Latest / Giant Ideas / “Dive Headfirst”: Isomorphic Labs President Max Jaderberg on his Biggest Decision

Transcript

“Dive Headfirst”: Isomorphic Labs President Max Jaderberg on his Biggest Decision · Giant Ideas

• 0:00
  SPEAKER_00: The world will be revolutionized in terms of how we make medicines with AI, but I don't want that to take 20 years or 50 years to just happen passively.
• 0:07
  SPEAKER_00: Like, we need to make that happen over the next five years.
• 0:10
  SPEAKER_00: Always I've had a very clear goal of like, what's the superpower that I want to create out of this thing for AI and machine learning?
• 0:18
  SPEAKER_00: You have to translate that end goal superpower into a leaderboard.
• 0:23
  SPEAKER_00: But if you can do that, then this becomes a game.
• 0:26
  SPEAKER_00: I work 24-7, whether it's physically or in my head on this.
• 0:29
  SPEAKER_00: Like my work has always been my life and passion.
• 0:32
  SPEAKER_00: We feel like this is a generational opportunity to really change the way humanity looks for the positive.
• 0:38
  SPEAKER_02: Welcome back to part two of Giant Ideas, The Max from Isomorphic Labs.
• 0:42
  SPEAKER_02: Um, this episode, we're gonna talk a bit more about your journey to where you are today, some of the lessons from working with Demis.
• 0:50
  SPEAKER_02: So give us just a brief overview of how you came to end up at Isomorphic Labs, how kind of the spin-out happened from Google, and uh what it's been like working with Demis.
• 0:59
  SPEAKER_00: Yeah, absolutely.
• 1:00
  SPEAKER_00: So um I was you know deep in DeepMind for quite a few years.
• 1:07
  SPEAKER_00: I, you know, I worked on all of these sort of big uh breakthrough projects.
• 1:10
  SPEAKER_00: I was working on like core deep learning and then um generative models and then just scaling up reinforcement learning and you know, smashing all these game domains like AlphaGo, Alpha Star, you know, capture the flag, all of these sort of things.
• 1:23
  SPEAKER_00: Um I've always been most interested to see the impact of all of this stuff rather than just the creation of the machines themselves.
• 1:34
  SPEAKER_00: Um and for me it got to a point where um we were solving like impossible problem after impossible problem at DeepMind.
• 1:44
  SPEAKER_00: These things that like the top professors were like, no, this is 10 years away, and we know we've already solved it internally.
• 1:50
  SPEAKER_00: Um so sure, like we're not at AGI yet, and this is maybe back in 2020, 2021.
• 1:56
  SPEAKER_00: We're not at AGI yet, but we definitely have some incredibly powerful technology that can be used for really interesting applications.
• 2:04
  SPEAKER_00: And I was starting to think, okay, so are we going to be driving on these?
• 2:07
  SPEAKER_00: How are we gonna be driving on these?
• 2:09
  SPEAKER_00: At the same time, you know, in DeepMind, of course, we're we're starting to see the Alpha Fold results emerging.
• 2:14
  SPEAKER_00: So this is like the you know, for us, really the first proof of concept that you can have this deep learning model, this AI system that actually generalizes in the space of biology and help you can for help you for real novel science.
• 2:28
  SPEAKER_00: And so this got to a, you know, came to a point where um there was so clearly now the proof point that we could have a massive impact on drug discovery and more broadly the future of human health.
• 2:43
  SPEAKER_00: Um and this really needed to be its own mission.
• 2:48
  SPEAKER_00: It's such a big problem, you know, that we wanted to create, you know, a separate company that can have its own culture, its own mission, hire the right sort of people and put them together in the right ways to really, you know, maximize the potential uh of the isomorphic mission.
• 3:02
  SPEAKER_00: And so that was really the nucleus of that jumping off point, which was in 2021 where isomorphic started.
• 3:08
  SPEAKER_02: And I mean, Google has a pretty remarkable track record now of taking pure research and you know, sort of spinning it out into companies that are standalone, whether it's Waymo or DeepMind or now isomorphic.
• 3:19
  SPEAKER_02: Are there any lessons for those more technically minded scientific founders listening to this episode of how you kind of take research, you know, productize it, and then ultimately commercialize it into a business that they could you could share?
• 3:33
  SPEAKER_00: Yeah, it's interesting.
• 3:34
  SPEAKER_00: And and and you know, there's there's many different approaches here.
• 3:37
  SPEAKER_00: Um the the one that I've always been obsessed with is, and I think that a lot of these sort of projects have have born out of is the idea of like always be developing for that, you know, for the actual end application you care about.
• 3:53
  SPEAKER_02: So not research for research sake in a way.
• 3:55
  SPEAKER_00: People need to do research for research sake.
• 3:57
  SPEAKER_02: Yeah.
• 3:58
  SPEAKER_00: Um, but I think even if you're doing research for research sake, like think like what are you actually trying to achieve with this?
• 4:05
  SPEAKER_00: Um, your tinkering should be for an end objective.
• 4:08
  SPEAKER_00: That's always been my research philosophy.
• 4:10
  SPEAKER_00: I've done a lot of research for research sake, and I've done a lot of research for application, but always I've had a very clear goal of like, what's that, you know, in my head, I think, what's the superpower that I want to create out of this thing?
• 4:25
  SPEAKER_00: What's that super capability that I want to create out of this?
• 4:28
  SPEAKER_00: And of course, then you, you know, you have an idea of the how this might all come about, which is what gives you conviction that you can actually get there.
• 4:35
  SPEAKER_00: But I think really indexing on that end point is really, really strong.
• 4:39
  SPEAKER_00: And then crucially for AI and machine learning, you have to translate that end goal superpower into a leaderboard, you know, something where you can have a half-decent proxy for are you making progress against that superpower?
• 4:55
  SPEAKER_00: And that's easier said than done, especially in the space that we're in in biology and chemistry and medicine.
• 5:01
  SPEAKER_00: But if you can do that, then this becomes a game.
• 5:04
  SPEAKER_00: You can just grind away at making new models, making new um AI models that improve performance against your leaderboard, and then know in the back of your mind that you're actually making progress against the really crazy superpower we're trying to create.
• 5:18
  SPEAKER_01: I think it is interesting with Alphabet and Google, as Cameron said, you know, some of the most interesting parts of Alphabet or the some of the most interesting things that Alphabet owns or has shares in are outside of the core, right?
• 5:29
  SPEAKER_01: You've got Waymo driverless cars, you've got isomorphic with what you're doing, which just isomorphic alone could be the biggest company in the world.
• 5:35
  SPEAKER_01: If you make good on your mission and you solve all disease, it's going to be the biggest company in the world.
• 5:39
  SPEAKER_01: Waymo is obviously a very big company already.
• 5:41
  SPEAKER_01: Um, how has Alphabet got the balance right between, on the one hand, providing access to the kind of might of the capital and human resources and data that they they can provide?
• 5:53
  SPEAKER_01: And on the other hand, just letting you and the Waymo leadership, for example, just get on with it and be independent.
• 6:00
  SPEAKER_01: Uh, have they got the balance right?
• 6:01
  SPEAKER_01: And and what's been hard to do to do that?
• 6:04
  SPEAKER_00: I I think they've you know got a really good balance here.
• 6:08
  SPEAKER_00: And I think this is actually emerged, this is like a core part of Google culture, which has become alphabet culture, of course, of um bringing like just fantastic technologists and science and people together and giving them resources and the ability to explore what what's what's necessary.
• 6:30
  SPEAKER_00: Um, and you know, that's been a like uh core sort of core part of Google culture from almost like day one, right?
• 6:37
  SPEAKER_00: Those really early days of Google.
• 6:39
  SPEAKER_00: And then I think you see that with the approach to bets.
• 6:43
  SPEAKER_00: And you know, Alphabet, given the size of Alphabet, doesn't take a bet to be a small company, doesn't create this these new company bets to be small companies.
• 6:53
  SPEAKER_00: These have to be trillion-dollar company potential, uh potential to be bigger than you know um many other things.
• 7:01
  SPEAKER_00: And um, so obviously with that comes a lot of responsibility.
• 7:04
  SPEAKER_00: But again, I think core culture of Alphabet is you know, bring the best people together and give them the resources and the you know the frameworks that they need, and then let people create.
• 7:17
  SPEAKER_00: Um, and I think there have been some amazing things that have come out of that alphabet ecosystem, both through the the more formal bets separate company framework, but also, of course, internally within Google itself.
• 7:27
  SPEAKER_01: You you were at DeepMind before.
• 7:28
  SPEAKER_01: Yeah.
• 7:29
  SPEAKER_01: There's kind of quite a lot of debate in the UK and kind of um head scratching about you know, DeepMind sold for under a billion.
• 7:35
  SPEAKER_01: And isn't that a shame?
• 7:36
  SPEAKER_01: And if it stayed independent, we would have had a trillion dollar technology company in in Europe.
• 7:40
  SPEAKER_01: From the outside, I kind of think, yes, but would it have reached its potential had it not been part of the kind of Alphabet family and had all that resource?
• 7:48
  SPEAKER_01: Where do you stand on that?
• 7:49
  SPEAKER_01: Do you think DeepMind could have done it alone?
• 7:52
  SPEAKER_00: It's it's great to be able to say that in hindsight, but you know, back in 20, you know, what was it, 2014, yeah, no one really knew what AI was.
• 8:00
  SPEAKER_00: AI wasn't a thing.
• 8:02
  SPEAKER_00: People didn't really fully understand what the potential of this was.
• 8:06
  SPEAKER_00: Of course, you know, Demis Asavis very much convicted this is going to be uh, you know, incredible.
• 8:12
  SPEAKER_00: And I think to their credit, a lot of Google leadership at that point in time was understanding the potential because they'd had, you know, some pre-work in this space.
• 8:21
  SPEAKER_00: Sure.
• 8:21
  SPEAKER_02: Um in some ways at the original AI company, right?
• 8:24
  SPEAKER_02: With with the original.
• 8:25
  SPEAKER_00: Exactly.
• 8:26
  SPEAKER_00: Back to back back to PageRank.
• 8:27
  SPEAKER_00: Like how do you how do you understand the utility of of different web pages and and and you know, these are these are early recommender systems.
• 8:35
  SPEAKER_00: Um and of course, they they were creating Google Brain um internally as well in in Google before DeepMind um and alongside DeepMind.
• 8:45
  SPEAKER_00: And so, and then of course, as we know, the creation of AI uh requires you know a lot of amazing people and a lot of compute and data, um, which costs a lot.
• 8:59
  SPEAKER_00: And so how are you able to fund that, especially in an environment in a world that doesn't understand AI at all, which is what the world was like in 2014.
• 9:08
  SPEAKER_00: So, you know, I think by joining Google, there was actually the vehicle to realize, you know, what that full potential of AI could be.
• 9:19
  SPEAKER_00: So it's really hard to say in retrospect how how long would it have taken to realize that if it hadn't to have been within Google?
• 9:25
  SPEAKER_02: My my personal view is is no, just because if you look at all the breakthroughs in recently in AI, they came from, you know, let's look at OpenAI, who was a foundation model, foundation model for a long time, is in a nonprofit, and DeepMind developed its technology under the kind of large S of the Google balance sheet.
• 9:42
  SPEAKER_02: I think to do that for such a prolonged period of time, to get to those levels of research breakthroughs without real commercial applications, I think that's really, really hard to do.
• 9:50
  SPEAKER_02: So I think the legacy will be the spin-outs.
• 9:52
  SPEAKER_02: I mean, I think the bigger legacy will be all the amazing companies spinning out isomorphic, all the rest of it.
• 9:57
  SPEAKER_01: And but I think what's interesting is like you benefiting from the capital and the kind of scale of it all, but also retaining kind of startup culture and hunger within Waymore or within isomorphic.
• 10:07
  SPEAKER_01: You were a startup founder and CEO before you you joined.
• 10:10
  SPEAKER_01: Um, how have you retained that hunger when is it almost like artificial, artificial creation of hunger, given that there is this kind of vast scale?
• 10:18
  SPEAKER_01: Anyway, if you want to be in a kind of general alphabet just like at Microsoft or, you know, or wherever else, you can go and have a kind of no ping pong tables allowed.
• 10:26
  SPEAKER_01: Yeah.
• 10:27
  SPEAKER_00: You're allowed a ping pong table.
• 10:28
  SPEAKER_01: Do you have a separate, kind of hungrier, harder charging culture within isomorphic than you would have in the parent company?
• 10:34
  SPEAKER_00: I mean, that's the whole point of starting this as a separate company.
• 10:39
  SPEAKER_00: Separate company, right?
• 10:40
  SPEAKER_00: It has to feel like a startup.
• 10:42
  SPEAKER_00: You have to, you know, um be able to hire different types of people.
• 10:46
  SPEAKER_00: You have to be able to incentivize them.
• 10:48
  SPEAKER_00: Incentivize them, you know, in the way that startups normally would.
• 10:52
  SPEAKER_00: Um, and create a, you know, an organization and a and a body that is almost self-selective for you know the types of people who come in.
• 11:01
  SPEAKER_00: Um, so we're incredibly driven.
• 11:03
  SPEAKER_00: We're incredibly driven by the mission of solving all disease.
• 11:06
  SPEAKER_00: Like everyone who comes in, I think, has you know their own personal reasons to get behind that.
• 11:11
  SPEAKER_00: Um and so um, you know, we've been really uh you know cognizant and conscious from day one to create a culture that is you know really fast moving.
• 11:27
  SPEAKER_00: Um we, you know, really understand the opportunity that we have and the privilege because we have the alphabet backing and the amazing funding and resources behind us, but the responsibility as a result that this is, you know, we feel like this is a generational opportunity to really change the way humanity looks for the positive.
• 11:46
  SPEAKER_00: Um, but it's not going to happen passively.
• 11:49
  SPEAKER_00: I don't want this to, you know, the world will be revolutionized in terms of how we make medicines with AI, but I don't want that to take 20 years or 50 years to just happen passively.
• 12:00
  SPEAKER_00: Like we need to make that happen over the next five years.
• 12:02
  SPEAKER_00: And so it's that sort of you know injection that we I think we have internally.
• 12:06
  SPEAKER_02: And um, you did recently did, I think, a$2.1 billion financing round with uh predominantly with Alphabet, I think Google Ventures and Thrive.
• 12:13
  SPEAKER_02: How did you think about, well, how did that round come together?
• 12:16
  SPEAKER_02: What was the rationale for it?
• 12:17
  SPEAKER_02: And how, you know, how I imagine you're pretty selective in your investors, right?
• 12:20
  SPEAKER_02: What are you looking for in your investors?
• 12:22
  SPEAKER_02: Why did you choose to work with them?
• 12:23
  SPEAKER_00: Yeah, we you know, we had our existing investors, that was Alphabet, Thrive, um, you know, Google Ventures.
• 12:29
  SPEAKER_00: And then in that round, um, we also had CapG come in uh and MGX, Tamasek, and Sov AI here.
• 12:37
  SPEAKER_00: Um, and so that round was um, you know, when we thought about the investors for that, you know, we've had amazing experiences with our existing investors.
• 12:45
  SPEAKER_00: Of course, Alphabet have just been incredible backers from day one.
• 12:48
  SPEAKER_00: They understand the big ambition.
• 12:50
  SPEAKER_00: Um, and you know, they help us every day to drive towards that.
• 12:54
  SPEAKER_00: Um, Thrive have been amazing partners on that, on that journey coming in at the Series A.
• 12:59
  SPEAKER_00: Um, and I think with that last round, we really wanted to get some of these um, you know, fantastic geographies, you know, behind us um with with funds associated with them.
• 13:09
  SPEAKER_00: Uh, this is inherently a global mission.
• 13:11
  SPEAKER_00: We think of ourselves as a global company and we want to impact, impact patients all over the world.
• 13:18
  SPEAKER_00: Um, and so we want to wire that into our DNA from very early on in the company.
• 13:23
  SPEAKER_02: And and Demis recently, I think, stepped down from being kind of CEO of Deep Mind, um, but you have a very close relationship with him.
• 13:30
  SPEAKER_02: You're sort of day-to-day operational, but he's sort of you know very, very involved and isomorphic.
• 13:34
  SPEAKER_02: How is your guy's working relationship and what have you what have you learned from him?
• 13:38
  SPEAKER_00: Yeah, I mean, you know, we've worked together now.
• 13:41
  SPEAKER_00: Um, it's probably like coming up to 15 years.
• 13:44
  SPEAKER_00: Um, and I think that's, you know, it it's been, I've I've learned so much from him, you know, over the years, been very, very inspiring.
• 13:51
  SPEAKER_00: Um, I, you know, I actually remember one of my first uh all hands in Deep Mind.
• 13:57
  SPEAKER_00: This was this must have been in like 2014 or 2015, uh, where Demis was toasting the um uh the publishing of the Nature Paper for the Atari uh uh work.
• 14:10
  SPEAKER_00: And it was like, yeah, this is the first nature paper, but by the way, this is going to be the first of dozens of nature papers, and one day we'll win a Nobel Prize.
• 14:17
  SPEAKER_00: And uh, you know, I was like, Yeah, sure, sure, whatever.
• 14:20
  SPEAKER_00: Um, but of course, like all of that transpired to be to be true.
• 14:24
  SPEAKER_00: So I've learned over the years to like really um, you know, trust Demis's vision.
• 14:29
  SPEAKER_00: I think he's able to think uh far further into the future in a in a sort of grounded and concrete way than a lot of other people are able to do.
• 14:38
  SPEAKER_00: Um and so that's been an amazing person for me to work with um over over the years and now an isomorphic to help you know actually like translate all of that to okay, okay, how do we actually move forwards with this um and and build this amazing company?
• 14:56
  SPEAKER_00: Um and yeah, I think uh it it's been a you know, he he him stepping away from the CEO role at DeepMind.
• 15:05
  SPEAKER_00: Um he you know, he's obviously got a huge amount of responsibility in alphabet as well.
• 15:10
  SPEAKER_00: Um and then for us at Isomorphic, I think having just more cycles um for Isomorphic uh is you know is fantastic for us.
• 15:18
  SPEAKER_01: Amazing.
• 15:19
  SPEAKER_01: Let's talk a little about yourself and your journey.
• 15:22
  SPEAKER_01: Startup founder went to DeepMind, um, and are now obviously running Isomorphic.
• 15:27
  SPEAKER_01: We kind of ask all of our guests if they had to look back on their life at kind of key moments, these kind of photographs or snapshots that would define your life so far or the wisdom that you've accrued along the way.
• 15:38
  SPEAKER_01: What would kind of one or two of those photographs be for you?
• 15:41
  SPEAKER_00: There's probably two images that are both um they're both sort of fork in the road points.
• 15:48
  SPEAKER_00: Um, the first is is probably me sitting at the top of the like engineering science building in Oxford.
• 15:53
  SPEAKER_00: The second is probably me sitting at the top of my house with my wife.
• 15:57
  SPEAKER_00: Um, both were these like big decisions of which direction do I go in.
• 16:03
  SPEAKER_00: Um the first was actually back when I was finishing my sort of master's undergrad in engineering, I'd signed a job contract already with a management consultancy firm.
• 16:15
  SPEAKER_00: Um and so I'd I'd committed to go that route.
• 16:20
  SPEAKER_00: Um, but I was sitting at the top of the engineering science building, Oxford, looking out, and there was the these amazing lectures on Karman filters and you know machine learning.
• 16:28
  SPEAKER_00: Uh I was doing a dissertation project in uh computer vision at the time.
• 16:32
  SPEAKER_00: Um I was just like, this is so magical and interesting.
• 16:37
  SPEAKER_00: Um and I was starting to think to myself, like, why wouldn't I keep working on this stuff?
• 16:43
  SPEAKER_00: Um and I I had my professor, this amazing guy called Andrew Zissiman.
• 16:48
  SPEAKER_00: Um, every time I met him, he he was saying, You need to stay and do a PhD, you need to stay and do a pre-PHD.
• 16:54
  SPEAKER_00: And I was always like, No, no, no, I'm I'm I'm going to go and get a job and and you know get a proper job.
• 17:00
  SPEAKER_00: Um, but this was like a point in time, I think, where um it just tipped over and was like, actually, no, I need to take this really seriously.
• 17:09
  SPEAKER_00: And I did a massive uh sort of uh sort of pro con and and uh sort of list and and ended up tearing up the job contract for the consultancy firm, um, accepted the PhD position.
• 17:23
  SPEAKER_00: Uh I accepted the PhD position on one condition though, which was that I was gonna, I wanted to start a company out of the PhD.
• 17:30
  SPEAKER_00: And my professor had to let me do that.
• 17:32
  SPEAKER_00: Um and yeah, you know, he stayed true to his word and let me do that later down the line, because I was always obsessed with starting companies out of that.
• 17:40
  SPEAKER_00: But I think that was that first image.
• 17:42
  SPEAKER_01: And just on that first kind of journey, I guess, that you've described, what have you done to teach yourself to become uh a commercial leader, somebody who can actually run a company, run a team, lead people?
• 17:53
  SPEAKER_01: Has that been from reading books?
• 17:54
  SPEAKER_01: Has it been from learning from people like Demis?
• 17:57
  SPEAKER_01: Like what if just for the kind of technical-minded people listening to this, what would you recommend and what's worked for you?
• 18:05
  SPEAKER_00: Um Yeah, I think I've always been obsessed with uh, you know, seeing the impact and companies being fantastic vehicles to drive that impact.
• 18:14
  SPEAKER_00: Um reading loads of books and stories about entrepreneurs and uh you know podcasts and videos and and these sort of things.
• 18:25
  SPEAKER_00: And then, you know, honestly, like just you know, building stuff and trying to make it work.
• 18:33
  SPEAKER_00: And uh, you know, I I built countless websites and businesses around those and you know, trying to monetize those and uh over and over again, and and slowly you start to just get a feel intuitively for what your own style is in this space.
• 18:48
  SPEAKER_00: That's when I was you know much younger.
• 18:51
  SPEAKER_00: Um, and then over the years, like really learning from incredible people, you know, firsthand, people like Demis, people like my old boss Korai, who who now runs Gemini, um, coaches along the way, mentors along the way, um, and just constantly being open to and and self-reflective, I think, on on what works and how you want to operate as well, because everyone's style is different.
• 19:15
  SPEAKER_00: Yeah, that's for sure.
• 19:16
  SPEAKER_00: Amazing.
• 19:16
  SPEAKER_00: What would be the second image?
• 19:18
  SPEAKER_00: The second image um would actually be um in my loft with my wife uh on a whiteboard with a again a big pro-con list of like, am I gonna dive headfirst into this crazy isomorphic labs idea?
• 19:34
  SPEAKER_00: Um or stay in sort of core AI research.
• 19:39
  SPEAKER_00: And, you know, this was the point in time where uh I could see, you know, again, like we've solved the impossible problems again and again.
• 19:48
  SPEAKER_00: I was kind of getting bored of just solving impossible problems for the sake of to prove to the world that AI was a thing.
• 19:54
  SPEAKER_00: I want to see it have impact.
• 19:56
  SPEAKER_00: I could see large language models brewing, you know, this was pre-GPT.
• 20:00
  SPEAKER_00: Um but you know, we were obviously working on on the precursors of these things, and I have very close friends who've who've who've got who've been the core developers of of these models.
• 20:10
  SPEAKER_00: Um but fundamentally I think I was most excited to see the application of all the work that I'd you know poured into core deep learning, go into go into medicine and changing human health.
• 20:23
  SPEAKER_00: And so I think that second image was just really like trying to rationalize this, probably rationalize what is probably quite an emotional thing at the end of the day of just like what excites you the most.
• 20:34
  SPEAKER_00: Um, and I think actually for both of those images, it's it's been that thread of following what excites me most versus maybe what's you know more uh if you looked outside, what's the more common uh angle to go down?
• 20:49
  SPEAKER_02: Was your wife supportive of the pro to go join an isomorphic?
• 20:52
  SPEAKER_02: And do you have a picture of that pro and con list?
• 20:54
  SPEAKER_02: Because I think that would uh age quite well.
• 20:56
  SPEAKER_00: Yeah, I wish I I I probably did take a photo, yeah.
• 20:59
  SPEAKER_00: Um no, my wife was was obviously so supportive, you know.
• 21:03
  SPEAKER_00: Um she wants to, yeah, she doesn't want me to be grumpy at home going back from work.
• 21:07
  SPEAKER_01: So um one question about the journey.
• 21:10
  SPEAKER_01: We've heard from quite a few of our guests on John ideas recently, this idea that they've had to basically fully prioritize work and family and maybe a bit of health, although that slipped quite a lot for a lot of them.
• 21:21
  SPEAKER_01: And a lot of them seem to have sacrificed social life essentially.
• 21:24
  SPEAKER_01: Um probably the most extreme example was the founder uh Kesar Unis.
• 21:29
  SPEAKER_01: Um, and he basically said, Yeah, I have I've decided to have no friends because I just want to focus on my company.
• 21:34
  SPEAKER_01: What's your take on that?
• 21:35
  SPEAKER_01: Do you do you manage to have some sort of social life, some kind of calm away from work?
• 21:39
  SPEAKER_00: I think I do.
• 21:40
  SPEAKER_00: Yeah.
• 21:40
  SPEAKER_00: I mean, I I work 24-7, whether it's physically or in my head on this, like my work has always been my life and passion.
• 21:48
  SPEAKER_00: Um, because it's to me, it's not work.
• 21:50
  SPEAKER_00: It's like what we're able to just do in the universe as a you know, as a as a species, and and you know, being able to build this technology and see it being used for good is just so exciting.
• 22:00
  SPEAKER_00: Um and, you know, my family is everything, you know, both my immediate family, my wife, my kids, and my also all my close family, my siblings, um, parents, we spend a lot of time together.
• 22:15
  SPEAKER_00: Um, I I don't see it as like, and and my friends as well.
• 22:18
  SPEAKER_00: I don't see it as one or the other.
• 22:19
  SPEAKER_00: It's all like creating that full spectrum environment that every individual needs to thrive.
• 22:26
  SPEAKER_00: Yeah.
• 22:27
  SPEAKER_00: Um, different people, you know, sit on different needs different amounts of different things.
• 22:32
  SPEAKER_00: Um, you know, one weird thing for me is I I need time by myself to just like disconnect and just reset part of my brain.
• 22:42
  SPEAKER_00: Uh, I always have done.
• 22:43
  SPEAKER_00: I still make time to play Counter-Strike.
• 22:45
  SPEAKER_00: It's weird, but uh my wife takes the piss out of me all the time.
• 22:49
  SPEAKER_01: Um, but it's you know It's very important for the future of humanity that you are able to do that.
• 22:54
  SPEAKER_01: And I'm also very glad that someone who has such immense power over you know human health is someone who is a bit more balanced than some of some of the other leaders you hear from.
• 23:02
  SPEAKER_02: Give this man a counter counter-strike subscription for the rest of his life.
• 23:07
  SPEAKER_02: I I also I heard from a little birdie but that demisils are still uh regularly sending emails to a m.
• 23:14
  SPEAKER_02: They're back on again around 6 a.m.
• 23:16
  SPEAKER_02: Yeah.
• 23:17
  SPEAKER_02: True, true or false.
• 23:18
  SPEAKER_00: Um that happened last night.
• 23:20
  SPEAKER_00: There we go.
• 23:21
  SPEAKER_00: We like we we you know he is he's we went like yeah, he he his sleeping pattern I I can't pin down.
• 23:28
  SPEAKER_00: Yeah.
• 23:28
  SPEAKER_00: Um and sadly I'm not on the other end of the spectrum.
• 23:32
  SPEAKER_00: I need a lot of sleep uh to feel normal.
• 23:35
  SPEAKER_00: Um and we ping all hours of the day all the time, which is quite interesting.
• 23:40
  SPEAKER_00: Yeah.
• 23:40
  SPEAKER_01: Extremely reassuring, again, to hear from a normal person who does actually sleep.
• 23:44
  SPEAKER_00: It's sort of not one of these uh Well, I I I I I take solace in Einstein, who is known as the sleepy genius.
• 23:50
  SPEAKER_02: There's definitely models of both working.
• 23:52
  SPEAKER_01: So Max, thank you so much for sharing a bit about your journey and about the giant idea of solving all disease.
• 23:57
  SPEAKER_01: It's it's really exciting, and I think everyone should be rooting for you.
• 24:00
  SPEAKER_01: Thank you.
• 24:01
  SPEAKER_01: I appreciate the time.