A team at the University of Cambridge reports that, for the first time, a vaccine’s key immune-stimulating component was created entirely by AI before being tested in humans. The experimental vaccine is designed to target a wide range of coronaviruses, including all known Covid-19 variants, as well as related viruses found in animals that could potentially spill over into humans.

Although still in the early stages of development, scientists are also working on similar AI-designed vaccines aimed at influenza and Ebola, raising hopes for more universal protection against rapidly evolving diseases.

Vaccines work by training the immune system to recognise and fight infections. However, many viruses mutate quickly, changing their surface structures and reducing the effectiveness of existing vaccines. This is why flu and Covid vaccines need regular updates.

“We’re always behind,” said Professor Jonathan Heeney from the University of Cambridge to the BBC. “What we’re trying to do is get ahead of the curve.”

How the AI-designed vaccine works

Traditional vaccines are usually developed based on a specific current strain of a virus. In this approach, researchers instead used genetic data from multiple coronaviruses collected through global surveillance of potential pandemic threats.

This genetic information was processed by artificial intelligence, which then designed a “super-antigen” intended to train the immune system to recognise a broad family of related viruses. The goal is to provide protection even if the virus mutates or crosses from animals into humans.

Antigens are the active components in vaccines that trigger immune responses. According to the researchers, this is the first time an AI-designed antigen has been tested in human volunteers.

Professor Heeney described the results as unexpectedly encouraging, saying the technology was “surprising all of us” and showing “what we can do with it for the good of humanity”.

He added: “This is about making vaccines that protect us not just against today’s viruses, but also those that could cause future outbreaks. This represents a fundamental shift in pandemic preparedness.”

Early trial results

The initial clinical trials involved 39 participants and primarily focused on safety. A second, larger trial involving around 200 volunteers is underway to better assess how effectively the vaccine stimulates the immune system.

Findings published in the Journal of Infection suggest the immune response was modest, but researchers say the results are still encouraging at this early stage.

Professor Saul Faust of the University of Southampton, who was involved in the trials, said the AI approach shows clear promise. “The technology is much better at designing vaccines for potential pandemics when viruses are changing,” he said.

Broader applications

The Cambridge team is also testing AI-designed vaccines for other diseases. These include a universal flu vaccine that would not require annual updates, a vaccine for H5N1 bird flu in case it spreads to humans, and experimental vaccines targeting viral haemorrhagic fevers, including Ebola-related viruses.

One current Ebola outbreak in the Democratic Republic of Congo involves a strain for which no approved vaccine exists.

Professor Andy Pollard of the Oxford Vaccine Group, who was not involved in the research, said early animal studies have produced promising results.

He noted that the real challenge will be understanding how the approach performs in humans, whose immune systems are shaped by years of exposure to infections.

A potential shift in vaccine development

Experts believe artificial intelligence could transform how vaccines are developed in the future.

Professor Pollard described AI as a potential “game changer” that could predict immune responses more accurately and speed up vaccine development significantly.

Professor Marian Knight, scientific director at the National Institute for Health and Care Research, said the study represents a major step forward in developing broad, long-lasting protection against viruses.

UK Science Minister Lord Vallance also welcomed the findings, calling them an example of how combining scientific expertise with artificial intelligence can accelerate medical innovation.

While researchers caution that the technology is still in its early phases, they say the initial human trial results suggest AI could play a major role in preparing for and preventing future pandemics.