Scientists at the University of Cambridge have created a new type of vaccine developed with artificial intelligence. Its main element was generated entirely by AI and has now undergone initial human testing. The aim is a single shot that protects against all known human coronavirus variants as well as related bat viruses that might spark future outbreaks.
Standard vaccines prepare the immune system against one particular virus. Mutations can render them ineffective, requiring annual flu shots and repeated COVID-19 updates. AI addresses this issue by examining genetic sequences from many related viruses to find stable regions unlikely to change. Targeting these conserved parts could yield protection across an entire virus family.
The Cambridge group applied AI to sarbecoviruses, the group that includes SARS and COVID-19 pathogens plus several animal coronaviruses, identifying evolutionarily preserved features. These became the foundation of the new vaccine.
Unlike the mRNA vaccines used in the pandemic, this candidate uses DNA. DNA vaccines are more stable, easing storage and transport in regions with limited refrigeration. They can also be given without needles through a high-pressure liquid stream, reducing pain and simplifying mass administration.
Such practical benefits matter if the vaccine can shield against viruses not yet encountered. Broad-spectrum products might allow faster responses to emerging threats and reduce the need for annual strain predictions, as seen with influenza. They could also address outbreaks like the recent Bundibugyo Ebola cases that evade current shots.
Early trial results indicate the DNA vaccine safely triggered antibodies that recognize multiple sarbecoviruses. Responses were modest, however, and the duration of protection remains unclear. Larger studies are required to confirm real-world effectiveness and the possible need for boosters. A universal vaccine is still several years from availability, yet the findings suggest AI can accelerate progress toward broader protection.
