The closing days of 2016 brought great news: The world now has an Ebola vaccine that’s 100 percent effective at preventing infections from the strain behind the recent west African outbreak.

But that vaccine, known as rVSV-ZEBOV, was actually created back in 2003, and first tested in monkeys in 2005. Then, it just sat in a freezer. It stalled because developing vaccines is incredibly expensive. It can cost up to $1 billion to test them in large clinical trials, and to build the manufacturing facilities needed to make them. Pharmaceutical companies are unlikely to recoup that investment, especially when it comes to diseases like Ebola, which are rare and tend to hit poor countries.

So, when news broke of the Ebola outbreak in March 2014, at least seven potential vaccines had been created, but none were ready. They had been tested in monkeys, but none had passed even the most preliminary safety checks in humans. The rVSV vaccine was ushered into those safety trials in October 2014. After passing, it was deployed in Guinea the following March, to see if it actually worked. The preliminary results were in by June 2015, and were so promising that health workers immediately started vaccinating anyone who had contact with an Ebola patient.

That timeline was astonishingly fast. “For the first time in history, a vaccine was tested in the middle of an epidemic,” says Jeremy Farrar, the director of the Wellcome Trust, a U.K.-based biomedical research charity. “Maybe five to seven years of work were concentrated into one.”

But it was still too slow. The epidemic was already waning by the time the vaccine was ready for field trials. By the time it finally ended, 11,000 people were dead. “Everyone who went through the Ebola situation agrees that it can’t happen again,” says Nancy Lee, a senior policy advisor at the Wellcome Trust.

It won’t, if a new international coalition called CEPI—the Coalition for Epidemic Preparedness Innovations—has its way. CEPI (pronounced “seppy”) is a global vaccine-development fund, devoted to readying pandemic defenses during peacetime. With $460 million from the Wellcome Trust, the Bill & Melinda Gates Foundation, and the governments of Germany, Japan, and Norway, it will fund the development of vaccines against the likely pandemics of the future, testing them as far as possible, and stockpiling millions of doses. When outbreaks happen, the vaccines will be immediately ready for field-testing and mass-manufacture.

“If CEPI had existed before the Ebola outbreak, the outcome would have been different,” says Farrar. The rVSV vaccine could have been used a full year earlier, and thousands of lives might have been saved.

He and others proposed the idea for CEPI back in July 2015. If the pharmaceutical industry wasn’t interested in funding research and development of vaccines, they wrote, then governments and NGOs would have to step up.

Where to begin? The list of potential pandemic diseases is long and growing. But in December 2016, Mark Woolhouse, from the University of Edinburgh, developed a list of 37 “priority viruses” that are most likely to cause major epidemics in the future. The list includes familiar names like Ebola and rabies, and more obscure ones like Machupo, Oropouche, and monkeypox.  “CEPI can’t take on 37 diseases tomorrow,” says Farrar, so they picked three to focus on within the next five years: Middle East Respiratory Syndrome (MERS), Lassa fever, and Nipah.

MERS hides out in camels and has been circulating in the Middle East since 2012. It has killed a third of the people it has infected, and although it doesn’t seem to spread efficiently from person to person, that might change. Nipah persists in fruit bats and caused a major epidemic in Malaysia in 1999; outbreaks still regularly occur in Bangladesh and India. Lassa fever is spread by rats and is endemic to the same parts of west Africa that were recently hit by Ebola; hundreds of thousands of infections happen every year. “They’re not the only choices you could make,” says Farrar, but they are all severe risks, with potential vaccines already in the early stages of development.

CEPI will fund researchers to test vaccines against these three diseases in mice and monkeys. It will run early trials in humans to show that the vaccines are safe, and that they correctly educate the immune system. And it will plan the final large-scale trials that will run during an outbreak to see if the vaccines actually work. Such trials involve a morass of logistics, ethical approvals, and legal arrangements. “The Ebola outbreak showed that this can happen very quickly,” says Lee. “Legal agreements that could take years were agreed in months. But if things can be agreed ahead of time, it would speed things up even more.”

That agreement is vital: In an outbreak, bureaucracy costs time and lives. By agreeing on pathways for getting their vaccines approved, manufactured, and distributed, CEPI’s goal is to ensure that there are as few obstacles as possible between their vaccines and someone’s arm. When an epidemic hits, health workers can be given a ready-made product, which they know is safe and probably effective, and deploy it according to a ready-made plan.

Speaking of Ebola, CEPI will also be working with the American pharmaceutical company Merck to get the rVSV vaccine through regulatory approval, so that it’s ready for the next outbreak. And it will fund the development of more Ebola vaccines, since rVSV only works against one of the several dangerous strains. “The vaccine was a turning point for global health, but we haven’t finished the job,” says Farrar.

He has even bigger ambitions for CEPI. “My dream would be that 10 to 20 years from now, we have a vaccine for every one of the 37 infections on Mark Woolhouse’s list,” he says. “That won’t be possible for everything, but Ebola showed that there are many infections for which a vaccine is eminently makeable.”

Trevor Mundel, the president of the Global Health Division at the Bill & Melinda Gates Foundation, thinks that dream’s within reach. In creating vaccines for MERS, Nipah, and Lassa fever, he believes that CEPI will push techniques that could be used to rapidly make vaccines for any number of viral diseases—including ones we don’t expect.

A new class of vaccines, known as RNA vaccines, can help with that. Most current vaccines work by presenting the immune system with dead microbes, weakened microbes, or bits of microbes. The Ebola rVSV vaccine, for example, was created by genetically engineering a different virus to make an Ebola protein. When you inject someone with that virus, their immune system can study the protein and prepare defenses against an actual Ebola infection.

But proteins are made using instructions encoded in a molecule called RNA. So instead of injecting someone with the Ebola protein, you could instead give them the RNA instructions and get their own cells to make the protein. In this way, you’re transforming the patient’s own body into a vaccine factory.

The beauty of these RNA vaccines is that once you develop a way of delivering the RNA into a patient—some kind of scaffold or shell—you can theoretically customize it to deal with all kinds of diseases. Just swap the Ebola RNA for that of some other virus, and you’d have a new vaccine in a matter of weeks. And if you could show that the scaffold is safe, no matter whose RNA it carries, then you could speed that new vaccine through the regulatory process.

RNA vaccines are still being developed, and Mundel hopes that CEPI’s efforts will speed them along. If they work, we could eventually bypass the lengthy, expensive business of making artisanal vaccines, and enter the era of customizable, plug-and-play ones. “Everyone agrees that the pathogen that’s going to be of real concern won’t be on our list,” Mundel says. But hopefully, with RNA platforms, “we can rapidly produce a vaccine for whatever that pathogen might be.”