Building the Platform for Scientific Breakthrough, with Noubar Afeyan of Moderna and Flagship Pioneering

This week on No Priors, Sarah sits down with Noubar Afeyan, Co-founder and CEO of Flagship Pioneering, the biotech firm behind groundbreaking companies like Moderna. They explore how Flagship creates the conditions for scientific breakthroughs, tackles regulatory uncertainty, and pushes the boundaries of discovery. Noubar shares insights on AI’s role in healthcare, the challenges of bringing new therapies to market, and lessons learned from past pandemics. He also discusses Flagship’s platform approach to biotech innovation and introduces the idea of polyintelligence. Sign up for new podcasts every week. Email feedback to [email protected] Follow us on Twitter: @NoPriorsPod | @Saranormous | @EladGil | @NoubarAfeyan Show Notes: 0:00 Introduction 0:48 Founding Flagship 5:51 Fostering environments for emergence 11:17 Expanding into new frontiers 14:26 Developing technology amid regulatory uncertainty and risk 19:12 How Flagship has evolved 22:47 AI applications in healthcare 27:30 Bottlenecks in bringing new therapies to market 32:20 Lessons for the next pandemic 34:11 Building a platform 38:10 Polyintelligence

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:05Sarah Guo?

  Hi listeners and welcome to No Priors. Today, we're joined by Noubar Afeyan, founder and CEO of Flagship Pioneering, the firm behind Moderna and over 100 other biotech companies. We'll talk about his approach to building biotech startups, how AI is reshaping drug development, and his theory of polyintelligence. Welcome, Noubar.

• 0:24Speaker 01

  Thanks again for doing this and I look forward to the discussion.

• 0:27Sarah Guo?

  Noubar, let's start with the roots. You've had this incredible journey personally. You arrived as a teenager after your family fled war-torn Beirut. You earned MIT's first PhD in biochemical engineering. And you know, over the past three decades, you've created a force with Flagship that's changed the trajectory of global health with many important biotech companies and more than $100 billion of value. Can you just talk a little bit about your motivation to start Flagship originally and what you thought it might become?

• 0:53Speaker 01

  I will work hard to fit the description you just gave of what I've done or what I'm trying to do. The motivation for Flagship stems from what I was doing before, which was that I started a company in 1987 when 24-year-old immigrants didn't start companies in this country. But instead, it was kind of like former Merck senior executives or IBM senior executives were the only ones who were entrusted with the massive amounts of venture capital, namely $2, $3 million per round, used to go into venture capital. So this was very early days. And I had the kind of chance, opportunity to start a company right out of my graduate school and ended up raising quite a bit of venture money

• 1:35Speaker 01

  and eventually kind of went down a path of entrepreneurship. Along the way, one of the things that interested me was why it is that kind of the entrepreneurial process was supposed to be random, improvisational, kind of idiosyncratic, almost emotional, gamy. All of those things, I kind of thought was a bit of a put-off when it comes to actually doing things in a serious professional way. And I kind of used to go around in the very early 90s saying, "Why isn't entrepreneurship a profession? And if it was going to be a profession, how could it be a profession?" And at the time, you know, there were largely one or two competitions, you know, giving prizes, which of course reinforces the gamy nature of it.

• 2:20Speaker 01

  Uh. And so I started thinking about that. I then started thinking, well, one way to know you're doing it as a professional is if you can do many of them in parallel. And my motivation for that was investing. Because of course, venture capital is a parallel investing activity. If it was the case that you could only do one of these at a time, you would have serial venture capitalists, not parallel entrepreneurship capitalists. And yet, people think entrepreneurs are supposed to be serial, but investors or lawyers or everybody else can operate in parallel. The learning cycles of doing things at same time is completely different than if you actually force yourself to think about the essence of what you're doing and what's reproducible, what has to be different in each case. And so I got interested in that.

• 3:04Speaker 01

  And parallel entrepreneurship led me to say, well, how do you do that even? So I spent the late nineties alongside running my first company, getting involved in the co-founding of several other companies, where I kind of tried this out individually. And then I realized that it's pretty hard to do as a solo player. And so then I created the first company creation company. It used to be called New Cogen when it started. It stood for new company generation. And that's the name we operated under for three years until people told me, it sounds like a disease. And then I kind of changed it to flagship. By the way, that's exactly what happened. But from then, we've been on a journey to figure out not only how to do professional entrepreneurial activity, but institutional, which means act just like investing became institutional,

• 3:49Speaker 01

  act as teams, uh, with, with company kind of objectives, et cetera. So that's a complicated way to describe the motivation. But maybe I'll simplify it to say, I thought that the activity, the, the most value creating activity that I know of in, in the current human endeavor is starting companies. I used to teach entrepreneurship at MIT for sixteen years and then Harvard Business School taught for three years on innovation. And I always ask people, what's the single most value creating activity? What's the biggest invention that humans have made in that regard? And they start telling me internet and satellites and this, that and the other. And I'll tell them it's just the startup, the human invention and the value created from the startup

• 4:32Speaker 01

  includes Google and Tesla and Facebook and Genentech and every single one of those companies. And I thought, why are we outsourcing that to some random, you know, gamey activity? And that's what motivated me to start Flagship.

• 4:46Sarah Guo?

  What do you mean by gamey?

• 4:48Speaker 01

  Because it's like supposed to fail most of the time. And once in a while you win and then you celebrate the win. And what I mean is like it.

• 4:55Sarah Guo?

  It's random.

• 4:56Speaker 01

  But not only random, but there's like winners and losers and keeping score. I don't know, it's maybe the wrong word, but I just mean like people even call gamification in the in the software space, there is a version of this, like I don't mind being playful. Cause if you're overly serious, sometimes you miss things, but it can't just all be play. We take hard earned money. We deploy it to do things that are damn near impossible. Once in a while we reduce them to practice so they become not only possible but valuable. And yet people treat it like, oh, well, you know, it didn't work. There's 20 different things we tried, one of them worked. That, I don't know, as an engineer by background, as a scientist, I just thought that what we do, especially listen

• 5:38Speaker 01

  in healthcare, especially in climate, especially in kind of like agriculture, food security. You can't think of this as, you know, like shots on goal and this, that. You've got to kind of say, hey, we can get better at this.

• 5:50Sarah Guo?

  One of the things that Flagship has sort of famously espoused is this idea of allowing breakthroughs to emerge rather than being made by people. How do you square that? Like, how do you create an environment where breakthroughs like emerge naturally and are also sort of predictable and controllable?

• 6:08Speaker 01

  First of all, if anybody listening to the podcast wants to go read more about it in Harvard Business Review. I finally, after 23 years of doing it, kind of like we published a paper a couple, three years ago, uh, together with Gary Pisanow that laid out the idea of emergent innovation. And what we mean by that is the following. If you think about human design, it's goal based, right? So what is the goal of creating a new company? Well, you could say selling it. But but if you said, what was the goal of creating? I don't know, Nvidia. And I don't think you can define a goal, right? Because they just kind of thought they could create valuable products that would impact people initially in the gaming specs. And then all of a sudden, the AI opportunity. What happened to the AI opportunity? It emerged.

• 6:51Speaker 01

  It was not a predictable, kind of like in the business plan, except it was in the five year plan, not the thing. Not at all. The opportunity emerged. So where else do we see activities where novelty happens in this unpredictable way? Answer, nature. So how does it happen in nature? In nature, the forces of variation, selection and iteration create unbelievable novelty and really esoteric complexity of all sorts of kinds. Life, I mean, that's a pretty, we still have a lot of living things that I know, and I don't think there's that many living silicon things. And so that is emergent. So I would say that if you do variation, selection, iteration in anything, you get emergence.

• 7:35Speaker 01

  You do it in human thought. You get, you know, kind of like revolutions. You get memes, you get political thought, you get religions, you get lots of things. You do it in running shoes. You get, you know, Air Jordans, you get, I mean, that's what you do. All of this is emergent. It's just that people who participate in it describe it as though they came up with it. And and they kind of describe everything in this unbelievably, sorry to say, conceitful, characteristically human way. And so the fact that we have our own language and therefore like now large, large language models that is really good at taking reality and converting it into how it is that the human agent presided over it.

• 8:19Speaker 01

  That's like saying, you know, whoever wins the war gets the right history. I'm saying this, having practiced this for thirty eight years. I don't believe anything that I've been involved in innovating and creating actually is the product of my work. It's been emergent. What I try to do and what our whole organization tries to do is to create an environment within which emergence can happen. And then have the humility not to describe it like some act of genius. You will not hear me giving interviews ever describing these things as some like super intelligence that we wielded. Versus it's an emergent thing. And by the way, generative AI is one hell of a technology for emergence, which we're using for that in pretty cool ways these days. Whether it's emerging new proteins or emerging new consumer products.

• 9:04Speaker 01

  So I couldn't be more excited, but it's emergence. That's what we harness and have used for twenty five years. You know, ever so gradually more systematically, but we got a lot to learn.

• 9:14Sarah Guo?

  I think that's an incredibly intellectually honest point of view. I think maybe one of the things you're alluding to is the very best entrepreneurs tend also to be guilty of this. The narrative changes to have the story in a sort of a clearer narrative light from the beginning, where people had much better predictive power than, you know, anybody who's been around real companies that have success like really, really understand.

• 9:37Speaker 01

  I used to tell people that most um successful entrepreneurs who've done, you know, kind of like unexpected things, also secretly worship at the altar of chance. Because they also realize that at the end of the day, their explanations of what happened do not suffice for actually what happened. And while they'll never describe it saying, we did this, then we did that, then we did this, and then we did that. And then something happened that out of the blue, I have no idea why it happened, and that's what led us to success. Nobody writes books about that. And yet, if you interview, as I know you have, lots and lots of people, they'll kind of say, look, a lot of things that happened, we got fortunate. But we were ready for it because we did everything else as best we could.

• 10:22Speaker 01

  That's what I believe.

• 10:23Sarah Guo?

  Yes, and I think, you know, we'll get to talking about AI, but at least for us at Conviction, the view is it is very hard as an environment to predict even, you know, what happens with the technology, what applications are relevant. But you can say like, there was alpha in the interest to begin with. And we can be prepared for the opportunity and we'd be fast to adapt and identify and sort of cultivate that opportunity or or select down from the what ifs in flagship terminology.

• 10:50Speaker 01

  And you use the word adapt very correctly because, you know, a lot of times people say, what comes first, the variation or the selection? And the answer is just the selection pressure in nature comes from the sum total of everything that exists in nature. It's the same here, you know, consumer preference often reacts to what's put in front of it, which then forms confused consumer preference. And so it's really hard to separate these things. You have to create an environment in which these two interact and then you get new products and new services.

• 11:17Sarah Guo?

  Flagship, just in terms of ambition and scope, doesn't just do therapeutics. You've done ventures in nutrition and agriculture and climate, these different areas. How do you think about opportunities to extend into beyond medical biotech?

• 11:31Speaker 01

  We're very careful in that regard, but but where we have a core advantage, whether it's intellectual property we've created or maybe a daring that comes from not knowing enough about the space, we will kind of venture into it. And, you know, it's for us, the first thing we do in a space informs the next five things we do. And then if all five things don't succeed, we'll kind of say, you know what? Maybe we can't get, you know, paid for the innovation in this space. Our methodology is all about trying to bring to life today what might otherwise exist five years from now. Not everything that will exist five years from now will be valuable. So on top of that, we got to come up with something today that's also going to be valuable.

• 12:15Speaker 01

  And that is not for every sector. So for example, we worked for many, many years in renewable energy. Turns out that, you know, one of the most advanced ways we could make carbon neutral liquid fuels was to engineer photosynthetic bacteria that usually grows in the depths of oceans, really cool bacterium that sees a little bit of photons. We engineered these things to make diesel. It could literally consume CO2. And and make diesel fuel and secrete it. And people thought it's impossible and we did it. But and then we created reactor systems dirt cheap to do this in. Next thing you know, we go out and this was in the two thousand eight to two thousand twelve timeframe. So a company called Jewel. Well, guess what? You know, it's a commodity business.

• 13:00Speaker 01

  The price of diesel, whether it's renewable or not, is the price of diesel. The price of carbon when we started was fifty bucks a ton and when we ended it was five. And when we started, the US was energy dependent. And when we were done, the US was gushing with energy, you know, liquid fuels and oil and gas. So in the end, you realize that no matter what you do in that space as an innovator, you're not going to actually be able to get a premium. So we learned a lesson that says that sector, at least back then, didn't deserve the kind of cutting, leaping innovation that we could make. Since then, people have said to us, why don't you go back and do that? And the answer is, I still don't know if it if it's going to pay for it. Because just when you thought green and carbon was going to be priced, here we go again. So to your question, you might be surprised.

• 13:45Speaker 01

  We started companies in in supercomputing back in the two thousands. We started companies in networking.

• 13:50Sarah Guo?

  Oh, I didn't know that.

• 13:51Speaker 01

  Yeah, it's crazy, but we have. And there were like one offs where we said, you know what? Let's see what we can do in a space, not not out of lack of discipline, but since we were doing it in-house, we could kind of take our time figuring it out. And as we speak now, we have started things recently in the material space. So kind of semiconducting materials, carbon capturing materials. So we'll experiment. We do have this kind of almost emergent mindset as well about where we will apply ourselves. But we know that not everything needs scientific leaps. Not everything lends itself to this kind of activity, so we experiment.

• 14:25Sarah Guo?

  Many flagship companies are pioneering new categories that regulators don't know what to do with. And you know, we face some parallels here with some of our companies now. But if I think about microbiome therapies or gene editors or these categories, this seems like another like somewhat brutal market dynamic along with like the cyclicality that you're describing. How do you think about the the risk of this? These new markets that you have to shape policy and public opinion and pulling the future, you know, five years in is not just a technology problem, right?

• 14:58Speaker 01

  It's a very good question, and that is one of the things that we. Think carefully about and and obviously in hindsight inadequately about because we can't control all the variables. I think that risk is a concept that is best applied to adjacencies of what already exists. I usually tell people, imagine a circle. Everything inside the circle exists today and is known today. Everything just outside that circle, the adjacent circle around it, is what's going to be known over the next interval. And I don't know which way the innovations will happen, but that's where most innovations happen, in this circle of adjacency. In that circle, there's an advantage in that people sitting today can make estimates of risk and reward.

• 15:42Speaker 01

  And that's what due diligence is all about. That's why you ask multiple key opinion leaders, and they tell you what they think, and then you aggregate, and you just make an investment decision, you judge how the management, etc. But then take that distance from the current further and further and further out. At some point, people can no longer estimate the likelihood of success and the rewards of success. So now, what do you call that? I would not call that risk anymore. I would call that uncertainty. And uncertainty is things you cannot attribute probability of success to. What we do as humans is that we consider those things nevertheless risk, because there's been this economic kind of like drive largely by Wall Street and others, that everything can be put on a risk matrix. I don't believe that.

• 16:26Speaker 01

  And and therefore, we kind of just view that as super high risk. So for example, fusion today, even though there's a few people who are claiming they could do fusion in a shorter timeframe than thirty five years, is, what is it? Is it risk or is it uncertainty? I would say it's not risk, because you can't tell me what the probability of success is. I'm going to come to your question about market success, regulatory success, policy, because this just adds more layers of of this uncertainty. But what we do is we kind of say, you know what? Why would you think that in adjacencies, there are extraordinary value pools? Because one thing you can be sure of is that everybody is working on adjacencies. Every startup is, every academic lab is, every large company is. Everybody is in the adjacencies.

• 17:10Speaker 01

  So I worry in the scientific fields. I'm not talking about business model innovations and service models. I don't, frankly, fortunately, I don't understand any of that, but so I stay out of it. But in the science-driven, technology-driven space, I think adjacencies have their own problem, which is the risk of commoditization, which is a risk they usually don't take into account. But in our world, there's an uncertainty. So what do you do when you face uncertainty? A subset of things that are uncertain, that are not yet known whether they're valuable or not, or they can be done or not, you just go do the experiment. And if you can come up with the right experiments that can first bring it to life and see if it can be made real, you still haven't done anything about some of the uncertainties and risks you mentioned,

• 17:53Speaker 01

  but at least you kind of can control what you control. So the way we look at it is, yes, if I had a choice between a really hard technical feat, but that once done had a ready market versus one that also then had to be totally mRNA, which was our 18th company that became Moderna, is a good example of this. There was no mRNA drug or vaccine before. There was nobody working on it. When we started, there were academic labs that had worked on it and given up. And so like, what is that? You need regulatory change or at least acceptance. You need market pricing. You need manufacturing that we had no idea how to do. But it turns out that if the reward that you could try to foresee with or without a pandemic,

• 18:36Speaker 01

  frankly, we had plenty of value it created without the pandemic, kind of distorted the path. But nevertheless, I think that that's how we think about it. So we carefully embrace uncertainty and try to resolve it on the way to understanding what the risks are. And then we try to mitigate the risk. And if you're not willing to do that, you're going to work on me too valuable. And I just, we, one thing we know, and now everybody who works at Flagship knows, is that we're no smarter, no harder working, no better connected than anybody. What we can do is to underwrite uncertainty.

• 19:10Sarah Guo?

  It's a really interesting orientation. One more question on flagship. What has changed over the last 25 years? Like, how's the flagship approach different today than in 1999? As somebody who started a fund two years ago and, you know, an incubator and a bunch of things.

• 19:24Speaker 01

  In nineteen ninety nine, two thousand, you have to realize that the world was being overtaken by the internet, in particular e-commerce. Ironically, at the same time, the human genome sequence was being done and I was involved in the company that did that, the private effort, Solera, that did that. You know, it was a weird time because all the money was kind of being siphoned into, you know, sunglasses dot com and diapers dot com and all these things. That was the fad of that day. It was really hard to get money to do anything with life science and medicine, et cetera. And that's important because we kind of realized that nevertheless, uh, there was a big need for medicines, et cetera. So we kind of started focusing a lot on this intersection between biology and technology twenty five years ago.

• 20:09Speaker 01

  From day one, we had this notion that you can conceive and create companies systematically. And we wanted to learn how to do that and get better and better. What we didn't do is to initially bet on being able to systematically make breakthrough innovations. The way we've learned how to do, we believe, since that time. So today, flagship is five hundred and fifty people. About two hundred plus are scientists, engineers, M Ds. We file six, seven hundred patents a year centrally. And every one of the things we work on has essentially no connection to what's been done before. That second innovation, which really came about in the late two thousands, is what's changed. The other thing that's changed is that we were a small organization. We were about fifty people as recently as seven years ago.

• 20:54Sarah Guo?

  Wow.

• 20:55Speaker 01

  It's only in the recent past that we've brought in-house the capabilities to scale companies internally, not just to conceive them. And so that has led us to have, you know, a, a, an internal engine through which we now have many, many people who know how to build companies in parallel and many of them. And so the learning cycles have accelerated. And I'd say what's changing much more is that we have a technology tool through generative AI. By the way, we've worked with AI-based companies for twenty five years. This is not like the revisionism you hear these days. Everybody wants. We literally started a company in two thousand and one called AfiNova.

• 21:38Speaker 01

  And if you go back and look at what it did, it basically used machine learning evolutionary algorithms to evolve consumer products online, starting in two thousand and one. And we worked for six years developing the machine learning tools, which were basically dynamic evolutionary algorithms. And then this has nothing to do with, by the way, uh, DNA or RNA, except it's metaphorically doing what nature does. And we've gone back to doing that. Our hundredth company, FO100 now, is basically developing that with now modern generative AI. But what you could do with generative AI in this kind of like leaping conception hypothesis generation, it's really remarkable. So there isn't a day that goes by that doesn't change. You know, we've also changed how we think about our space in the innovation kind of value chain.

• 22:23Speaker 01

  So we more and more think that we can generate breakthroughs that then the incumbents in the space can benefit from. So we've set up large partnerships with the likes of Pfizer and Novo and GSK and Pharma, the likes of Thermo Analog Devices, Samsung in the tech space, all aimed at kind of expanding the reach of our innovations so that we can have more impact. So lots of learnings along the way.

• 22:48Sarah Guo?

  Let's talk about the AI piece. What do you believe are the most exciting applications of AI in healthcare?

• 22:54Speaker 01

  If you think about AI largely as data, large data driven models that can do kind of things that begin to look like human cognition and things than we were able to do with just correlation, kind of statistical kind of things and optimization. And some of the things that machine learning used to be developed for, you start getting more and more ambitious in how to use it. So among the very first things we did, beyond Moderna's use of kind of really early deep neural nets, followed by the other techniques, to take data in the whole mRNA space and inform the design of the next and the next generations of what we did, as well as how we manufactured. But they were, Moderna was doing that on the one hand.

• 23:37Speaker 01

  But we started looking at that as a way to design proteins. So back about six, almost seven years ago now, we created a project, the what if, that said, what if you could computationally design a protein of any desired function? And you might say, ah, of course, now, you know, we have AlphaFold this and we have quantum models of folding. So we didn't want to use any of that. We literally said, can you show enough instances of the desired function with the underlying DNA sequence for a learning algorithm to generate new ones? And people said, no, no, no, but you need to know the DNA sequence, then the protein sequence, then the folding structure, then the this and the that. And we said, yeah, but last we checked, every generation that DNA gets handed to

• 24:20Speaker 01

  the next generation, there's no manual that describes all these things. The DNA has no idea what a protein is, has no idea what folding is, and yet, boom, the function follows. So we said, there must be an encoding of that knowledge in the DNA.

• 24:34Sarah Guo?

  There's enough data in there somewhere. Yeah.

• 24:36Speaker 01

  A hundred percent. There's patterns that are basically encoded, which we don't understand. So on that whim, otherwise known as hypothesis, we literally just started and tried it. Because that's what's changed now, is that the incremental cost of actually asking this question has come down, not just with AI, but also with experimental setups. And in fact, within a couple of years, we could show that at least for antibodies and how they bind to their targets, we could begin to show some pretty interesting leaps that you can make computationally that you couldn't make experimentally, at least not in the same time. And then that has led now to a company called Generate Biomedicines, which, by the way, is one of the very first large partnerships that Nvidia did in the biology space. So that's how we know them quite well.

• 25:19Speaker 01

  And this was several years ago, three, four years ago. And now, you know, uh, uh, Generate has more than 15 different, you know, computationally designed antibody programs, some in the clinic, some advancing to the clinic. And, you know, a lot of people kind of go, well, show me a drug that's been competition designed. The answer is showing a drug is a lot more than the design and early testing. But we have done that in space. We've then applied that to cell models, we've then applied that to DNA, RNA, all sorts of molecules, to lipid nanoparticle design. And now we've expanded some of the very cool advanced things we're doing is literally to create. Novel platforms that can essentially create autonomous ways to doing scientific discovery.

• 26:02Speaker 01

  That is generate hypotheses, specify experiments, run them, collect data, interpret the data, generate hypotheses, and just do science, the way Waymo drives cars. And while we're nowhere near having Waymo on the streets with these things, we definitely can show the elements and we can tie them together. And in narrow spaces, we could show what the future of that could look like, what it looks like if people did in chess and go train such that, you know, a million people for a thousand years might play enough games to get to that level. We're beginning to see hints of that. So that's very, very cool. And then, like I said, to finish, the area that it really interests us is these multi-agent kind of systems that can do emergence.

• 26:46Speaker 01

  We're doing that in the product space, just flat out kind of new brands, new products. We're doing that, obviously, learning carefully about those business models. We're doing that in the mental health space, so kind of early intervention so that you can use agent-based interventions. But again, I don't mean by that training, you know, what a doctor would do. I mean, let the system interplay and kind of learn from the dynamic between different types of agent-based models. Very little we're not excited about. You'll notice I didn't talk about all the productivity gains, you know, automated document writing, summarization. I mean, we're doing all of that like everybody else is, but that doesn't require our extra effort. These things feel like pioneering to us.

• 27:30Sarah Guo?

  You were just describing hypothesis generation and other advancements of this type. They're leading to a quickly increasing number of candidates, but that's, as you said, like not the, not show me a drug. That's just kind of top of funnel. Maybe it's higher quality top of funnel as well as volume. What are the biggest bottlenecks in translating these innovations into like market-ready therapies? And how do we address these challenges more effectively? If the goal is to get more therapies to market now that we have more potential therapies?

• 27:58Speaker 01

  Let's work backwards. The last thing you have to do is to file with the FDA, some BLA or NDA to get your approval. The step before that is you have to conduct the phase three trial to show at the right dose level across a large enough population statistically significant. Superiority without toxicity and then we can work backwards. Those last steps are regulated and unless there's a rethink that with data driven kind of approaches, we don't need quite as much of the kind of analog testing that we do because we can actually create models of what the data is telling us. That day will come, that day, until that day comes, we're going to be waiting, you know,

• 28:43Speaker 01

  for the large scale trials and the hundreds of millions of dollars that it costs to do that.

• 28:47Sarah Guo?

  Don't you believe that should, that should come sooner?

• 28:50Speaker 01

  Of course it should, especially if it, if it could have improved the disease I'm going to get to die from. So it's totally nuts. And by the way, the way to do that to me is pretty straightforward. It's called Operation Warp Speed. What we saw during COVID.

• 29:04Sarah Guo?

  It's clearly possible. Yeah.

• 29:05Speaker 01

  Completely is the way of thinking that would result if you are, you know, under an avalanche of, of disease threat and you're very, kind of like you get organized such that the private, the public, the regulator, everybody doesn't take cut corners, but essentially realizes the objective is a solution, not slowing down the process just to be on the safe side, in which case people are dying. So when the, when the consequence of not doing anything goes up, people act. The sad part is the consequence of not doing anything in cancer, in, in neurological diseases is playing out every day.

• 29:39Sarah Guo?

  Yeah, it's very large. It's just slower.

• 29:41Speaker 01

  We could do better, but it doesn't feel like a, an urgent threat. I wish it did. So look, I, I don't have a magic wand, but if I did, I would at least run some experiments in how we think about data informed ways to interpret the results of trials that can either adaptive trials or et cetera. So that's one category. The area we are working on, just so you know, is that we can do better at understanding the state of a particular patient. Like right now, people say stage one cancer, stage two cancer, as though there's such a thing. The medical profession has created stages. Every once in a while, they change them. It's a joke. That's got nothing to do with underlying science and biology. No offense to the colleagues who work in that area. It's just kind of a super macro approximation.

• 30:28Speaker 01

  But I think we now have the tools to be able to, in a very kind of molecular way, micro stage, what I call a biostage disease, on a trajectory that isn't four stages, but it's seventy five thousand stages, if you want. And the key for that is that you can then look at what mechanisms are turned on and off in any one disease during that trajectory. And what's the subset of people that you should choose to homogeneously test your hypothesis or a drug, such that the other people you tested it on doesn't defeat your trial, so that you might do a smaller trial, get a smaller indication approved, and then expand. That with AI tools and with measurements can be done. We're working hard to do them.

• 31:12Speaker 01

  That could substantially increase the productivity and lower the, lower the time to get a drug out. Unfortunately, it doesn't suit the interests of a large company who wants to give the drug to everybody, but for biotech companies, who otherwise would be dead on the way kind of to waiting, this might be an interesting innovation. So we got to get the regulators to be open to that. There's a lot of information in patient data that could inform us in ways to try to find the right mechanisms to go after. Accessing that patient data, given HIPAA rules and everything else, using it to train models to say, wow, now we know that it's this particular mechanism in a subset of Parkinson's diseases that you got to go after.

• 31:57Speaker 01

  That stuff's happening slowly, but that can also increase not just the top of the funnel in, in, in molecules, but the selecting down the things worth working on, because you already have human data that says that if you can do this, we already know the consequence because we have human data showing it in the kind of genetic testing that we can do. So lots of exciting things, but it's, it's slow.

• 32:20Sarah Guo?

  So you mentioned Operation Warp Speed. Like, we have to talk a little bit about Moderna. What do we do if we have another pandemic in terms of a different response?

• 32:29Speaker 01

  Before the pandemic, everybody thought it takes years and years to develop a vaccine. So after the pandemic, there was no hurry developing a vaccine because everybody thought it takes four years to develop a vaccine. So what's the rush? Of course, there were technology advances that made it possible to develop vaccine in like three, four months. But until people got their head around, why not try? Unfortunately, enough people had to die and enough calamity and economic shutdown had to happen. But people say, you know what? We have the money, let's just try it. And if it doesn't work, then we're nowhere else. That, I'd hope, we don't have to go through the same gestation period of debating. You know, there were a lot of people who said that you can't do any of this stuff. Unfortunately, there's people to this day who kind of think that vaccines can't work, don't work, et cetera. That's a problem.

• 33:12Speaker 01

  But in any case, I really hope that if it happens, there will be the coordinated response of multiple approaches that are thrown at it. I feel very confident that there isn't a life form disease that we can't find an appropriate antidote to, or an appropriate vaccine to, if we can deploy the best technology in the most coordinated way with the right incentives. And by the way, a key thing that Operation Warp Speed did was to set a value outcome for people who tried. Basically, the government said, we will buy X many doses at this price. And so we could go to investors and say, listen, if we can, we don't know what the probability of success is.

• 33:57Speaker 01

  But if we win, there's a clear market signal. That also would go a very long way outside of pandemics. But in a pandemic, that mechanism is tried and true now.

• 34:07Sarah Guo?

  There's a reason to accept the uncertainty.

• 34:08Speaker 01

  Exactly, to your earlier question.

• 34:10Sarah Guo?

  Moderna made many non-obvious bets along the way. One of them was, you know, every, every AI and biotech company I talked to, you know, they want to be a platform versus a, you know, single or two or three clinical asset type company. How do you think about this trade off between early investment in platform versus clinical assets? Because, you know, in a, in a different field, as you and I were talking about before, investor climate and the macro changes around you too. Not every biotech survives. And so, you know, how do you think about that, having been through several cycles of that investor climate?

• 34:43Speaker 01

  Let me answer how we think about it at Flagship. And then let me answer it about how one should think about it, because, you know, we of course only represent one subset. How we think about it is that because we go to far out places looking for undiscovered value, the notion that you do that to come back with one asset is the definition of insanity. Because if you're going to do that, you might as well bet on well-known proven technologies, just a slightly different version, and you could then bet on an asset and hope that your lottery ticket gets pulled. Sorry for being a bit rash, but that's how I view it. But if you're going to go and do RNA for the first time, DNA for the first time, gene writing, gene editing,

• 35:23Speaker 01

  computational proteins, you need to diversify that because you don't know which ones of these are going to get knocked out for reasons that have nothing to do with the underlying technology. Hence, every single of the 110 companies we've been involved with for 25 years are a platform. Every single one, there's no exception to that. So we embraced from day one, but for a different reason. Namely, we wanted to go beyond adjacencies, beyond kind of the reasonable zone into unreasonable things. So that's why we do it that way. Now, if you said, well, why doesn't, why does everybody not do it? Or most people not do that way, even though they want to, as you said. And then the answer is what you said, which is the capital it takes to do one thing is already high. To do two, three things is even higher.

• 36:08Speaker 01

  And oh, by the way, to do a platform that could potentially do 15 things, even if you don't use it for that, is very high. Second, investors don't properly value platforms because they don't give assigned value to the correlated option value of one program that becomes derisked based on three other programs. And so you don't get the credit and you get the deduction because it's now of a sudden overly expensive and overly this and that. The last thing that I'd say the reason investors don't like it often is because they figure it's a strain on management's capabilities. You know, it's one thing to execute one program. But if you're going to execute two, three, four, I fear that that's creating a lot of

• 36:53Speaker 01

  companies that are essentially going to be knowingly dead because that's just a numbers game. And in an environment like this, there's almost like a mass extinction event going on, unfortunately, as we speak, for a lot of these single asset companies accentuated by the recent bets that China has made, both centrally, governmentally, and in the form of results in startups, because they've gone after that same space at a much lower cost, much different barrier to entry in terms of clinical data. And I don't know how you compete in a single asset by a tech company against kind of a Chinese invasion of assets that we're seeing in the business development front. Good for the pharma companies to access those. Those are great assets.

• 37:38Speaker 01

  But remember what I said earlier about commoditization. I don't know how you don't get commoditized at a higher structure of cost that we have here. So that logic brings me back to doing platforms because at least it gives you a chance to do partnerships and to try to kind of come up with ways to stay alive. It's not that our companies are dying. Our companies are, some of them are meeting an unfortunate end as well, but at least some of them are not.

• 38:02Sarah Guo?

  Maybe that represents an opportunity for, you know, the entrepreneurs who truly have a platform and the investors who are able to invest through that.

• 38:10Speaker 01

  Exactly.

• 38:10Sarah Guo?

  Last question for you then, because it's an interesting concept. In your annual letter for 2025, you described this concept of polyintelligence, integration of human, nature-derived intelligence and machine intelligence. Where do you think human intuition will remain most relevant?

• 38:24Speaker 01

  You switched from intelligence to intuition. Intuition is just a, basically a model. So all of us generate our own models and then we use them in short form. So it's actually the closest thing to a computer large language model that humans have. We don't call it a large language model, but it's exactly to me what it is. And so, yeah, anywhere you want to use it, L L F, you could use human intuition. And the problem is the L L M's have a lot more data and they've been trained across, you know, millions of people's worth of data and yours is trained on yours.

• 38:54Sarah Guo?

  Okay, so intuition was the wrong thing, actually.

• 38:56Speaker 01

  No, no, no, no, no, I wasn't criticizing you. I was just taking, picking off from your concept and saying, like, so where does human, let's just say, or human, is that what you mean?

• 39:05Sarah Guo?

  Yes.

• 39:05Speaker 01

  I mean, look, I think it remains to be seen. I describe in this letter. The notion that thinking that the real frontier is between human and machine misses out on the fact that most of science is between human and nature. And now we have a machine that actually can intellectually support our inquiries of nature. So it really is a triangle. It's not a line between human and computer, but it really is a triangle where human intelligence and machine intelligence, coupled with nature's intelligence, will inform each other. And these three actors will adapt to each other. The role of the human in that, I think, is very important. Humans compute completely differently, and they have a way to act that's also quite different

• 39:50Speaker 01

  than the way computers act today and the way other forces of nature act. So I kind of choose to think of it as a three-way thing. That's like a beautiful new axis of emergence. And, you know, what's going to come out of it is the future of life.

• 40:06Sarah Guo?

  I love it. What a wonderful note to end on. Thanks so much for talking to us, Noubar.

• 40:11Speaker 01

  Happy to chat.

• 40:13Sarah Guo?

  Find us on Twitter at NoPriorsPod. Subscribe to our YouTube channel if you want to see our faces. Follow the show on Apple Podcasts, Spotify, or wherever you listen. That way you get a new episode every week. And sign up for emails or find transcripts for every episode at no-priors.com.