Editor’s Note: This feature draws from a conversation with Priscilla Chan and Mark Zuckerberg, joined by Ben Horowitz, Erik Torenberg, and Vineeta Agarwala, exploring the groundbreaking work of the Chan Zuckerberg Initiative. The discussion highlights how the Initiative is creating computational tools to accelerate the prevention, treatment, and management of disease. Chan and Zuckerberg detail why basic science requires $100 million-scale projects beyond the scope of traditional NIH funding, how their Cell Atlas has become biology’s “missing periodic table” with millions of cells catalogued in open-source format, and how innovative virtual cell models enable scientists to test high-risk hypotheses in silico before committing to costly wet lab experiments.
In recent years, the Chan Zuckerberg Initiative (CZI) has emerged as one of the most ambitious philanthropic projects in the realm of biomedical science, driven by the shared vision of Priscilla Chan and Mark Zuckerberg. Their overarching goal—to cure, prevent, and manage all disease by the end of the century—is as audacious as it is visionary, reflecting both an extraordinary commitment to scientific progress and a sophisticated understanding of the technological tools required to achieve such a monumental task. As Mark Zuckerberg explains, “When we first set out that goal to cure and prevent disease by the end of the century, honestly, most scientists couldn’t look at us with a straight face. And they’re like, you’re crazy. Yes.” This statement encapsulates the skepticism the couple initially faced from the scientific community. Many researchers found the ambition almost laughably unrealistic, not because the goal itself was undesirable, but because the pathway to achieve it seemed obscure and unattainable through conventional scientific funding mechanisms. In response to this challenge, Chan and Zuckerberg adopted a strategy focused on accelerating the pace of basic science through the development of transformative tools rather than attempting to directly solve every disease themselves.
Priscilla Chan’s background as a pediatrician informs much of the ethos behind CZI’s approach. She reflects, “I trained as a pediatrician, and people always think like, oh, it must be about medicine. And for me, I went into medicine because I wanted to improve people’s lives. I wanted to make a difference. I wanted to be able to help others.” Chan’s clinical experiences exposed her to the profound limitations of existing medical knowledge and therapeutic approaches, particularly when confronted with rare or poorly understood conditions. “I met a lot of patients and frankly, little kids and families for which we just had no idea what the problem was,” she recounts. Often, these cases could be traced to a specific gene if one were lucky, but more commonly, families received generalized information with little actionable guidance. This reality illustrated to Chan the importance of basic scientific research as the foundational pipeline for hope and innovation. It is precisely this recognition of the gaps in understanding at the most fundamental level of biology that shaped the CZI’s strategic focus on tool creation rather than immediate therapeutic intervention.
Zuckerberg elaborates on this concept, drawing a parallel between scientific discovery and technological advancement: “Most major breakthroughs are basically preceded by the invention of a new tool to observe phenomena in a new way. Think about things like the microscope, being able to observe bacteria… or the telescope.” In other words, major leaps in understanding often depend less on incremental experimentation and more on the ability to view the world in fundamentally new ways. For CZI, the central hypothesis was that by enabling scientists with powerful new tools—particularly computational models powered by artificial intelligence—they could accelerate the pace of discovery across biology. This perspective reflects a shift in focus from traditional philanthropy, which often emphasizes funding individual researchers or incremental studies, toward a model that invests in infrastructure and platforms that can multiply scientific productivity exponentially.
Traditional funding models, particularly government grants administered by institutions like the National Institutes of Health (NIH), tend to favor smaller, incremental projects with relatively near-term goals. As Zuckerberg points out, “The vast majority of funding comes from the government and NIH grants. It’s parceled out into these relatively small grants that allow individual investigators to investigate usually pretty near-term things.” While such grants are essential for maintaining the day-to-day operations of laboratories and supporting incremental discovery, they are ill-suited for the development of large-scale, transformative tools—projects that may require hundreds of millions of dollars over a decade or more. In response, CZI has embraced a model that funds $100 million-scale projects, aiming to create foundational tools that can be widely shared with the scientific community. These include initiatives such as the Cell Atlas and virtual cell models, which serve as platforms for accelerating research rather than attempting to cure diseases directly.
The Cell Atlas, a cornerstone of CZI’s efforts, exemplifies this philosophy. Zuckerberg describes it as “biology’s missing periodic table,” a reference to the way it systematically catalogs millions of individual cells in a standardized, open-source format. The rationale behind this project is simple yet profound: to enable scientists to have a common reference for cellular diversity, allowing discoveries to be more easily contextualized and replicated. The process of building the Atlas began a decade ago with modest pilot studies, funding methodological standardization, and seeding initial labs. “When we were starting off, we didn’t even necessarily have in mind that we were going to use that to build virtual cell models. I think that’s sort of just come into focus as the AI work has advanced,” Zuckerberg notes, highlighting how foundational data infrastructure can give rise to new applications once technological capabilities evolve. The resulting network effect, where a shared annotation tool led to widespread standardization and a collaborative scientific community, mirrors the democratizing effect of the Internet: “Come for the annotation, stay for the virtual cell model,” Chan observes, underscoring the unanticipated yet transformative consequences of building open, interoperable platforms.
The emphasis on AI integration represents another layer of innovation. Chan and Zuckerberg have recognized that advances in computational modeling, particularly in large language models and reasoning systems, create unprecedented opportunities for simulating biological processes. These virtual cell models allow scientists to test high-risk hypotheses in silico before committing to costly and time-consuming wet lab experiments. As Zuckerberg explains, “Right now…you have to choose something that you think is going to have some likelihood of success to keep your lab career going. But if you had a virtual cell model…you could actually start testing it and tinkering on the computational side and ask riskier questions, things that would have been expensive and costly in terms of time and resources to do in the lab.” This approach fundamentally reconfigures the scientific method, shifting some of the risk and experimentation into the computational domain and thereby accelerating the potential for discovery.
The ambition to build virtual cell models also embodies the principle of hierarchical modeling in biology, where understanding smaller subcomponents informs larger system behaviors. Zuckerberg elaborates: “Our view is that you basically want to build up a state-of-the-art protein model and then have that be a part of the state-of-the-art cellular model. And then once you have that, you build things like the virtual immune system, which allows you to simulate much more complicated systems.” This stepwise construction mirrors approaches in engineering and artificial intelligence, emphasizing the layered accumulation of knowledge, from proteins to cells to tissues and ultimately entire physiological systems. By linking these hierarchical models, CZI aims to create a versatile platform that not only aids in basic research but also enables more precise, individualized approaches to medicine, reflecting the growing recognition that “most diseases should be thought of as rare diseases because each one of our biology is different,” as Chan remarks.
Personalization is indeed a critical component of CZI’s vision. Chan emphasizes that traditional treatment paradigms often rely on broad categorizations based on age, ancestry, or demographic markers, with little regard for individual variability. In contrast, the tools developed through CZI’s initiatives allow researchers to track the effects of specific genetic variants on cellular behavior and protein expression, creating the potential for highly targeted interventions. “If you look at the way we’ve been able to look at variants and look at single cell transcriptomics, we’re starting to be able to say, okay, this variant actually impacts this set of downstream cells. And then we start looking at the proteins that get expressed and how it looks similar or different to what a healthy cell would look like. Then you can start targeting, okay, like, let’s look at that as a target,” Chan explains. By enabling a deep understanding of individual biological pathways, the work of CZI promises to revolutionize the principles of precision medicine, moving away from one-size-fits-all treatments toward truly individualized therapies.
Strategic focus and long-term planning are also integral to the Initiative’s success. Unlike conventional grant-making, which often operates on short timelines, CZI’s approach emphasizes “10 to 15-year horizons,” chosen to align with both the pace of scientific discovery and the collaborative potential of research teams. As Zuckerberg states, “When we looked at the grand challenges on the 10 to 15-year time horizon, it needs to be like, when you look at it, you’re like, I see a path. Not everything needs to be solved for us to take it on.” This framework allows for the pursuit of ambitious yet achievable projects that can generate meaningful data and tools while maintaining credibility with the scientific community. The Biohub locations—San Francisco, Chicago, and New York—each reflect deliberate choices about specialization and collaboration: cellular engineering in New York, tissue-cell communications in Chicago, and deep imaging and transcriptomics in San Francisco. The Biohubs are intentionally positioned to combine cutting-edge research with collaboration across academic institutions, fostering an interdisciplinary approach unconstrained by traditional laboratory boundaries.
The integration of AI at the Biohubs further amplifies their impact. Chan notes, “We were already building tools to measure interesting data, building the data sets, but we didn’t really know what to do with them yet. And large language models coming onto the scene, we’re like, wow, we can make sense of all of this now.” By leveraging AI to interpret vast datasets and simulate complex cellular behavior, researchers can iterate hypotheses more rapidly and precisely than ever before. Specific AI models, such as the variant former and diffusion models, enable prediction of cellular responses to genetic modifications and the creation of synthetic cell types for simulation. Cryo models, capturing spatial relationships within tissues, add yet another layer of detail, allowing researchers to examine interactions in realistic, three-dimensional contexts. Collectively, these models constitute a new computational toolkit for biology, analogous to how the microscope and telescope historically transformed scientific observation.
The commitment to open-source principles has been central to CZI’s philosophy. By making tools and datasets widely available, the Initiative has fostered a collaborative scientific ecosystem. As Chan highlights, “We would be thrilled if people picked up the models that we build to be able to build the diagnostics, the therapeutics that need to come.” The Cell by Gene platform exemplifies this ethos, where a tool initially developed to solve annotation bottlenecks evolved into a widely adopted, standardized framework for single-cell analysis. This collaborative approach not only accelerates research but also amplifies the impact of CZI’s investment, as community adoption scales beyond what any single organization could achieve independently.
Moreover, the focus on reasoning within AI models reflects a nuanced understanding of scientific inquiry. Zuckerberg describes early efforts to create reasoning models over biology, noting, “You effectively have these models that simulate world models in different ways, and then you want it to be able to not just be able to spit out correlations… but actually be able to kind of reason through how things would evolve and why things would happen.” This emphasis on causality and prediction distinguishes CZI’s work from conventional AI applications, where pattern recognition is often prioritized over mechanistic understanding. By fostering models capable of reasoning about biological processes, the Initiative aims to create computational partners that can assist scientists in generating novel hypotheses and guiding experimental design.
Underlying all of these endeavors is a fundamental philosophy of risk and innovation. By creating tools that allow high-risk experiments to be simulated virtually, CZI seeks to overcome one of the intrinsic constraints of modern research: the conservatism induced by grant funding and the high cost of wet lab work. “Right now…you have to choose something that you think is going to have some likelihood of success…But if you had a virtual cell model…you could actually start testing it and tinkering on the computational side and ask riskier questions,” Zuckerberg observes. This approach not only accelerates discovery but also democratizes the research process, enabling a wider range of scientists and startups to pursue transformative ideas without the prohibitive financial risk that has traditionally constrained innovation.
In conclusion, the Chan Zuckerberg Initiative represents a novel paradigm in scientific philanthropy and research strategy, blending bold ambition, technological sophistication, and strategic long-term planning. By focusing on tool creation, open data, AI integration, and hierarchical modeling, Chan and Zuckerberg are constructing a framework designed to accelerate the pace of basic science, empower individual researchers, and ultimately transform our understanding and treatment of disease. As Chan summarizes, “We think a lot about understanding biology…And truly, each one of our biology is different…What should really happen is being able to precisely and accurately and quickly treat people by looking at individuals’ biology.” This vision, underpinned by methodical investment, technological innovation, and a commitment to open collaboration, positions CZI as a pioneering force in the quest to cure, prevent, and manage disease—a mission that, while historically inconceivable, may become attainable within this century due to their innovative approach.
