A vaccine whose active component was designed entirely by artificial intelligence has been safely tested on humans for the first time, according to a study published in the June 2026 issue of the Journal of Infection. The trial represents a significant step in using AI to accelerate vaccine development against rapidly mutating viruses.
The vaccine, called pEVAC-PS, was developed by researchers from the University of Southampton, the University of Cambridge, and DIOSynVax Ltd. As Healthline reported, the trial is the first time a vaccine designed by computer simulations has been administered to human volunteers.
The technology works by using a super-antigen designed by AI to create immune protection against a wide range of viruses, even as those viruses mutate. Current vaccine development methods do not have this capability, which is why vaccines for viruses like influenza and COVID-19 require frequent reformulations to stay effective against new variants.
The goal of pEVAC-PS is broader than just current COVID-19 strains. Researchers designed it to defend against future coronavirus threats that might jump from animals to humans, a concern that public health officials have raised in the context of pandemic preparedness.
The phase one clinical trial enrolled 39 healthy adults between the ages of 18 and 50, all of whom had received two or three prior COVID-19 vaccine doses but had not recently contracted the virus. Participants were divided into four groups and given different doses: 0.2 mg, 0.4 mg, 0.8 mg, and 1.2 mg. Each participant received two doses, the first on day zero and a second on day 28.
The vaccine was delivered without a needle, using a device called the PharmaJet Tropis, which administers the vaccine intradermally, just under the skin. The needle-free method reduces the need for specialized medical equipment and lowers the risk of needle-related injuries and infections. Researchers noted this makes the vaccine easier to administer and store in places with limited medical resources.
Volunteers were monitored for side effects for seven days after each dose. They recorded local reactions such as redness or pain at the injection site, as well as systemic symptoms like fever or fatigue. Blood samples were collected at multiple points before and after vaccination to measure immune responses.
The vaccine was found to be safe and produced no significant side effects. However, the authors cautioned that a larger trial is needed to determine whether pEVAC-PS can provide strong protection across a more diverse population. The trial was conducted in the UK between December 2021 and September 2023.
Researchers also noted that the use of AI in vaccine design dramatically shortens the development timeline compared to traditional methods, which could be critical in responding to future outbreaks.
