# Priscilla Chan and Mark Zuckerberg: Frontier AI + Virtual Biology To Solve All Diseases Latent Space · 2025-11-06 Priscilla Chan and Mark Zuckerberg, co-founders of CZI's Biohub, explain their ten-year shift from broad philanthropy to a focused mission of building frontier AI and virtual biology to cure all diseases. They argue that tool-building — from 12-foot microscopes to the 125-million-cell CELLxGENE atlas — is the essential, underfunded work that enables scientific breakthroughs. The couple details how their Biohub model combines frontier biology (e.g., spatial imaging, cellular engineering) with frontier AI (models like rBio and VariantFormer) to create a hierarchical virtual cell, eventually expanding to a virtual immune system. They emphasize that data generation must precede modeling, citing the decade-long Human Cell Atlas as foundational, and note that AI timelines may accelerate their 100-year goal significantly sooner. The episode closes with a call for biologists and engineers to collaborate, use their open models, and help generate data that grounds these next-generation tools. ## Questions this episode answers ### How is CZI shifting its strategy after 10 years, and why are they teaming up with Evolutionary Scale? Mark Zuckerberg and Priscilla Chan explain that after a decade, CZI will focus on science as the main philanthropic effort, specifically through the Biohub model. They are combining their AI and biology teams with the renowned Evolutionary Scale team, led by Alex Rives, to create an integrated frontier AI and frontier biology lab that builds virtual cell models and aims to accelerate scientific progress. [28:32](https://addtry.com/faf7c064-0156-4301-b233-429e213f6a51?t=1712000) ### How many cells has CZI's Human Cell Atlas cataloged, and what's next? Priscilla Chan says the Human Cell Atlas now includes 125 million RNA transcriptomes, with CZI contributing about 25% of the data and the rest from the ecosystem. They are now working on a billion-cell project, which is taking only months and at a fraction of the price, demonstrating the rapid acceleration in data generation for AI models. [14:34](https://addtry.com/faf7c064-0156-4301-b233-429e213f6a51?t=874000) ### What is the virtual immune system model, and how could it lead to new treatments? Priscilla Chan describes the virtual immune system as modeling individual immune cells to understand their interactions. This could enable engineered immune cells, like CAR T cells for cancer and custom cells that detect plaques in hearts. By understanding the immune system's balance, they aim to treat autoimmune diseases and use the body's own cells to maintain health. [45:17](https://addtry.com/faf7c064-0156-4301-b233-429e213f6a51?t=2717000) ### What is the timeline for curing all diseases, and what factors could speed it up? Mark Zuckerberg notes that while CZI's mission aims to cure or prevent all diseases by the end of the century, AI progress could shorten that significantly. He says the timeline depends more on AI advancements than pure biology. Priscilla Chan adds that big, collaborative efforts like the Cell Atlas, though not glamorous, are crucial to generate the data needed for these models. [48:28](https://addtry.com/faf7c064-0156-4301-b233-429e213f6a51?t=2908000) ## Key moments - **[0:00] Intro** - [2:18] Priscilla Chan: Philanthropy lacks a clear dashboard like companies, making it hard to gauge momentum. - [3:03] CZI's niche is bringing AI researchers and wet-lab scientists together with physicians, says Priscilla Chan. - **[4:05] Science vs Translation** - [4:31] Mark Zuckerberg: Gates Foundation focuses on public health translation, while CZI fills the gap in long-term tool development for basic science. - [5:47] Mark Zuckerberg: Major scientific advances often follow new tools, and CZI builds its own labs to develop those tools over 10-15 years. - [7:40] Mark Zuckerberg: 'We're not actually curing the diseases. We're just trying to build the tools.' - [8:06] Mark Zuckerberg: Biologists doubt curing all diseases by 2100, but AI researchers think that timeline is unambitious. - [9:05] Priscilla Chan: Connecting AI researchers with data collectors makes models better because data context matters. - **[10:45] Physical Proximity** - [10:45] Mark Zuckerberg: We are at the early stages of building virtual cell models, hierarchically from proteins to cells to immune system. - [11:23] Mark Zuckerberg: Physically co-locating biologists and engineers is an overlooked but fundamental part of accelerating interdisciplinary science. - **[13:55] Virtual Cell** - [14:20] Priscilla Chan: The Human Cell Atlas took 10 years and cost a lot, but the upcoming billion-cell project will take months at a fraction of the price. - **[15:51] Microscopes** - [15:51] Q: Are the custom microscopes still a significant bottleneck for imaging all cells? - [19:01] Mark Zuckerberg: AI models can interpolate between high-resolution dead-tissue imaging and live clinical scans to approximate molecular imaging in living organisms. - [20:01] Mark Zuckerberg: CZI uses transparent zebrafish for live imaging and applies AI to translate findings to human biology. - [21:03] Priscilla Chan: Until recently, scientists didn't know how many cell types exist in the human body, and we've only imaged a fraction, mostly healthy cells. - [22:19] Mark Zuckerberg: CZI's cell diffusion model generates synthetic cells from descriptions, but grounding them requires spatial and engineering data from the Biohub network. - [23:16] Mark Zuckerberg: CZI's frontier biology lab designs experiments to feed data into AI models, contrasting with AlphaFold's use of existing data. - **[23:18] Frontier Lab** - [25:12] Q: How does CZI create feedback loops between AI models and wet-lab experiments? - [26:36] Mark Zuckerberg: 'The idea that AI will automate all wet-lab experiments is the biological version of saying AI will automate all of society.' - **[27:25] Ambitious Research** - [27:25] Priscilla Chan: High wet-lab costs force scientists to pursue safe hypotheses; AI models can de-risk bolder, potentially transformative experiments. - **[28:45] Evolutionary Scale** - [28:47] The Evolutionary Scale team, led by Alex Rives, is joining CZI to lead the combined AI-biology program, bringing world-leading protein models. - **[30:15] Precision Medicine** - [30:23] Mark Zuckerberg: 'AI people are always in a hurry' — they expect frontier biological models to emerge much faster than a decade. - [30:41] Priscilla Chan: The ultimate measure of success is not better AI models, but their clinical impact — enabling precision medicine for individuals. - [31:35] Priscilla Chan: CZI's virtual cell models will address 'variants of unknown significance' by modeling their impact on disease pathways, transforming diagnostics. - [33:08] Priscilla Chan: Depression treatment is empirical with months-long trial-and-error; precision medicine would predict the right antidepressant per patient. - [34:59] Mark Zuckerberg: CZI plans to merge multiple virtual cell models into a 'biological omnimodel,' analogous to how language and vision models were unified. - [37:04] Priscilla Chan: In an AI-driven future, doctors will focus on care and compassion, using models for diagnosis and treatment, while being data inputs themselves. - **[45:17] Immune System** - [45:17] Priscilla Chan: CZI's virtual immune system models immune cell interactions, enabling engineered cells to detect and treat cancer and heart disease. - **[48:27] Timeline** - [48:28] Mark Zuckerberg: The timeline for curing all diseases hinges on AI progress; strong AI could enable virtual cells and precision treatments far sooner than 100 years. - [49:52] Priscilla Chan: The Cell Atlas work was not glamorous or tenure-track material, but big data generation is essential for training powerful biological models. - **[52:43] Closing** - [52:52] Mark Zuckerberg: 'Check out the models' — CZI invites biologists and engineers to test their early AI biology models and provide feedback. ## Speakers - **Alessio** (host) - **Swyx** (host) - **Mark Zuckerberg** (guest) - **Priscilla Chan** (guest) ## Topics Biology, Healthcare, World Models ## Mentioned BioHub (company), CZI (company), DeepMind (company), Evolutionary Scale (company), Gates Foundation (company), AlphaFold (product), CellxGene (product), ESM-3 (product), cryo model (product), rBio (product) ## Transcript ### Intro **Alessio** [0:05] Hey, everyone. Welcome to the Latent Space Podcast. This is Alessio, founder of Kernel Labs, and I'm joined by Swyx, editor of Latent Space. **Swyx** [0:11] Hello. We're so delighted to be in the Imaging Institute of CZI with literally C and Z. Uh, welcome Mark and Priscilla. **Priscilla Chan** [0:18] Thanks for having us. Thanks for getting nerdy. **Alessio** [0:20] Yes. **Mark Zuckerberg** [0:20] Yeah, we're excited to do this. **Swyx** [0:22] We, we so don't often get to see this side of you, and, uh, so thank you for taking some time, uh, out to talk about this. And, and it's like, you know, sort of the 10-year anniversary kind of, of, of CZI. So I just wanted to introduce people if, if, you know, people have not been caught up. Um, one of the interesting things that we found out just from talking to your teams is there's an interesting difference between how you guys started CZI and the Gates Foundation. Uh, and I, I heard that Bill Gates is a mentor of yours. So maybe you could tell that story on, of like deciding to start CZI and deciding to pursue basic science instead of translational work. **Mark Zuckerberg** [0:56] Well, I mean, I think one of the core things for us with CZI was just getting started earlier, right? We got some advice that basically philanthropy and doing science, just like any other discipline, requires practice, and you're not gonna be good at it overnight. So, um, so we should just kinda dig in and start doing a, a few different iterations on it and see what we enjoy and where we think we can have an impact and go from there. So, um, yeah, I mean, like you mentioned, I mean, this is-- we're coming up on, in November, the 10-year anniversary of when we started CZI. And, you know, there's a lot of work that we've-- we're really proud of that we've been a part of inc-including, you know, work in education and supporting communities. But, you know, when we reflect on it, we feel like the work that we've done in science really has had the biggest impact and in a lot of ways is accelerating. And with, um, especially with all the advances in AI that are coming, I think the ability to have an even bigger impact over the, the coming decade is, um... It, it just seems really clear, like this is coming into focus. So, you know, for the next period, we really wanna make science the main focus of, of what we're doing, and, and specifically the Biohub organization, that we're really proud of this model that we've helped pioneer that we can go into detail on, is really gonna be like the main focus of our philanthropy, and it's just something that we're very excited about. **Priscilla Chan** [2:18] Yeah. When we started 10 years ago, we had this idea like, okay, I bring experience as a physician, Mark's an engineer, and he builds things, and we have an, an opportunity to give back resources to make an impact on this world. And we sort of just-- we tried a bunch of things. And the thing that in running a philanthropy I'm incredibly envious of people who run companies is that, like, you guys can have a dashboard, and there's like financial results, and people tell you if you're on the right track, on the wrong track, and there's clarity. But in philanthropy, there's so much you can do, and it takes a long time for you to get a sense of like, what has momentum? What are we doing that is actually bringing all of our both skills and resources to maximal impact? So over the past 10 years, I would say we've been getting a sense of what is that thing that really allows us to have the impact and makes the most of what we bring to the table, and it, it's really been around AI and biology, where we're like, "Oh my gosh, this is it." And, you know, the ecosystem is big. And we really think our ability to bring great scientists, great AI researchers together between the wet lab and the compute, the ability to bring physicians and patients into the picture, that's a unique niche for us at the Biohub. And, um, that's-- we, you know, we need others to take the work to translation. The Gates Foundation has a strong focus on translation and the field, and we have had a number of really awesome collaborations and continue to, where we really look at sort of the basic fund- fundamental research and being able to partner with someone who's thinking about the translation layer is incredible. ### Science vs Translation **Alessio** [4:06] We kinda see the first decade, and I would love to get your take as a decade of creating data. **Mark Zuckerberg** [4:12] Mm-hmm. **Alessio** [4:12] Creating a science ecosystem and then starting to work on some of the models, and the next decade maybe is more of the applied modeling side. At what point did you decide that just doing the tooling was better versus, you know, you could have cured malaria in Africa too or some other disease, you know? But- **Mark Zuckerberg** [4:31] Yeah. Yeah, I mean, take a step back, and this is kind of related to your, your first question too. I mean, there's-- like Priscilla was saying, the space is huge. You know, there are lots of other philanthropies, including Gates, who I think they would say that they're primarily focused on public health and sort of administering, uh, like once you know what a, what a cure is, just getting it out to the world is a huge thing too, and someone needs to do that, and that's a lot of work and a lot of resources, and it's good that, that they're doing that. Basic science is another completely different part of the kind of innovation funnel to enable that. And our view is that the federal government basically dwarfs everyone else in terms of how much they invest through NIH. But there's a certain pattern to how they invest, which is really enabling a lot of individual investigators to do work. And our kind of observation was that if you look at the history of science, a lot of major advances are basically preceded by new tools or new ways of observing things, right? So the initial telescope allowed a lot of advances in astronomy. The microscope, the invention of that allowed a lot of understanding of biology. And similarly, I think we're at a point in history where a lot of new tools are being built, computational tools, tools to instrument the body in different ways and, and understand things. And often, those tool development It just takes a longer-term timeframe and a sometimes a larger commitment of capital, including the way to do it isn't necessarily just to make grants to a lot of different people. You need to really operate it yourself, which I think is one thing that's different about the way that we've operated than, than others, is, you know, most times when you think about philanthropy, you think about kind of giving money away in terms of grants, and a lot of what we're doing is actually building up these institutes and, and kind of building labs to do that kind of research our- ourselves by bringing in leading scientists and engineers and, and, and all that. But that's kind of the strategy is, um, we feel like there's a lot of new tools to develop. There's sort of been a hole in the ecosystem where tool development and kind of the 10 to 15-year runway that you need to do that, and often hundreds of millions of dollars to build things like the microscopes and that you're, and imaging that you're seeing in this institute here, I think that that's been sort of underfunded, and that's where we think that if we, we do that kind of work, it can just give all these other scientists way more tools to accelerate the pace of, of, of research, hopefully discover cures, and then you have folks who are focused on public health who bring that out to the world and kind of deploy it to everyone. **Priscilla Chan** [7:01] Yeah, I mean, our mission is to cure, prevent all diseases and, like, that's not gonna happen just in our four walls. So the strategy has to be how do we make every single scientist and everyone better and more effective? And, you know, the strategy Mark talked about is sort of where we landed on how to actually maximally move the field forward. **Mark Zuckerberg** [7:23] Yeah. **Swyx** [7:23] Yeah. The mission is cure, prevent all diseases. By, by the way, a lot of people outside of the CZI worlds are still kind of find this concept very alien, but talking to the CZI people, they really truly believe it, and, uh, it's, it's im- impressive how you pick the right mission to motivate everyone to work towards this enormous task. **Mark Zuckerberg** [7:40] Well, it's kind of a funny thing. I mean, it's, uh, I mean, I, I kind of... We like to talk about the mission as, like, helping scientists do it, right? 'Cause we're not actually- **Swyx** [7:47] Correct **Mark Zuckerberg** [7:47] ... curing the diseases. We're just trying to build the tools- **Swyx** [7:49] Tools **Mark Zuckerberg** [7:49] ... advancing the- **Swyx** [7:49] Data- **Mark Zuckerberg** [7:50] Yeah **Swyx** [7:50] ... models. **Priscilla Chan** [7:51] Mm-hmm. **Mark Zuckerberg** [7:51] Yeah, like, basically accelerating scientists work towards that. But, you know, a funny thing about it is we had this initial timeframe of by the end of the century, and, you know, when you ask biologists, there's a lot of questions around, okay, that's really ambitious. Are we gonna be able to do that? And then when you ask AI people, um- **Swyx** [8:10] They're like, "Yeah, yeah." **Mark Zuckerberg** [8:11] ... it's like, "That should be really easy. Like, why are you so unambitious that you're- ... you know, shooting for just the end of the, the century?" And I, I do think that at the pace that, that AI is improving things, I mean, I think it might be possible significantly sooner than that. I mean, I don't think it's, it's necessarily worth putting a, a number on it or a date, but I think that that's, you know, to your point about the first decade was, you know, sort of about doing work like the Cell Atlas to, to be able to help understand basically all of the, the kind of specifics and data about all the different configurations of every cell in the body. When we did that, we kind of had this vague notion that that would be useful to advance science, but I think that, you know, like a lot of people in the tech industry, we, we have even been impressed by how quickly AI has accelerated. But that ended up being a really valuable thing to have done over the last 10 years, especially for where AI is now and now the models that can get built with that. **Priscilla Chan** [9:05] But the thing that's interesting, don't you agree, is like, okay, so from a te- I totally agree that in our intersection of AI and biology, the AI folks are like, "Yep." The biologists are like, "Hmm." Um, and I think it's actually that confluence of conversations that lead both the biologists to be like, "Okay, I'm really uncomfortable about this idea and timeline, but if I'm really pinned down to think about it, what are..." Like, you really force people to think through like, okay, what are actually the barriers? What would you need to do? And you're forcing that conversation from the biologist side and from the AI side, really getting a sense of, okay, what-- Like, data is not just data. You guys know this. Like, you need to know sort of how the data was collected and from where and being able to connect the AI researchers to the folks who are actually gathering the data on a daily basis makes their work better. And so it's, it's that conversation that's happening here that, um, I think makes people outside so excited about this 'cause it's credible, and they sort of have worked, really dug in and thought through how to, how that would work and, um, and they're excited, and they believe, and believing is the first step. **Swyx** [10:20] Believing is the first step. **Mark Zuckerberg** [10:21] Yeah. **Swyx** [10:21] There, there's a general pattern of software eating the world, and I think AI eating the world is kind of like the next version of this. I was talking with Garrett outside who, uh, says he's a biologist, but, you know, I, I think he's using models like SAM from, from Meta. **Priscilla Chan** [10:32] You're like, "You don't look like only a biologist." **Mark Zuckerberg** [10:35] What does a biologist look like? **Priscilla Chan** [10:37] Like, no, I was- ... he said, like, he's working on models out there, and that's like biologists are working using models, right? **Mark Zuckerberg** [10:42] Yeah, yeah. **Priscilla Chan** [10:42] They're not just, like, in imagination, like, just using in the wet lab. Um- ### Physical Proximity **Swyx** [10:45] Yeah, totally. Yeah, I think, uh, one of those things that, you know, referencing the wet lab is all of the, the key approaches that you're pursuing is turning things in, uh, pursuing the virtual cell, turning things from mostly wet lab into something in silico. How far along are we? **Mark Zuckerberg** [11:01] I mean, it's pretty early, right? I mean, I think the, the first step, which I think is easy to overlook, is basically what Priscilla was talking about of just getting these folks together. It, it almost-- It's worth taking a beat just to talk about this just because I think most people assume that this is like obviously you would go do that, but it's somewhat novel in science because of, I think, the way that a lot of funding has been done. That is, basically you grant individual teams small, like, relatively small grants, and people do a lot of science independently. It is, I think, pretty amazing how much progress you can make if you just have people from different disciplines sit together, right? It's, I mean, this is like over my career, I mean, both at Meta and, and here. It's, um, it's like you have, you have teams that, like, are not working together for some reason, or they disagree on something. It's like, okay, physically just have them next to each other, and it, like, actually is super helpful. So- Here, what, what are we doing? It's not just bringing together the biologists and the engineers, which was a core part of the initial Biohub model, but it was also unlocking the ability for people to work together across institutions. So the first Biohub that we started out here between Stanford, UCSF, and, and Berkeley allowed a lot more collaboration between scientists and engineers at those universities than was in practice happening before. And it's like you can look at this and be like, "All right, that seems really obvious," but it actually was sort of an interesting and novel experiment, and one that I'm really happy to see others also implementing, because I think it's just such a clear win, um, just the kind of the human side of bringing people together and having them sit together. So anyway, that I, I would say is kind of step one or step zero, and is probably quite overlooked, but is sort of a fundamental part of the model that I, I guess also goes back to this idea of like we're not just kind of like granting funds to other people, we're building an institution and we're having people sit together. So then you get that, and then you get these people who are like half biologist, half AI engineer because they, they kind of have some experience doing it. And I mean, I don't know. I mean, we can, we can talk through the specific models and, and there's a lot of exciting stuff there, but I'd say it's, um, it's an early glimpse of where this is all going. I think, like, you wanna kind of build up these models hierarchically, so you give them a lot of data about specific proteins, and they can model specific proteins in the cells, and then you can model different cell behavior, and then eventually you get-- you kind of zoom out and you're modeling like a virtual immune system or something like that. And it's sort of hard to simulate the immune system without having a good understanding of how a cell might work, and it's kind of hard to understand or simulate how a cell might work if you don't really understand how the proteins interact. So you kind of need systems that understand data at all different levels of this, and then you kind of pull them together. And then if you look at the different models, there's, you know, there are versions that are kind of focused on, all right, like which parts of the genome are, are kind of being expressed in different ways. ### Virtual Cell **Mark Zuckerberg** [13:59] I mean, the cryo model that I think is very interesting that's built off of the data here, the only model that I'm aware of that's like a, a spatial model of like, of basically like how, how these cells work. And, and you kind of, you just wanna be able to look at stuff from different perspectives and then put them together, and you build like a richer and richer model of, of kind of how these cells work. But we are definitely at the beginning of this journey. **Priscilla Chan** [14:20] But it's like slow and fast, slow and fast, right? So when we built the Human Cell Atlas, we started ten years ago. It was one of our far-- first RFAs, and we actually f-- the first RFA was to fund the methodologies of how you would get a single-cell transcriptome. And it took us about 10 years to get to a place where we had, uh, we now have one of the largest, uh, corpus of tr-- uh, RNA transcriptomes, a hundred and twenty-five million cells. Cost a lot of money. And the really cool thing we discovered through that process was if we could seed the effort and make it easy for people to contribute, it happened. That's CELLxGENE. We actually-- We're responsible for maybe twenty-five percent of the data, and the rest of the ecosystem contributed seventy-five percent of that. That's an incredible asset and has been very important in modeling work. Similarly, if you look at AlphaFold, they, they, they built off publicly available data that was collected for thirty years prior, right? So that takes a long time. But now we're doing the billion cell project, and that is taking months and at a fraction of the price. You know, really slow to fast, but it's a single dimension, and cells are so complicated. And here we're looking, like Mark said, at the three-dimensional imaging structures. That's an-- And it's slow and expensive. But with cryo-- with the cryo model, it will get fast again. And you just have to repeat it, and so I think we'll get growth spurts, but a- it's all happening just faster and faster. ### Microscopes **Alessio** [15:51] How do you think about the layers? So you have compute, and we'll talk about that later. On the data side, you build these amazing microscopes. I learned that they're all built for you by spec. They're not off-the-shelf things that anybody- **Mark Zuckerberg** [16:03] These are design partners. **Alessio** [16:04] Yeah. **Priscilla Chan** [16:05] Yeah. **Alessio** [16:05] How, how much of a bottleneck is that still? Like, can we convert the world of atoms into bits now at the right precision- **Mark Zuckerberg** [16:14] Some acceptable rate **Alessio** [16:15] ... or do we need- **Mark Zuckerberg** [16:16] Yeah **Alessio** [16:16] ... do we need more work on the microscopes themselves too? **Mark Zuckerberg** [16:18] I mean, you're never done. **Alessio** [16:19] Right. Yeah. **Priscilla Chan** [16:20] Well, speed, uh, for here, speed has been a big question of how just getting the process through. So here we've worked on sort of the speed at which we can look at tomograms and the sort of c- contrast and resolution, and that's where the laser phase plate comes in, so to be able to make the data better and faster to get the data. Um, but it's a bottleneck in so much as there's only a f-- I, I, I don't know the exact number. There are like maybe tens of these microscopes in the world, so that's one bottleneck. And I think really is, uh, like when I was saying it's slow and then fast, there's so many other dimensions that we don't have yet of like the cool thing here is with s- the transcriptome work, we're looking at cellular expression, and with the imaging work, you're being-- you're able to localize it in space, and now you wanna connect those two. But that's still like two dimensions connected. Time is another comm-- dimension. We need to get dynamic imaging in place. Um- **Mark Zuckerberg** [17:17] Oh, God. **Alessio** [17:18] That's so much- **Priscilla Chan** [17:19] I know **Alessio** [17:19] ... resolution. Yeah. **Priscilla Chan** [17:20] Right? But like really cool biological innovation. B- We need innovation in the way we can look at things, like stain free, dye-free, so we can look at things without sort of human intervention with time as a dimension is another-- 'cause, like, we are not frozen slices. Um, so I think it's just continuously looking at what the next dimension we wanna sort of be able to either understand deeply or connect to our existing corpus of data and knowledge. **Mark Zuckerberg** [17:50] And obviously the, the ideal would be you wanna increasingly be able to image things inside living cells, right? So I mean, you can kind of, you can simulate it a bit by, okay, you can take a cell out or, or- Some culture and it's- **Alessio** [18:03] It's all destructive, yeah **Mark Zuckerberg** [18:04] ... it's like, okay, it's living for a little bit or something. But I mean, you really wanna be able to kind of as much as possible actually understand what's going on in living organisms. **Alessio** [18:12] Can that be done? Is there-- What, what, what are the approaches? **Mark Zuckerberg** [18:14] Well, the better it gets. **Priscilla Chan** [18:15] Well, there's this cool methodology. So there is a really high-intensity X-ray methodology you can use. The, the organ has to be dead. So, like, you can just shoot X-rays, high-intensity X-rays at, like, a lung and understand at, like, a sort of molecular level how the lung has assembled. And then you can correlate that with living imagery, right? MRIs of the lungs, CTs of the lungs, and look at the associations between the living images in real patients with the sample that you put- **Mark Zuckerberg** [18:48] Yeah **Priscilla Chan** [18:48] ... into the high-intensity X-ray. So that's another example of, like, correlating data types so that we can get that sort of high-level specificity with clinical data that impacts humans. **Mark Zuckerberg** [19:01] But I mean, at some level, that's sort of the point about building these AI biological models, is you can have a lot of data, and you can interpolate that on that space- **Priscilla Chan** [19:11] Yes, that's, yes **Mark Zuckerberg** [19:11] ... and understand that. **Priscilla Chan** [19:11] Totally. **Mark Zuckerberg** [19:12] And then there's, you know, so one of the models that, a-again, I mean, this is, it's really early work, um, but the, the rBio model, the idea of doing reasoning is that then you don't just get correlation, but you get some understanding of, like, logic over how these things get together too. So yeah, I mean, I, I think it's probably gonna be a while, and people don't have great hypotheses on how you'd actually do, like, molecular imaging, like of a cell deep inside a living organism. But the goal is to be able to approximate that as much as possible with, like, this kind of surround view of, of, of, of different things that you can image. **Priscilla Chan** [19:49] You guys like to see cool stuff. It's not here, but at our San Francisco site, we do image see-through fish called zebrafish. **Alessio** [19:58] Zebrafish, yeah. **Mark Zuckerberg** [19:58] That's another-- It's another good example of, like, it's another good hack. **Alessio** [20:01] We walk through so many different models here. **Mark Zuckerberg** [20:01] It's like, all right, it's like how, what's a good way to- **Priscilla Chan** [20:03] Yeah **Mark Zuckerberg** [20:03] ... imaging a living thing? It's like, take a see-through thing. **Priscilla Chan** [20:06] Take a see-through thing. And then use a model to say, how does this see-through thing actually relate to us, right? Like, I'm, like, not that interested in curing disease, cure, prevent, manage all disease for zebrafish. I am very interested- **Mark Zuckerberg** [20:19] I might speak for yourself. **Alessio** [20:20] For zebrafish. **Mark Zuckerberg** [20:21] Yeah. **Priscilla Chan** [20:22] Mark, Mark's pro-zebrafish. I'm okay on zebrafish. But you, you need to u- Another application of large language models is looking at how, what is conserved and what is actually relevant and important to the way human biology works in a fish model. And so being able to have that translation be m-more effective so we don't waste our time on things that won't apply in a model organism is another really interesting way to elevate biology. **Alessio** [20:49] On the data side, can you just give a overview of how far we are? Like, what percentage of all cells have we imaged and do we have-- What's the distribution of them? You know, like when you say hundred fifty million to one billion cells, is that a lot? Is that, uh, ten percent? **Priscilla Chan** [21:06] The funny thing is, until recently, we didn't know how many cell types- **Mark Zuckerberg** [21:09] Yeah **Priscilla Chan** [21:09] ... the human body had. **Mark Zuckerberg** [21:10] I mean, this is kind of a wild thing. I mean, this was a big part of the Cell Atlas project- **Priscilla Chan** [21:12] Yeah **Mark Zuckerberg** [21:12] ... is like we-- there wasn't even... It's kind of like imagine the periodic table in chemistry, but you, you know- **Priscilla Chan** [21:17] You just don't know **Mark Zuckerberg** [21:17] ... it's like- **Alessio** [21:17] It doesn't end. **Priscilla Chan** [21:18] It-- Well, it's what I know- **Alessio** [21:20] You don't have the squares. **Priscilla Chan** [21:20] We know it's billions. We know there are billions of cell types in a human, and we've only truly looked at a fraction of them, and we looked at it in largely healthy cells. And so, like, just the number of permutations of, like, age, well, species, 'cause not all research is in humans, right? So species, ancestries, like what is your sort of genetic background, age, like babies are different than old people, gender. All of those things actually are permutations, environmental exposures. All of those things are permutations on the cell that actually you, you wanna be able to understand in healthy and diseased states. I feel confident that we are at the beginning of this. **Alessio** [22:03] I'll ask a little bit of, um, obvious question in terms of the intersection of AI and, and bio, which is, don't we want precision in biology? Don't we want, uh, some grounding in a world model maybe that we don't normally get in a language model? **Mark Zuckerberg** [22:19] Yeah, I mean, I, I think that that's sort of the, the point of doing all the measurement and being able to have all this real... You know, and it's like, so you have the, um, the diffusion model for generating cells that we, that we put out. I mean, it's like one of the, one of the recent models. And it's like, it's cool because you can basically, you have a model now that you can describe, you know, like, the conditions, and it'll basically give you a synthetic cell. But yeah, you, you want it to be increasingly grounded, and that's a lot of the point of the biology and the engineering that we're doing, is to be able to have these different facets of that. So the Imaging Institute is one part that gets you the spatial data that's, that's very helpful. And the work that, uh, we're doing in the other Biohubs on cellular engineering and instrumenting inflammation and things like that, it's basically, it's scientific work to build new types of tools that allow us to measure new types of things that generate data that allow us to ground the models in different ways. One framing that we have on this that I, I think is, is pretty interesting is that, you know, there's this concept of a frontier AI lab that is like, okay, it's, it's building AI models that are sort of at the frontier of what's possible. ### Frontier Lab **Mark Zuckerberg** [23:31] And I think you can think about biology in that way too, and there's sort of a concept of a frontier biology lab. Like, what is the idea of a, you know, it's, uh, like labs that are kind of at the cutting edge of, like, building the most advanced imaging, like measuring, you know, inflammation or doing cellular engineering in the most advanced ways. What-whatever the, the problem space is that you're at. And then I think that there's this interesting problem space of what happens if you're at the intersection of those two areas, right? So I mean, you mentioned, um, the work that DeepMind did on AlphaFold, which was great. Um, that's an example of a frontier AI lab- Using a data set that was just generated by other scientists, like over decades, right? But I think part of what we're trying to unlock here with Biohub is the idea of what actually-- what happens if you do frontier biology and frontier AI in sync together, and you're designing the tools on the frontier biology side in order to specifically collect and be able to learn types of data that you then want to feed into specific types of models that you wanna build so that it can understand the, the cells and the body at different types of resolution. I think you can just kind of, um, I don't know, you, you-- It's like a much more integrated approach that, that allows, you know, designing the things that you need that, that should eventually get towards more grounding and not just allowing, you know, folks who are good at AI to do the best they can with whatever biological data happens to be available. **Alessio** [24:57] What's the hill climbing in this scenario? So like with language models, you have benchmarks, you look at the benchmark, you just make that go better. With these things, you have to bring it back to the real world. So as you build these models, like how do you bring the two teams together to get feedback? **Priscilla Chan** [25:12] I, I think it's very similar to what Mark just said. You wanna be able to va-validate on the accuracy question. We don't expect that these models... Th-they will get increasingly accurate, but you wanna be able to have feedback, and it's not as easy as being like, you know, "This, yeah, this output doesn't make sense." You have to actually take it to the wet lab, run the experiment, find out if it actually happened as predicted, and feed it back into the model, and that's the virtuous cycle we wanna build to help the AI best serve the biologists and the biologists be part of continuously improving the models. **Alessio** [25:48] From like a numbers perspective, in a language model, you can run tens of thousands of tests. **Mark Zuckerberg** [25:53] Yeah. And they're very false. Very false. **Alessio** [25:55] Yeah. **Mark Zuckerberg** [25:55] I mean, and we have to build a lot of them out. Yeah. **Alessio** [25:57] Yeah. And then on going to the wet lab, what do you think that's gonna be like the feedback cycle? Like, as you start to have more of these things to be tested in the wet lab, do you feel like that's gonna be a bottleneck that like we cannot take that many or? **Priscilla Chan** [26:09] Um, I don't know the answer to that yet. I think the, the, the throughput on sort of established metrics in the wet lab is actually getting quite fast. You can run-- parallelize a lot of experimentation. Um, so, uh, it-- but it's not at the te- easily at the tens of thousands of verifications. But it's-- We'll have to s-- Well, we actually have to see. We'll probably need to be smart about how we do it. **Mark Zuckerberg** [26:36] But I mean, there's, you know, a lot of people I think often take these things to the extreme and are like, "Okay, pretty soon if you have these models, you're just gonna be able to run experiments with the models without even having to go to a wet lab." And it's like- **Priscilla Chan** [26:47] Yeah. **Mark Zuckerberg** [26:48] No, I mean, I think that- ... that's kind of like-- I think that that's sort of the biological version of like eventually AI is gonna automate every single thing in society. It's like, look, it may be you get there, right? And I think that there's like some chance over time. But well before you do, you're going to be able to have models that can help generate hypotheses, and scientists can apply their taste on which ideas or kind of suggestions come from this are worth testing, and then you test them, and then you feed it back into the model, which I think is basically the way that every AI model is deployed into- **Alessio** [27:23] Even in coding **Mark Zuckerberg** [27:23] ... work in other places. **Alessio** [27:24] Yeah. **Priscilla Chan** [27:24] Totally. **Mark Zuckerberg** [27:24] Yeah. **Priscilla Chan** [27:25] Like you-- Right now, because the wet lab is so expensive and relatively slow compared to sort of computational experimentation, like people are choosing like, "I need something to hit." So people are going for hypotheses or ideas that are like, you know, uh, to use a sports analogy, like singles or doubles. ### Ambitious Research **Priscilla Chan** [27:45] And but like they-- It's just too risky. They only have so much grant funding, and they need something to help move their work along. But like if we have a model that can help de-risk some of the bigger, riskier ideas, that's gonna move science faster. Um, and, uh, I think makes the scient... and, and those ideas both, you know, can be sourced with AI as a tool, but really, it's really about making the scientist less hesitant to explore big ideas. **Alessio** [28:13] Yeah. Obviously, that's a lot of the success of the model CZI, which is, uh, serving this part of research that is underserved because there was basically no benefactor or no fun- no funding mechanism, uh, by which to do this. One thing that we're announcing when we release this podcast is this unification of the sort of Biohub model. **Priscilla Chan** [28:31] Mm-hmm. **Mark Zuckerberg** [28:32] Yeah. **Alessio** [28:32] Um, I think it's very analogous to the foundation model and frontier lab approach, right, where you bring together people disci- different disciplines. You have much longer time horizons than a- than anyone else. Uh, are there any other key elements to the strategy of the Biohub that you're taking? ### Evolutionary Scale **Mark Zuckerberg** [28:47] Well, I mean, one thing that we haven't talked about is the Evolutionary Scale team and Alex Reeves and, and his team joining, and they're like- **Alessio** [28:53] Well, let's, let's talk about the announcement. Yeah. **Mark Zuckerberg** [28:54] Yes. I mean, I mean, this is like probably the most talented team working on AI and biology, right? And, and like at the intersection of doing of like basically good biology background and also, you know, they've just been working on- **Alessio** [29:08] ESM-3 **Mark Zuckerberg** [29:09] ... yes, some of the top protein models for a long period of time. Yeah, I mean, I, I think if you wanna build a, an organization that is doing frontier biology and frontier AI, you need to have like world-leading AI researchers, and we're doing that by basically combining the team that we have that's already put out all the models that we're talking about today, plus having the Evolutionary Scale team, which is just like very renowned, um, join, and Alex is basically going to be running the program. So I think it's, it's sort of an interesting decision, I, I think, to have the AI person basically be running the, the overall program, partnering with these leading, leading biologists, I think gives a sense of how optimistic we are about the AI work being very fundamental to this. But we're very serious about building out like a leading part, a leading lab on the AI side as well. That goes for both the talent and the compute. I think we were probably the first to build out a large-scale compute cluster for, um, for biological research. I think now, um, there are some others who are doing it too, but we're also building on that and, you know, we really-- we plan to release frontier models on this. ### Precision Medicine **Alessio** [30:16] Do you see that as the 10-year output? Like, in the next 10 years, we look back at that and then- **Priscilla Chan** [30:21] I just talked to them yester-yesterday. They say it's faster than that. **Mark Zuckerberg** [30:23] But AI people are, are, uh- **Priscilla Chan** [30:24] They're always- **Alessio** [30:25] Yeah. **Mark Zuckerberg** [30:25] Yeah. **Priscilla Chan** [30:26] They're always in a hurry. **Alessio** [30:27] We have AGI in two years, so right. Would that be a satisfactory result for you guys? You fast-forward 10 years, you have, like, the best, you know, the three best models in, in biology, or is there, like, a further goal that you wanna have as an output of the foundation? **Priscilla Chan** [30:41] I have to bring it back to the patient. Uh, I think, like, the AI models are-- I think we will be very excited both if we have great models and scientists are using them, but you really wanna make sure that it's, like, accelerating clinical impact. Like, that's the goal, right? Like, um, that the AI models is a very challenging milestone that we've worked-- we are working very hard on, and we will get there. But how do you actually take those models and apply them to actually change the way people live? And, uh, there's so-- there's two va-variants that I think about in the application of these models. Why are they important? One is, like, each one of our genetics is incredibly diverse and different. Uh, first, like, first of all, we are just all-- f-the four of us are unique people, but we also have things, like, that are sort of known indicators of disease and unknown indicators of disease. And I actually find the variants of unknown significance to be the most es-interesting and the most frustrating. Say s- you know, someone that you love, it's sort of a diagnostic mystery. They need to go in and look at the genetics. Most likely, they'll come back and be like, "There are these three things that are not usual, but we also don't know why." And you're like, "Okay," like, "Should I panic? Should I not panic?" **Mark Zuckerberg** [31:56] Yeah. **Priscilla Chan** [31:56] "Like, what do I do now?" And what you really wanna do, and I think these models will be able to do, is look at those variants and actually model out what is the impact in the different cells, how it influences cellular behavior, and whether or not that is tied to a pathway to disease or not. Like, that's a big deal, and I think we should be doing that. That is actually the future of medicine, where we think about each one of your biology based on your genetics, your exposure, and how that predisposes you or not to disease. Like, that's huge, and we wanna be able to see that clinical application, but we can't. It's too expensive, too hard to model each person, i-impossible to model each person in a lab. But if we can build models around this, it is possible, and then we can start thinking with extreme precision. And I'm just-- not just talking about rare disease. Like, there are, like, common diseases. I'll just say depression. Right now, it's empirical, right? We just say, like, "You're depressed. Like, here, let's try this antidepressant." And it's, like, usually the one that the pat-- the doctor's more familiar with or maybe one that you've heard of, but, like-- and then you have to try it for months before it's like, did it work? Did it not work? **Mark Zuckerberg** [33:09] Months? **Priscilla Chan** [33:10] Yes. **Mark Zuckerberg** [33:10] That's the c-cycle? I, I don't have familiarity- **Priscilla Chan** [33:13] That is the cycle **Mark Zuckerberg** [33:13] ... with this. It's, it's horrible. **Priscilla Chan** [33:14] And, and meanwhile, if it doesn't work, it means the person's suffering, and this applies to, like, almost every disease, right? There has to be some biological explanation as to why some medications work and don't. So can we actually then look at each patient and say, "Based on who you are, we think this medication is gonna work best for you"? That's the future I wanna live in, where we can actually understand individuals in-- as individuals and use the biology and science very directly to keep them well. **Mark Zuckerberg** [33:45] Yeah. So, like, i-if there's a name for this tool that has the clinical impact- **Priscilla Chan** [33:50] Mm-hmm **Mark Zuckerberg** [33:51] ... that is o-on the scale of the electron, how do you envision it? I, I guess, like, um, I, I feel like it's almost gonna be, uh, the CZI app, I guess. **Priscilla Chan** [34:00] Oh, well, it won't be j- uh, first of all, that's not what we're building right now. We're building the basics. We're understanding, like- **Mark Zuckerberg** [34:06] Yeah, yeah **Priscilla Chan** [34:06] ... cells and molecules. Um, so I'm painting a s- **Mark Zuckerberg** [34:09] Someone else will do it. **Priscilla Chan** [34:10] Uh, we're painting a picture, like, we need partnerships. This is- **Mark Zuckerberg** [34:13] Yeah, yeah **Priscilla Chan** [34:13] ... you asked about the ecosystem before. **Mark Zuckerberg** [34:14] Yeah, yeah. **Priscilla Chan** [34:15] Like, there are experts along the way, uh, of this pathway, and so we sort of are at the fundamental research side. **Mark Zuckerberg** [34:21] Yeah, yeah. **Priscilla Chan** [34:21] And you need to be able to partner with folks to bring this all the way through impact. But the way I think about it, people call it different things, but essentially you want to get to, uh, medicine where we-- it's truly precision medicine. It's N of one. We're understanding you and designing therapeutics for you. **Mark Zuckerberg** [34:39] Yeah. I like the mission of Rare as one as well. That's, that's a great framing. **Alessio** [34:44] Do you feel like that's possible, like almost treating the body as like a compiler? It's like, because I know exactly what it looks like, I know exactly what's gonna happen, or is the body just like there's too many outside inputs, and like over time, it kind of deviates from what you have? **Mark Zuckerberg** [34:59] Well, I think we'll see how far we can get, but I, I mean, I'm pretty optimistic that we'll be able to make a bunch of progress. And yeah, I mean, there's-- like you basically-- I mean, what format does this take technologically? I, I would imagine you're taking these different types of virtual cell models and eventually merging them into the equivalent of like a biological omnimodel. Kind of like how on the language model side you had people that did language, and then, you know, people who did different- **Alessio** [35:24] Vision **Mark Zuckerberg** [35:24] ... kinds of media models- **Alessio** [35:25] Yeah **Mark Zuckerberg** [35:25] ... and perception and all that, and then eventually you just kind of merge that. And then, you know, you aim to get positive transfer by merging it, so that way it's not just combining capabilities, but getting everything else to be stronger. So yeah, I mean, technologically, I think that's basically what it looks like is, you know, over whatever it is, a five or 10-year period, we're building up a series of Biohub models that, uh, like increasingly get, um, all these different dimensions of, of data and capabilities that can be used to help run individual science experiments and potentially eventually help with finding individual therapies for patients. Although we're gonna be less on the clinical side. We're gonna be more on the k- kind of scientific tool development side and the kind of main tool, if you will, is this like these Biohub virtual cell models. **Priscilla Chan** [36:17] I would say five years ago, without sort of the large language model supporting this, I don't think it would've been possible to really-- 'cause biology's incredibly complex, and what we're essentially trying to do is break it down from a discovery-based science where you kinda get lucky, you kinda get clever, and you sort of figure out a hack to learn something new- **Mark Zuckerberg** [36:39] Yeah **Priscilla Chan** [36:39] ... to really making it, uh, closer to an engineering problem of like this is how the system works, and when this breaks, what happens to the rest of the system? But, like you said, there's just- there's far too many dimensions for to- for us to hold in our brains. That's why we're so excited about this intersection at this moment because it is possible to consider so many more dimensions matching the complexity of biology. **Mark Zuckerberg** [37:04] What is the role of the doctor in that future, right? If you can, like, predict everything out, and then if you take personal super intelligence seriously, do you kinda distribute some of the diagnosis and all of that work, or h- how do you envision that? **Priscilla Chan** [37:19] I've been thinking about this a lot. Um, and I think one is the model's not gonna take you all the way. You're still gonna need to really look at ind- individual clinical s-situations, and, and the doctor's gonna be a form of data input into the model, right? And so the doctor-- there's some judgment that comes into place. But there's already a lot of models that make doctors really good at what they do. For instance, looking at your skin. Like AI is really, really good at detecting lesions in your skin that are concerning. It's excellent. Retinal issues, it is excellent. So the AI modeling and mapping is really, really good. So it's already happening. So I, I think about, like, what should future doctors be trained to do? And I really think care and compassion and sort of walking patients through understanding, I think understanding why leads to trust in both the science and in the clinical pathway, and really walking alongside patients on that journey is, you know, it was the original calling of physicians, to be healers and to be using great tools to heal patients. **Swyx** [38:34] Well, so bedside manner ultimately is **Mark Zuckerberg** [38:37] I, I mean, I also think you can zoom out though from, like, the role of a doctor to-- I think everyone wants the health system to be more proactive and less reactive, right? So th- today it's like you show up when you're sick, and then, like, you have someone treat you or understand what's going on. I think the goal with a lot of these systems is to be much more proactive about this. So when we say that the vision is to try to help scientists cure and prevent all diseases, it doesn't mean that there's gonna be, like, no bacteria in the world and, like, no one ever starts to get an infection. It's just that, all right, ideally you can kind of understand all of that really early, right? Similarly, if someone, you know, if, if someone gets a mutation it, it looks like it might become cancerous, then, you know, you can just treat it a lot better if you n- if you know that early rather than, like, showing up to a doctor when it's, you know, already metastasized and you, you have a bunch of, of, of issues on that. So, so I don't know. I think that there are gonna be a lot of opportunities to fundamentally improve the healthcare system overall. But I, I, I think-- I mean, I agree with everything that you said on this, and I also just think that, like, when we say that, that we think it's gonna be possible to prevent and cure all diseases, it's not literally that no one ever gets the beginning of a sickness. It's just that it, like, it, it kind of can be managed in a way where everything is sort of manageable. **Swyx** [39:58] I think we discover more diseases the longer we live. **Mark Zuckerberg** [40:00] Yeah. **Swyx** [40:01] Um, is it possible to not die? Is, you know, obviously that's a, that's a meme that's coming to fruition. If you theoretically cure all diseases, maybe death is a disease. **Priscilla Chan** [40:12] Mark just said we had extreme alignment, which I love. Thank you, honey. Um, this is, this is- **Mark Zuckerberg** [40:18] This is one that we don't necessarily agree on **Priscilla Chan** [40:19] ... this is one that I'm not sure we have extreme alignment on. I, in fact, just haven't thought about this one very much- **Swyx** [40:24] Yeah **Priscilla Chan** [40:24] ... because I think there is so much- **Swyx** [40:27] There's other things to do **Priscilla Chan** [40:28] ... there, uh, there's so much to do in terms of like I, I-- you know, I'm a pediatrician. I think about babies, and, like, very sad things happen to very small people. And, like, I think a lot about that, and how do we, like, maximize life quality and the things that harm small people. I'm biased. And I haven't thought as much on the other end of the spectrum. But I don't know. I'm 40. Maybe I should, but I feel like I can still focus on the little ones. **Mark Zuckerberg** [40:56] I think the strategy is the same, right? I mean, it's like we're, we're basically choosing to not focus on any specific disease and, like, verticalize. Our strategy is one of trying to accelerate scientific progress overall, and I, I think that there are a lot of people who are gonna focus on each of these individual things. So I don't know. **Priscilla Chan** [41:14] But we don't have to 'cause that's not our strategy. Our strategy- **Mark Zuckerberg** [41:16] No **Priscilla Chan** [41:16] ... is to make sure that we have tools that make people do the best science possible out there. **Swyx** [41:21] I'll put to you that, you know, because of-- because aging and, and environments are s- and mutations are so diverse, like you actually-- you have a high concentration of grouping, uh, in, in the early years, and it should have more diversity in terms of, uh, the, the, the cell types and, and the problems that you face in, in the later years. And so that-- I mean, there might be some imbalance in terms of, uh, where all these things happen. Uh, but I, I'm not, you know, I'm not, I'm not pitching it any particular direction there. **Mark Zuckerberg** [41:47] No, I mean, I, I mean, I think it's-- clearly if you look at the trend over the last, I don't know what it is, a hundred years, I mean, there was this flip, and if you, like, pay attention to the history of science, where it changed to kind of hypothesis-driven scientific method of like we're gonna run tests and have controlled experiments. And since that happened, the average life expectancy has basically increased by I think it's about a quarter of a year every year over the last hundred years. Now, a lot of that, like Priscilla said, is basically making it so that a lot of people don't die young. ... so far had somewhat less of an impact on extending the maximum human life expectancy. Although the oldest people today, I do think in general are older than the oldest people, you know, 20 or 30 or 40 years ago. But I think that there's been a little bit less of an increase there, and more just, um, kind of making it so that people don't suffer and die prematurely from things. **Swyx** [42:44] Yeah. **Mark Zuckerberg** [42:44] But I mean, and then there's other things that you wanna focus on here, too. It's not just, like- ... how long you live, it's like- **Swyx** [42:50] Healthspan **Mark Zuckerberg** [42:50] ... the quality of the life while- **Swyx** [42:51] Yeah **Mark Zuckerberg** [42:51] ... you're, you know. So I think it's like you can live a full life and have that be high quality, and or you can get sick in different ways that kind of add up over time, and I, I think, like, there's lots of different ways to improve. Um, and it's, um, it's, there's like, you know, all these different analogies that you could throw at this, but I, I think that there's, there's just a lot of, lot of room to improve here. **Priscilla Chan** [43:14] Yeah. **Swyx** [43:14] And then the other element I wanted to come back to on the engineering side, which is when you present it a high dimensionality problem, you want to reduce things into little boxes that you can sort of manipulate at a higher abstraction. And that's something I, I try to do with, uh, the folks outside, and it's really we really struggled because- **Priscilla Chan** [43:30] Oh, yes **Swyx** [43:31] ... over here you're imaging on, like, the atomic level, and then you're also worrying about proteins, and then you're also trying to build a cell model. Is every abstraction leaky? Like, where is the b- boxes I can move around then and not worry about it? Now, my, my physics, physics analogy is in the regular world you don't have to worry about quantum physics, but here we kind of do. **Mark Zuckerberg** [43:48] Mm-hmm. I think you wanna build it up a little bit hierarchically. And like, I mean, when you're trying to understand proteins, understanding molecules makes a big difference, but at some level you can kind of just look at, like, correlations in cells. But if you wanna, like, really have the most accurate model, and if you wanna be able to reason about things, then you probably also wanna understand proteins well, and then I think that kind of extends. But yeah, I mean, that's part of the interesting challenge of this, is that it's not just, like, one le- one resolution that you're looking at it. I think in order to do it well, it's, y- yeah, I mean, it, you're, you're kind of, you have some amount of abstraction, but I think you want the models, just like language models or, I think, how our brains work, to basically build up different levels of abstraction and pattern matching, and I think that that's here, too, and you basically just need to be kind of like have some basic excellence and understanding at each of these different levels. **Swyx** [44:38] It's weird. The, the, the, the number of levels at which you have to telescope up and down, it's, uh, mind-boggling, and I think, like, that's, when, when people say dimensions, they typically mean orthogonal dimensions, but here it's sort of like nested- **Mark Zuckerberg** [44:50] Yes **Swyx** [44:50] ... and- **Mark Zuckerberg** [44:51] Yeah **Swyx** [44:51] ... and, yeah, it's just- **Mark Zuckerberg** [44:52] Just different scales that are, like, oddly different disciplines to understand each- ... specific scale. **Priscilla Chan** [44:57] Yes. **Mark Zuckerberg** [44:57] And it's like in a way that, like, the people who are good at understanding one scale are, like- **Priscilla Chan** [45:02] Have never spoken- **Swyx** [45:04] Yes **Priscilla Chan** [45:04] ... to people at the next scale. **Mark Zuckerberg** [45:05] Yeah. **Swyx** [45:06] Yeah. **Mark Zuckerberg** [45:06] Yeah. **Swyx** [45:06] Yeah, physics is there, chemistry is here- **Priscilla Chan** [45:07] Yeah. **Mark Zuckerberg** [45:08] Yeah **Swyx** [45:08] ... and yeah, bio's there. It's nice to hear about it, but when you see it and you meet the people, you're like, "Oh, this is real." Like, and they are, they are actually working together. **Mark Zuckerberg** [45:16] Yeah. ### Immune System **Priscilla Chan** [45:17] Yes. **Alessio** [45:17] And then there's this goal of the virtual immune system that you're working towards. I would love to, for you to chat about that. And also, like, if that happens, what should other people build? So there's obviously, you know, CRISPR and some of that technology, it's like the people should maybe ramp throughput for. Like- **Mark Zuckerberg** [45:33] Mm-hmm **Alessio** [45:33] ... uh, how do you think about the future? **Priscilla Chan** [45:35] The virtual immune system I think is, is, you know, obviously, uh, I, I think of a subset of sort of the generalized model eventually we'll get to. But the virtual immune system is super interesting for a couple of reasons. One, it's individual cells interacting with each other. There's, you know, a l- a number of, uh, uh, cells that we don't even fully understand what they do, B cells, T c- cells, NK cells. And so we can use our current technologies to understand these cells at a more granular level, so that's cool from a biology standpoint. But the clinical impact is huge of understanding the immune system. Because biology, turns out, has already given us a way to keep the body healthy, and it also sometimes goes awry and causes disease with autoimmune disease. **Alessio** [46:24] Yeah. **Priscilla Chan** [46:24] Right? And so it's a very complex system that has to stay in balance, and if it goes out of balance in either direction, you get sick. It can also go into your body, and it's a privileged system that is mobile and can go into places like your brain, your pancreas, your heart to sort of either do maintenance or to collect signal. That's, it's built in. So if we can understand this system, we can use it to keep people healthy. We already kind of do. So there's CAR T cells, uh, where we reprogram T cells to go in and fight cancer. In our New York Biohub, we're doing cellular engineering to say, like, "Hey, can you go in to this person's heart, check if they have, uh, plaques that are causing problems, read it into your DNA, self-lyse, and then we can read out the signal as cell-free DNA and give us a binary answer, yes or no." Then we can, uh, put in other engineered cells and imagine where you go in and you clear out the plaques using engineered immune cells that are your own. That is incredible. That is a tool that, like, can, is, is realistic, too. I know it sounds sci-fi. It is realistic. It is happening. And then on the other end of understanding the balance, like so many autoimmune diseases, MS, lupus, those are the ones, uh, examples of ones we know. I, I think there are other things that are autoimmune that we don't understand, like dementia can have a, autoimmunity can play a large role in that. And so if we can understand the fine balance that the system needs to be kept in, then we can actually impact a lot of the ways the, uh, the human body is maintained. Um, so I think it's both interesting from a biology perspective and feasible to model, and probably one of the highest sort of now impact systems if we can learn how to manipulate. **Alessio** [48:12] Amazing. Um- **Mark Zuckerberg** [48:12] But it's only one system, right? I mean, it's- **Priscilla Chan** [48:14] Yeah **Mark Zuckerberg** [48:14] ... I think it's like the- **Priscilla Chan** [48:15] So it's a subset, yeah **Mark Zuckerberg** [48:16] ... if you're focused on curing and preventing diseases, the immune system is a pretty important one, and I think it's also interesting for all the, and unique in a lot of the ways that you said, but there's, like, lots of other parts of the body to understand, too. ### Timeline **Alessio** [48:28] And I think we're running out of time, so we have two questions to close. One, again, 100 years maybe is too long, right? What would it take to do it in 50, in 25? And to make those happen, like, what should other people build to support, um, your work? **Mark Zuckerberg** [48:42] I mean, I think a lot of this is gonna end up coming down to how far a lot of these AI methods get, right? I think that there's, like, people have-- there's just this constant ongoing debate around what are the timeframes for getting to very strong AI, and I think if you, if you get that, then I think it's pretty optimistic that with the right investments in frontier biology, you should be able to get these systems that can allow you to have virtual cells, that allow you to do the kind of precision treatments and preventative care that can achieve this kind of mission, um, significantly sooner. But at the end of the day, I think a lot of that timeframe will probably come down to the AI timeframe. There's obviously a ton of stuff to do in biology, but, but it's-- so it's not, uh, I mean, like, I think that what, what, what should other people do? I mean, other people doing more frontier biology and, and helping to collect the, this type of data and solve these problems is super helpful to that too. It doesn't automatically happen. But I guess if we're, you know, predicting whether it's gonna take 10 or 20 or 40 years, that is probably more a function of the pace of AI development than it is a pace of the pure biology side. **Priscilla Chan** [49:52] Yeah, I was gonna agree with you. I think it's a, a lot needs to... It's-- I think we're on a path to ga- get a lot of important biological data through, uh, advances in laboratory technique, but it's not a given. Like- **Mark Zuckerberg** [50:05] Mm-hmm **Priscilla Chan** [50:05] ... and there are different groups that are expert at this all across the nation and across the world, and so we need to be continuing to push the research and the methodologies. And I wanna say that, like, you know, the Cell Atlas was not glamorous work. People were not gonna get their tenure track paper by sort of analyzing the hundredth and twenty millionth cell. That is just not it, right? And so rethinking the way that this work gets done in a collabo- like, doing big things together in science, like, that's what built-- is gonna need to happen to sort of get the knowledge we need to build models that give us this type of insight. **Mark Zuckerberg** [50:47] I guess one thought on, on, like, the type of biology that I think should get done is there is a certain orientation around choosing problems that will help generate data that can help make the models a lot smarter. I think that there's a-- you do that when you are very optimistic about the pace of progress and what AI is going to enable. 'Cause the classic reason that scientists generated datasets is so that they could basically look through the datasets to make advances. So it is a little bit of an inversion in the thinking, which is like, "I'm now going to do this so I can, like, help train this other thing to be better and create more advances." And I think in a world where you really believe that there's gonna be very significant AI progress, I think more frontier biology should be done in that way. But these datasets aren't going to get created by themselves. There's a lot of work that needs to get done and a lot of investment there. And at some level, you could probably have, you know, the smartest AI model in the world, but if it doesn't actually have the data to understand this stuff, it's like, okay, you can't just, like, reason from first principles about, a-about all these things. I mean, a lot of human knowledge comes empirically, not from first principles reasoning. I think that more-- this is kind of the whole Biohub network idea that we're building, and I've been really happy to see other folks, especially a lot of people in technology, I think have this orientation too. They're-- you know, they, they believe a lot in AI, they believe in the technological progress, they've generated some, some significant wealth for building their companies, and now they're investing in, in science research, and I think that's great. And I think doing it in this way where you're, like, building up these networks to solve specific-- to basically build specific tools that generate data that make the models better, it's one approach. It's not like measu- it's not that all science should go in that direction, but it's one of the things that I'm quite optimistic about that I, I think is gonna make a b- very big difference. **Alessio** [52:43] Cool. That's probably all the time we have, but I'll just leave it to you guys for any calls to action, anything that you want biologists or engineers to check out. ### Closing **Mark Zuckerberg** [52:52] I mean, check out the models. **Alessio** [52:53] Check out the models. **Mark Zuckerberg** [52:54] Um, you know what I mean? It's- **Alessio** [52:54] The tooling. **Mark Zuckerberg** [52:55] Yeah. I mean, it's-- they're, they're early, but I think it's-- they're, um, they're-- it's kind of an interesting dir-- sense of where things are going and, you know, we'd love feedback on it, and it'll kind of just help this feedback loop of, of, like, what we should build next. **Priscilla Chan** [53:09] Yeah. I would say let's do this together. Uh, like, we need lots of people coming together to do this work. **Alessio** [53:15] Well, thank you for- **Mark Zuckerberg** [53:16] Awesome **Alessio** [53:16] ... organizing it and solving and curing all diseases. **Mark Zuckerberg** [53:20] Try to help others do it. **Alessio** [53:22] Yes. All right. Thank you guys. **Mark Zuckerberg** [53:23] Thank you. **Priscilla Chan** [53:23] Thank you. --- This library is powered by PodHood (https://podhood.com), the podcast website platform.