It’s an ambitious goal set by the Coalition for Epidemic Preparedness Innovations (CEPI): develop safe, effective vaccines within 100 days of a new virus emerging by advancing rapid-response platform technologies and creating a global network of research and manufacturing that contain outbreaks before they escalate into crises.

In the aftermath of the COVID-19 pandemic, CEPI is looking to redefine global preparedness. Regionalized vaccine manufacturing is critical to that goal as production is highly concentrated globally, making lower- and middle-income countries vulnerable to supply shocks, regulatory bottlenecks, and slow pandemic response.  

To help address the disparity between wealthier industrialized nations and poorer developing ones, CEPI currently supports more than 150 R&D and manufacturing projects in nearly 50 countries. BioNTech, a CEPI partner, is building an mRNA vaccine manufacturing facility in Kigali, Rwanda, designed to strengthen local vaccine production capacity and expand access to essential vaccines across Africa.

Samsung Biologics has also recently partnered with CEPI to support vaccine manufacturing preparedness by leveraging its mammalian cell-based infrastructure and quality systems to create a process that can be activated quickly during an outbreak, providing access to up to 50 million vaccine doses during a future pandemic.  

However, while CEPI’s network of vaccine manufacturing partners on three continents is helping to boost sustainable production of vaccines and outbreak response, challenges remain in speeding up vaccine design, improving the efficiency of preclinical and clinical studies, and enhancing biomanufacturing processes to achieve its 100 Days Mission goal.

In February 2026, CEPI launched a new five-year strategy — CEPI 3.0 due to begin in 2027 — to “accelerate the development of vaccines and other biologic countermeasures against epidemic and pandemic threats so they are accessible to all people in need.”

Under the strategy, CEPI is looking to speed up the development of platform technologies such as artificial intelligence (AI) and embed them with regional manufacturers so countries can move quickly against outbreaks. A critical piece is what CEPI calls a Pandemic Preparedness Engine, described as a kind of ChatGPT for vaccine developers.

“Within a single, secure platform, the Pandemic Preparedness Engine will be designed to integrate multiple vast and disparate datasets including gene sequences and potential viral mutations to help design vaccines, supported by epidemiological data to inform preclinical and clinical trial findings, and key insights on regulatory assessments and state-of-the-art manufacturing technologies to enable rapid scale-up and supply,” according to CEPI.

AI and regional vaccine manufacturing

Stakeholders gathered earlier this month at CEPI’s Global Vaccine Manufacturing Summit in London to discuss what needs to happen next when it comes to innovation, regulatory policy, and financing to achieve its 100 Days Mission. Trevor Mundel, president of global health at the Gates Foundation, a CEPI founding member, touted the potential for AI to enable regional vaccine manufacturing and speed production by spreading the know-how of how to make vaccines.

“Technical expertise is no longer going to be the preserve of those places that have got a lot of resource and a lot of investment,” Mundel told the summit in a keynote address. “We are going to be able to diffuse this across the globe. So, wherever there is a need, we will have the expertise to do the kinds of vaccine manufacturing that we need locally.”

The vision is to take the accumulated wisdom around the production of vaccines and combine it in a system that can be downloaded wherever it is needed globally, according to Mundel.

At the same time, Mundel said CEPI is aware of the emerging biosecurity risks associated with AI. That’s why CEPI and its partners are taking a biosecurity-by-design approach for CEPI 3.0 that factors in the risks and looks to build robust mitigations, as the Pandemic Preparedness Engine concept and capabilities are being refined and tested.

Biosecurity is becoming “much more like cybersecurity,” Mundel warned, which he described as a continuous “cat and mouse game of attack and defense.” As a result, he said the current security environment raises the bar for CEPI 3.0 in terms of the timelines, countermeasures, and the surveillance techniques needed globally.  

“CEPI 3.0 is going to be one of the major vehicles that’s going to work this out,” Mundel concluded.

Chemistry, manufacturing and controls

The goal of CEPI 3.0 is to realize CEPI’s 100 Days Mission with vaccine manufacturing partners serving a critical role in “using real-world responses to test systems and to eliminate obstacles, bottlenecks, and barriers to speed,” according to CEO Richard Hatchett.

CEPI is working with partner manufacturing facilities to expand their preparedness capabilities so they “can shift into response mode very rapidly,” Hatchett said.   

Bruno Tricoire, director of chemistry, manufacturing and controls (CMC) at CEPI, led a session at this month’s summit on how AI and advanced data tools can strengthen vaccine manufacturing across development, scale-up, quality, and operations.

“In the CMC and manufacturing space, there are so many things we can do with AI,” Tricoire said, noting that the application of AI to process and efficiency data can speed up vaccine production. “Obviously, we have a lot of data and information to process” with the potential for “different ways to be able to manufacture and release material.”

However, Tricoire emphasized that any AI approach to regional vaccine manufacturing must be grounded in fit-for-purpose use cases, locally deployable solutions, and strong data foundations that work for low- and middle-income countries — not just high-income-market models.

Based on practical use cases discussed at the summit, including process optimization, predictive maintenance, quality assurance, and tech transfer, he said CEPI will prioritize the “low-hanging fruit” projects that will be activated in the coming year as part of a “CMC AI Engine” — its working name for the time being.