shots, this vaccine uses DNA — more stable for storage and transport, advantageous in lower-income countries with limited cold-chain infrastructure. Needle-free administration via high-pressure liquid stream through the skin.
Trial results
First human trial of an AI-designed vaccine. Results showed the DNA vaccine stimulated immune responses producing antibodies recognizing different sarbecovirus types. Technology was safe and well tolerated.
Immune responses were modest. Duration of protection, booster requirements, and real-world infection prevention remain uncertain. Larger trials needed.
Broader implications
Broad-spectrum vaccines could transform pandemic response and targets like influenza and Ebola (recent Bundibugyo strain outbreak bypassed existing vaccines). Universal vaccine remains years away but AI may accelerate development.
This article is republished from The Conversation under a Creative Commons license.