ИИ-платформа Catalyze помогает донорам находить и финансировать перспективные научные проекты

Back in the early years of the COVID-19 pandemic, Sandeep Patel was directing an R&D division of BARDA (the Biomedical Advanced Research and Development Authority), a government agency that prepares for pandemics and other major public health threats. While BARDA worked to fund development of the COVID vaccine, as well as related research, Patel often saw outside funders approach the agency, checkbooks in hand, offering to help. “A lot of people came to us, a lot of funders, basically raising their hand, saying, ‘We want to help. What do we do?’” Patel said. The problem was that there was no system to tell them which credible projects needed support. “Funders either just misallocated money or they didn’t know where to put money, or they just held back money, even though they wanted to deploy it.” Patel had seen similar disconnects between capital and research during his years as a consultant working with government agencies like the NIH. Though that agency gives out billions in research grants, it is typical for most research proposals to go unfunded — even many that receive high evaluations by the most rigorous review groups. In a tale that won’t be unfamiliar to grantseekers outside the scientific realm, this high rate of refusal means scientists seeking funding must continue to write grant application after application to potential new funders, tasks that can eat up major chunks of a scientist’s time — maybe as much as 30% or 40% of their work hours, rather than doing the actual science. For society at large, the end result is that technological innovation is slower, and biomedical research and treatments take longer to arrive or may never be developed. Catalyze leverages grant review work that’s already happening to draw in donors What was lacking, Patel realized, was a good way — a safe and efficient marketplace or infrastructure — to connect scientists with interested funders. “There is no shortage of ambitious ideas that researchers have, and yet we have this failure to connect these nodes together in any meaningful way,” he said. “We really need this information infrastructure that allows people to coordinate.” With a team of scientists, technologists and research funding professionals, Patel founded Catalyze. It’s a digital platform designed to allow researchers to post projects and proposals, while philanthropists and other funders use AI-assisted search to identify — and support — projects that fit their interests. A key element of Catalyze is its leveraging of the existing grant review process that has already been conducted by peer scientists at agencies like the NIH and NSF. It’s a detail that most of us laypeople may not appreciate: The grant review process by experts at the direction of the NIH and other agencies is no small thing. It involves considerable taxpayer expense, not to mention the volunteer assistance of scientists, as well as the time spent by the scientists writing the grant applications. Patel says that money and effort should not go to waste or be repeated time and time again every time a scientist applies to a different organization for a grant. “Rather than us trying to figure this out on our own, we can just pick off their highly rated, unfunded proposals,” said Patel. What kind of numbers are we talking about? Last year, wrote Patel, the NIH reviewed 62,592 grant proposals and funded only 8,161, or 13%. That left thousands of proposals in the top quartile that went unfunded — “due to a lack of money rather than a lack of merit,” said Patel. Patel, who has a Ph.D. in physical chemistry, has worked in science funding and biomedical innovation for much of his career. At BARDA, he was director of the Division of Research, Innovation, and Ventures (DRIVe), which deployed almost $1 billion toward research in response to COVID-19. While at BARDA, he also pioneered a public-private venture capital fund focused on health security. His partners at Catalyze include Amit Gupta, a specialist in AI and digital infrastructure, and David Bowen, former deputy director for global health policy at the Gates Foundation. The Catalyze platform uses AI to help potential funders find research proposals that align with their own interests. In other words, Catalyze’s AI function doesn’t vet the proposals — that’s the job of the human reviewers through NIH’s usual procedures — rather, the AI makes proposals easier to find and compare, and thus to allow funders to identify projects they want to support. Catalyze has launched quietly, said Patel, with some 80 to 100 projects in its database. At the time we spoke, he said one project had so far attracted funding from multiple supporters. Attracting new money to science funding Established philanthropic foundations and veteran funders that have been supporting science often have their own teams and scientific review boards to assess grant proposals or hire consultants to assist in the process. But the promise of a solution like Catalyze, Patel, said may lie in its ability to attract newer money to science, such as those just starting their philanthropy, DAF donors (perhaps some of those riding the AI high), or smaller foundations — in other words, potential givers who may have money and strong scientific interests but lack the staff and infrastructure of a large foundation. New ways to increase support for research are particularly important now, amid Trump-era disruptions and cuts for scientific research. As IP has reported, philanthropic funders have instituted a number of measures intended to keep research moving forward, including through increased grant support and support for scientist training and education. We’ve also been covering other new approaches to funding science. Back in 2024, for example, I wrote about Renaissance Philanthropy, which creates “thesis-driven philanthropic funds” to drive investment and financial support in areas including climate, education, energy and more. Meanwhile, outside of science funding, similar efforts have arisen in the Trump 2.0 era to connect individual donors and lean funders with funding opportunities, like Project Resource Optimization (PRO), a platform set up by former USAID staffers to raise money quickly for vetted, life-saving global health work at risk of defunding. As the ranks of very wealthy potential donors continue to grow, we expect to see a proliferation of efforts like Catalyze, playing the role of informational intermediaries, and even of funding intermediaries outright. The legacy philanthropic foundations that have supported scientific and biomedical research over the decades have enabled great advances, and their work and support has been and continues to be greatly valuable. But that doesn’t mean new approaches shouldn’t be explored: If ideas like Catalyze can succeed in accelerating progress or bringing significant new dollars to science funding, that’s a big win for the research community and for society. Related Inside Philanthropy Resources: For Subscribers Only