AI Vaccine Research Surges as the First AI-Designed Vaccine Enters Trials

AI-led approaches to vaccine design are expanding rapidly, with the number of research papers on AI-enabled vaccine development increasing by 5,200% over the past 10 years, according to data from Elsevier’s Scopus database. This coincides with the first human trial of a vaccine whose key component was designed entirely by AI.

The AI-designed vaccine, developed by researchers from the University of Cambridge and its spin-out DIOSynVax, uses a “super-antigen” intended to provide lasting protection against multiple Sarbeco coronaviruses—even as they mutate.

The researchers state that advancing to human trials demonstrates the safety of a fundamentally new approach to vaccine design that uses computer simulations.

To design the antigen, researchers used all the available genetic sequence data for Sarbeco coronaviruses logged by global viral surveillance programs. Machine learning was used to analyze the sequencing data and to design an antigen to protect against this group of viruses, including variants that are yet to emerge.

The AI-designed vaccine represents a step change from traditional vaccines, which use antigens from specific virus strains or variants already detected in humans. This typical approach to vaccine design necessitates annual vaccination updates as viruses mutate.

“We’ve converted vaccine development from being reactive to being future-proof. Our vaccines will continue to provide protection against viruses even as they mutate into new strains,” Professor Jonathan Heeney, the scientific lead of the research, said in a press release.

The Phase 1 trial of the vaccine involved 39 healthy volunteers aged 18 to 50 years who had received 2 or 3 prior doses of a COVID-19 vaccine and had no recent confirmed infection. Participants received escalating doses of the vaccine, administered at day 0 and day 28. The results were recently published in the Journal of Infection.

The DNA vaccine was administered via needle-free intradermal injection. This delivery method offers an alternative to those with a fear of needle-based injections. It could also help make vaccination easier to carry out in settings where conventional injections are more challenging to deliver.

The vaccine was well-tolerated and elicited no safety concerns. The majority of adverse events reported were mild or moderate. Fewer events were reported after subsequent doses, suggesting good tolerability of repeated injections.

Although the safety profile was encouraging, the vaccine’s impact on the immune system was only modest and variable. There was a small increase in neutralizing antibodies against the Delta and Omicron variants, while activity against the ancestral Wuhan strain and SARS-CoV-1 was limited.

Differences amongst the participants in prior SARS-CoV-2 exposure and vaccination history likely influenced these results. Recruitment for the study took place during successive waves of Omicron infections and booster campaigns, complicating the interpretation of vaccine effects.

In a peptide microarray analysis, the researchers confirmed that the vaccine was directing immunity towards conserved viral regions. It revealed antibody binding to receptor-binding domain epitopes, including known broadly reactive antibody sites. While such binding does not necessarily translate into strong in vitro neutralization, these same binding events have been associated with appreciable in vivo protective activity through Fc-mediated mechanisms.

Breaking the reactive vaccine development cycle

This new class of AI-designed universal vaccines has the potential to decouple vaccine development from crisis cycles and protect against many variants simultaneously.

Traditional vaccine research tends to follow outbreaks; funding and development increase when a threat becomes visible, then fall as the immediate risk declines.

Recent data from Scopus, Elsevier’s abstract and citation database of peer-reviewed academic literature, shows how Ebola vaccine research follows this trend (Figure 1).

In 2015, the number of research papers on Ebola vaccine development peaked, coinciding with the 2014–2016 West African Ebola crisis. Publications rose again in 2020–2021, around the time the first Ebola vaccine for Zaire ebolavirus was approved. Since then, papers have declined, likely reflecting the reduced urgency, despite other Ebola virus species continuing to pose threats.

Figure 1: The number of research papers present in the Scopus database published on Ebola vaccine development over time. Credit: Data courtesy of Elsevier.

More recently, growing concerns about potential spillover of H5N1 from birds to mammals and humans have resulted in a spike in H5N1 vaccine research (Figure 2). After fluctuating over the past decade, H5N1 vaccine publications rose sharply in 2024 and 2025, likely in part as a response to the multistate US dairy cattle outbreak reported in March 2024.

Figure 2: The number of research papers present in the Scopus database published on H5N1 vaccine development over time. Credit: Data courtesy of Elsevier.

Advancing RNA Delivery Beyond Vaccines

This eBook explores key scientific advances in LNP development that are shaping the future of RNA delivery, including repeat dosing, lyophilization, harnessing multi-antigen mRNA, and organ-specific targeting.
View eBook

Advertisement

While traditional vaccine design often responds to visible threats, the new vaccine technology is designed to proactively tackle outbreaks and unknown spillover events. Universal vaccine technology could mark the end of the “reactive” vaccine development system, which struggles to keep pace with evolving viruses.

“This new class of universal vaccines are future-proofed. They not only protect against many variants simultaneously, but potentially against related viruses that haven’t yet emerged and spilt over to humans,” said Prof. Saul Faust, the trial’s chief investigator. “If we can develop and clinically advance this new class of vaccines before a virus outbreak begins, millions of lives could be saved, lockdowns avoided, and the economy preserved.”

Momentum in AI vaccine research

The completion of the first human trial of an AI-designed vaccine is the culmination of increasing interest in the use of computational biology and machine learning in vaccine development.

According to Scopus data, in 2015, only four papers were published on AI-enabled vaccine development. A decade later, this figure jumps up to 212 papers. Leading the growth in AI vaccine research are India, the United States, and China, which account for over half of the papers published on the topic in the Scopus database since 2015 (Figure 3).

Figure 3: The number of research papers on AI and vaccine development included in the Scopus database by country (top 15 countries shown). Credit: Data courtesy of Elsevier.

The jump in AI-designed vaccine publications points to a larger pivot towards more predictive immunology and technology that can provide broad, lasting viral protection.

Further development of the Sarbeco coronavirus vaccine is needed before it can be rolled out to the public. A larger Phase 2 trial in a wider, more diverse population is required to confirm that the vaccine generates strong, broadly protective immunity.

“Given the ongoing risk of SARS-CoV-2 evolution and the potential for future zoonotic coronaviruses, developing broadly protective vaccines remains a critical global health priority,” the researchers concluded.

Reference: Munro AP, Ferrari M, Kinsley R, et al. A phase I, needle free, dose escalation clinical trial of pEVAC-PS, a candidate pan-Sarbecovirus Vaccine. J Infect. 2026;92(6). doi: 10.1016/j.jinf.2026.106759