Cambridge AI-Designed Coronavirus Vaccine Clears Phase I Safety Trial

A vaccine whose active ingredient was generated entirely by computer simulation has completed its first human trial, researchers at the University of Cambridge and spinout company DIOSynVax reported on June 5. The results were published in the Journal of Infection.

<cite index="14-1,14-2">The findings describe the first completed human clinical trial of a vaccine whose key component was designed entirely by artificial intelligence, according to findings published June 5, 2026, in the Journal of Infection. The Phase I trial tested the pEVAC-PS vaccine developed by Cambridge and DIOSynVax to protect against multiple Sarbeco coronaviruses.</cite>

The headline safety result is narrow but meaningful for a Phase I study: <cite index="16-1,16-2">the experimental vaccine was found to be safe and caused no significant side effects in a study involving 39 healthy volunteers.</cite> Safety and tolerability are the primary endpoints at this stage; the study was not powered to demonstrate efficacy.

The immune data are more striking in breadth than in magnitude. <cite index="19-6">The vaccine triggered immune responses in the volunteers not only to SARS-CoV-2 and SARS, but to related bat viruses that could potentially jump from animals to humans and cause future pandemics.</cite> That cross-reactivity to zoonotic strains that have not yet caused large human outbreaks is the central scientific claim the team is making. Independent replication will be needed before that claim can be considered settled.

The delivery mechanism is also worth flagging. <cite index="23-5,23-6">The Phase I study was conducted at National Institute for Health and Care Research Clinical Research Facilities in Southampton and Cambridge, and participants received a DNA-based vaccine through a needle-free microfluidic jet system.</cite> Needle-free DNA vaccines have a patchy clinical history, so the platform itself remains one of several variables to watch in later-phase work.

The antigen design process is the novel methodological piece. <cite index="17-7">The technology uses machine learning to design what the team calls a "super-antigen" -- a synthetic target built from computational analysis of massive genetic sequence databases collected through global animal virus surveillance programs.</cite> <cite index="17-11">DIOSynVax's platform creates a single antigen designed to represent broad features shared across entire viral families -- in this case, the Sarbecovirus subgenus, which includes SARS-CoV-2, SARS-CoV-1, and numerous bat viruses with pandemic potential.</cite>

<cite index="16-5">DIOSynVax, short for Digitally Immune Optimised Synthetic Vaccines, was founded in 2017 as a University of Cambridge spinout with support from Cambridge Enterprise, the university's commercialization arm.</cite>

The study enrolled 39 people aged 18 to 50 -- a sample size appropriate for a first-in-human dose-escalation design but far too small to draw conclusions about protection rates or durability of immune response. The paper's authors do not claim efficacy; they claim proof-of-concept for safety and immunogenicity. Those are different things, and the gap between a Phase I signal and a licensed vaccine remains large.

What this trial does establish is that a fully AI-designed antigen can pass regulatory review, enter humans, and not cause obvious harm while eliciting measurable immune responses. <cite index="22-6">Researchers describe this as the first time a vaccine active component designed entirely by computer simulations has been tested in people, according to the University of Cambridge and Euronews.</cite> That procedural milestone is real regardless of where the program goes next. Phase II work, with larger cohorts and immunogenicity comparisons against existing coronavirus vaccines, will determine whether the breadth of response seen here translates into anything clinically useful.

Sources cited:
- Journal of Infection (Munro et al., 2026) - via MedicalXpress (https://medicalxpress.com/news/2026-06-ai-universal-vaccine-human-trial.html)
- University of Cambridge official press release (https://www.cam.ac.uk/research/news/new-universal-vaccine-technology-could-protect-us-from-future-virus-outbreaks)
- Interesting Engineering (https://interestingengineering.com/ai-robotics/cambridge-ai-designed-universal-vaccine-phase-1-trial)
- Let's Data Science (https://letsdatascience.com/news/cambridge-tests-ai-designed-universal-coronavirus-vaccine-46ea2448)
- Planet Today (https://www.planet-today.com/2026/06/universal-vaccine-for-viruses-that.html)