The Next Phase of Human Evolution Is an Engineering Problem

Are we all going to die because of AI? It's a question that is top of mind. What does the future look like with us and AI? Are we becoming something different? Or are we no longer going to exist? Jonathan, my co-founder, just wrote a 10-part essay called "The Misframing of Longevity," which really talks through this thesis that, hey, there can be a positive and pragmatic view of the future We just need to invite people into that story and build it together. And that's exactly what we're doing today. Here on our team, we build Clarity. We solve problems around, does a system actually know who you are? Not a static profile, but a dynamic one, a living model that learns your goals, your patterns, your trajectory of becoming. Jonathan wrote this essay, and he is very much inculcated in neuroscience, computational biology, the field of longevity and all the different forums and folks that are experts in that space. He's really close to that space. And he put out this paper that I read, and that is exactly what we're going to scrutinize today. So for today, welcome to the Self Aligned Podcast. We bring on founders, researchers, builders of the future, people doing big things and big stories to tell. I'm very happy and grateful. Jonathan is my CTO and co-founder building out Clarity. And what we're doing at Team Epistemic Me. Today is a thought leadership piece where we're going to talk about the misframing of longevity. So Jonathan, tell me about the misframing of longevity. I know there's tons of PhD papers that have fed this synthesis of your writing to create this more positive, pragmatic view of what is this next phase. Yeah, uh, so if you could just talk through that and how— maybe like outline it at a high level and then go into how did how did you even get to these thoughts? And the before-after of you had a pessimistic view, and you wrote this series, and it gave you a very positive view.

Tell me about all that. Yeah, I guess that's the first thing I will say. Since I wrote this essay, I felt so much better about the world that we live in, about reality, how things might evolve, and what a hopeful future could look like. So I'll start there. Yeah. And I think we're okay. When I'm looking at, like longevity, uh, it's not like a lot of people think like, oh, you're trying to stay alive forever or something like that. I'm really looking at like, how does humanity and technology co-evolve, you know? Like, you know, we used to live very short lives before we discovered things like, uh, you know, doctors cleaning their hands and vaccines and stuff like this. And that, that was technology meeting humanity. And now that technology is, you know, accelerating so rapidly, uh, we have this question of like Well, you know, how are we going to co-evolve? And what are the historical patterns that have, you know, that we can use as a lens to understand how this will come together? And I was at some longevity conferences last year where I saw the longevity roadmaps where, you know, like, well, aren't pharmaceuticals plus AI just going to save us all and like, no diseases? And it's not feasible at all, actually. Um, and, and so I think every version of like, uh, repair or replace as, as like a way to sort of like maintain health has like catastrophic failures where like the evolution of technology won't fix those problems.

Um, like there's still going to be an expiration date no matter how many drugs or repairs you do. Yeah, because that's the system, right?

Right. Like you look at, uh, there's this idea called Eroom's Law in pharmaceuticals, the opposite of Moore's Law. And as time goes on, it gets more and more expensive to develop pharmaceutical solutions. And so what is AI going to just accelerate its end, or is it really just going to fundamentally shift the game where Eroom's Law goes away and there's some other law that appears, some other ceiling? Yeah, yeah. And in this, I, I had much sort of strong opinions about like where, um, our ideas of— it's our ideas of what is a human that really come down to, like, you know, how we define life and how we extend life.

Everything comes from that. Yeah, like what does it mean to be human?

Fundamentally a philosophical question. Yeah. And I think one of the favorite bodies of work I've been pulling on recently is sort of like a consciousness-first notion of what it means to be human. When you go to your doctor and they look at you and you are a straw man who has a disorder, whether they apply a drug therapy or surgery therapy to like fix that disorder and get you back to like being like the perfect human, I guess, right? And that metaphysical view is important. It got us to a lot of places, but it just sort of doesn't work, and it breaks down. We have to deal with the subjectivity of being a real person, or being any kind of real being really has like deep subjectivity. And I think, you know, I'll point to Michael Levin in particular as helping me evolve my own thought models to understand like, uh, you know, that we, we are embryos, right, at the very minimum. I mean, we're recapitulating like billions of years of evolution. We're part of a continuing part of our solar system.

Say more about that. We're recapitulating billions of years of our Yeah, say more about that. What does that mean to somebody that isn't into biology or anything? What does that mean?

Yeah, I think like, to me it means you didn't come— you didn't come from nowhere. Like, the world existed long before you, right? And, and all these lives existed before you, and ancestors existed before you. They learned patterns that were encoded in, like, in— they found evolutionary fitness and were passed on to you. You're the result of billions of years of cellular evolution, of mammalian evolution, of primate evolution, of human evolution. And when you first, you know, your lights come on, when you're conscious, that's the result of this continuing process. Yeah, right. And then through your own embryology, you start to— your cells express, like, you know, your epigenetics, and they find fitness for, like, how should, how should everything we've learned from this point You know, from the past, yeah, to the future, right?

Yeah, yeah.

And so when we talk about like, what does it mean to be human, and what is a self, and what is subjectivity, um, you know, well, yeah, there's, there's a, there's a deep context for it. And I am much more than like a set of disorders that, that show up on a pharmaceutical commercial. Yeah.

Um, are you sure though?

I mean, um, I don't think, I don't think the working model of even like most highly educated doctors, biologists, physicists account for that at all.

Yeah, subjective being.

Yeah, right, right.

I totally feel this. Like when you go to— actually, we were just talking about getting concierge primary care memberships because of, you know, basically how not great healthcare— yeah, the shortcomings of healthcare today and the lack of a subjective lens on the human experience for the patient.

Yeah.

And it's more like, here's your chart, this is what you do, this playbook is what the health insurance tells me that I'm in the bands to make a prescription for, or whatever, uh, versus like, hey, who are you? Like, what are your goals in life? Like, whoa, okay, you want to be an ultra runner? Oh, and you're trying to— okay, like, well, based on that becoming— yeah, here's maybe the steps you should take now, right? Is that what you mean?

And I think like when we ask ourselves the philosophical question this day, like, you know, how do, how do we co-evolve with, with technology? What does it mean? Yeah, what did, what are the humanities become and how should they evolve, right? Like, um, you know, these are a lot of hard questions if you don't have the right lens. And so that to me is the right lens. And when I was attending some of these longevity conferences last year, I kept, I kept seeing like, you know, presentations.

We went to Vitalist Bay, uh, Don't Die, Don't Die Summit.

Yeah, some conferences around the Bay Area.

Yeah.

So like the big ones and most forward-looking thinkers. And what you get is this sort of like, first there's excitement of like, people are working on this and this is a real problem and there's investment into it. Cool. And then you learn and like, oh, every roadmap is like a dead end.

It's pharmaceutical.

There's no, there's not like a hopeful vector.

No. Yeah.

Yeah. And so I don't know what I was carrying around in my mind during like, I remember Vitalis Bay in particular was like, how are we meant to co-evolve? And I kept thinking about, like, you know, um, how have we evolved so far and how is technology going to evolve that? And I saw one video this year that just— they put the pieces together for me, and it was called Consciousness Engineering, uh, by, by this entrepreneur named Max Hodek.

Okay.

He was one of the founders of Neuralink, and he's got a new company called The Science Corporation. Uh, they're very well established. They have $230— they just raised $230 million in additional funding. Series D. Um, and, uh, the idea that he expressed, amongst many other things that are really interesting, the idea that just really set me off was this idea of implanting a layer of synthetic neurons and grafting it onto our existing neurons so that, like, not only can your neurons communicate with these synthetic neurons, but the synthetic neurons can talk to computational devices. Yeah, right. And, and all of a sudden, like, you know, so, so Michael Levin describes, uh, like the, you know, the embryo is, is these kind of like germ layers, right? Um, well, he's not the one who initially— like, germ layers are a very old concept, but he takes that and he's— yeah, what he gives us is that like each of the germ layers is, is a communicating tissue system. So our nervous system sends ions to communicate, but, but so does our mesoderm and ectoderm and endoderm, and they're all just communicating layer systems that, that work to communicate with each other and and with each of the other germ layers.

Yeah.

And, um, you know, if you study, uh, especially I think starting with, you know, like, uh, reptilian and mammalian evolution, you can see the role the nervous system has, a profound, um, you know, shaping of the adaptation landscape of things that find fitness. And, uh, especially in hominids and through humans, almost all of the major, like, adaptations we have are driven by the nervous system and its interaction with our environment, including our culture. So, you know, I thought like, okay, what if this synthetic, like, these synthetic nerves could actually be like the 5th germ layer? And so that's why I wrote this essay. It's called The 5th— it's on a site called The 5th Layer. Yeah. And it's called Longevity as Metamorphosis. And we're actually adding this 5th layer onto our human embryonic system as a way to have sort of a human exoskeleton or exo— exonervous system, I guess.

Yeah, that's fascinating. So that's the take on, okay, these are patterns we've seen in how we've evolved down to the cellular level to pass information so that we find evolutionary fitness as a species. And hey, like, looking at that pattern kind of makes sense. We kind of go cyborg in the mind. That's kind of like what I'm getting from this. Right.

Yeah, I think the cyborg idea, I think you can think we'll have the ability to have neural engineering procedures where you have a BCI implanted, or you think about any part of what's BCI again, brain computer interface, brain computer.

So the thing that is the exoskeleton I'm making, I can make sense and talk to the outside world from here, compute it, synthesize it, and pass that data to my biological cells and neurons.

And you could, you could use these for some amount of communication with other BCIs, with other humans. Um, and you know, a lot of this is very, very early, and the, the amount of information we get through BCIs is very limited. Uh, but it's also real. And you have companies like Neuralink famously, you know, uh, kind of maybe at the head of the pack in terms of funding and research. But, um, yeah, Max Sodek and the Science Corp split off and they created their own company. And I love his roadmap. I just think It's a very strong roadmap.

So what's on the roadmap? I love—

well, I, I like how he describes, uh, because so right now, like, our traditional medicine, uh, interventions include pharmaceuticals and they include surgeries and include some other things too. But for the most part, if you're getting insurance to pay for something, you're getting a disorder diagnosed and you're getting one of those things, right?

Yeah.

Um, so what Max Hodak adds to that is, uh, is engineering or neural engineering therapy. Meaning that like we could have BCIs implanted to solve specific problems. And what they're actually selling today is a chip that goes into your eye, a very small one, and it solves macular degeneration.

Interesting.

For people like if you're above 80 years old, this is a very common thing. They're getting paid by insurance to install these. It is working with a very high efficacy. Wow. And so it's like both works as a business and it works as a BCI. BCI.

So it's not just vaporware, it's actually real. Yeah, yeah.

I wouldn't— it's actually not technically a BCI, but it is a, like, you know, they're synthetic cells that are interacting with your cells to solve a problem. Um, and then on that, later on the roadmap, they have that sort of like— it's a little disc with synthetic neurons on it that you graft into the brain.

Interesting.

He's got a specific plan for it. I think it's like your 14th cortical nerve, uh, to, to like have synthetic neurons. It basically becomes a new sensory organ. Um, yeah, the engineering of it is something I want to dive into more deeply, but like the fact that you have somebody who's, who's very, very real, very well funded, making solutions like this, uh, and he says consciousness engineering is just an engineering problem now. So all of these, these, uh, neural engineering therapies will become, uh, industry like the pharmaceutical industry where you have, uh, point solutions.

Yeah, you have lots of different eyes, whatever. Like people tackling those as domains. Yeah, that makes sense. That makes a lot of sense.

Yeah, think about that at scale now. Think about like, you know, oh, I need to go get my nervous system tuned up, like, to do X, Y, and Z. Yeah. And you can control, like, you know, how the synthetic layer interacts with your, your nervous system to a very high degree of accuracy. You can pull up your phone or whatever interface you're going to use, and you can see your body. Yeah. And like, on all the details, you know, that you can't see about your body right now.

Dude, it's like speedy oil change, but it's like Neural change, speedy neural change. Let me just— I gotta go to the— I gotta go get an engine tune-up real quick, right? Yeah. Wow, totally.

And I think it's like a certain phase of it is that you will like— you'll be— have a managed neural layer, and, uh, that this will solve a lot of health problems for you. It'll also give you abilities that you might not have had before. Um, these could be things like proprioception in terms of like being able to control your body. Oh, cool. There could be athletic performance improvements. Further down the line, you're going to see things like, you know, ideas that come through the brain and higher levels of thinking that might be able to be communicated at a bandwidth that we can like, yeah, understand each other's thoughts as a communication channel.

Yeah.

But you take this thing to its like logical conclusion, you say like, then you sort of have this ability to, you know, self-evolve your entire body, like yourself, through this fifth layer, through your nervous system, through all the other germ layers. Like you have You have control levers.

Agency.

Agency. Full agency down to the cellular level. Down to the cellular level. And like, we'll all be talking to each other. So when I take on an experiment of neural engineering and I try to change my body and I get an outcome that is efficacious, I can communicate that to other people that are like me and they can say, let me run that same experiment. And so like, you can see a population of humans like co-evolving with technology. In a way that, like, you know, they're more resistant to existential threat, that they're more capable, more productive, uh, better able to communicate and understand with one another. Um, and, uh, I don't know, I'm liking this vision a lot better than, like, AI overlords, you know.

You sure that you don't want to be an AI slave, like, to some robot, like, that predicts the next token? Like, are you sure that you don't want that? Totally.

I think that's it. We're bringing AI into our into our bodies through like a fifth layer nervous system exoskeleton.

Yeah.

Exo nervous system.

Yeah. And that is such an interesting vision because one of the things that I always think about when it comes to people is just how hard it is to communicate information, like to make meaning and then make that meaning go over to your brain to make the same meaning, so hard. And that's also, if you kind of in a meta way think about it, how many times, like I just, for example, very real human story. I didn't really know iron anemia was a thing. I just started ultra running last year. I'm vegan. I just found out I've been iron anemic. And I only found out, this is such a dude thing, but I like started having small like brain fog, depression cycles. And I was like, I mean, I'm starting a startup. And I'm exercising a lot, like I'm pushing my body a lot. So I was having weird issues. And then one day I like pulled some hair out and I was like, oh my God, what's this? Do I have cancer? Am I dying? And I started like talking to AI and being like, here's my stats. Like this is what I do as a human. I'm vegan. Here's my health stuff. Here's my health stuff. What might this be? Here's some other stuff I've been going through. And it was like, you probably have iron anemia based on all of this.

It knew that right away.

Yeah. It was like, well, based on this, This, it's really like the fact that you can even exercise means like you're generally healthy.

Yeah.

And you're getting stronger, which I was. And that's why I put off, you know, really looking at it. I was all like, I'm running better, I'm climbing better. Anyway, and if I think about 50 years ago, I would've just maybe wasted away or then maybe I went to a doctor and maybe they knew what it was or they referred me to another doctor who's maybe seen it. But today there are 2.3 billion East Indians. Ebooks worth of data being created every 10 seconds, right? That's a stat I really like. Many people don't even read a book a year. And so there's so much information out there. How do you find signal that's relevant to you? Like me to become myself, who is not iron anemic, who is healthy. And so AI in that very real use case helped me make sense of things, make meaning, and then make an action that's more aligned with me. And I didn't have to like go to a doctor and then go through insurance bullshit, and then maybe get to a specialist who then maybe gets me the right blood test or whatever. I was able to be more, you know, high agency about it. Okay, so my point of saying that is that's the right now state. The 100 years ago state was much different. And then it sounds like even the 10, 20, 30, 100 years later is gonna be, oh, it's just collapsing making meaning. Once I have problem, Oh, okay, well, like information flow and fidelity of information from cells to cells, neurons to neurons is a problem. But the parallel problem is human to human. It's like, what is iron anemia? Like, what, you know, how do I make sense of it? But what if in the future with what you said with this fifth layer, we've made sense of it as the data shows at the cellular level. When my cells lack oxygen, there is a flag that says maybe check into anemia. When my Robert cells say that for me, For my fifth layer, it's like, well, Robert's a vegan ultra runner. It's probably most likely to be this.

Right.

Right? Versus I had to right now go make sense of a bunch of stuff with AI and that's better, but the future world could be really proactive.

Right.

Right? Like really aligned because in that vision, the thing that stands out to me is there's a lower entropy in the information flow. Because of that, you can have higher agency.

Yeah.

Yeah.

Yeah.

I don't know if that resonates, but that's how I'm sort of piecing it together in my worldview.

Yeah, I think it's more information. It's more meaningful information. Yeah. 'Cause I mean, think about what your doctor has access to today through a blood panel, urine panel. Even if you're on top of your game and you're doing function health and you're testing your blood as frequently as possible, it's still not an accurate representation. Your hormone levels, for instance, fluctuate throughout the day.

Drifts.

They're just complete mysteries to start in terms of how— yeah, yeah. Whereas like with, with BCIs and like a true fifth layer kind of scenario, you're starting to model and map in real time the homeostatic systems of the body. So at the point, like, you call it iron anemia, but guess what? Like, there's probably a number of different homeostatic systems where it shows up that are responding to the fact that you might be lower on iron and they're adapting accordingly. And there's a whole, like, you know, breadth of information and depth of information that we're just lacking right now.

And all those logs, I can't even see them. Yeah, that's a good way to think about it. This is just uncover all the observability within the body. Yeah, germ layer, get all the traces.

Telemetry is—

yeah, instead of Datadog, it'll be data self.

Yeah, well, yeah, and like we talk about understanding the self, right? What does it— what does it mean? And you you get to like, you know, the logical conclusion, if you push it forward, how deeply can we understand ourselves? You realize we're just a fraction of ourselves we see today and how much more empowered we could be.

Tell me more about that, because I think, you know, I was just having a conversation with another AI founding engineer and he was talking about epistemology and subjectivity because he came from a learning sciences background, this Ariel. And then he moved into AI security. And so he has a, he was telling me, you know, there's cases where CTOs, CIOs, like they have to analyze risk differently than say like a congressperson will analyze risk differently than like somebody else, right? And so everybody's profile of things is subjective. And then we started nerding out about ontology, epistemology and stuff like that. But I think for the purpose of our conversation, most interested in about is this articulation of self. But it's something that most scientists don't really talk about, which is that informs your science, right? Yourself, the lens in which you view the world subjectively, informs how you try to find truth, right? Right. And I don't think most people are thinking about it that way.

So I want to hear more from you about that at the intersection of some of this work you're talking It's, it's a, I guess a lot we could talk about. I like to look at it through the lens of metabolism and metabolic health.

Yeah.

Yeah. And I think metabolism is, uh, is a model of how our homeostatic processes are working to get energy, you know, and to, uh, take our metabolic systems, our homeostatic systems are taking agency to manage metabolism and they're actuating their environment. And that happens at a cellular level. It happens happens at a tissue level, happens at an organ level, happens at a system level, happens at an organism level, right? And so there's like a kind of a congruence of information that happens there. And like, you know, if whether we're talking about health or mental health, the amount of like our metabolic context, like say I'm really hungry, is going to like give me one, you know, kind of view on whether or not I enjoy the media that I'm looking at right now or the human interaction that I'm having right now. Um, it matters a lot if there's like, you know, glucose getting to your brain or ketones getting to your brain. Yeah, these are subjectively different qualia, different feelings of how you're going to interact with the world. And, you know, if you're, if you're going to enter into like a state, like a metabolic state where your cells are actually processing different fuels, like, like a ketosis, or you're downregulating IGF-2 or something like that, um, you know, then your whole sort of like your whole metabolic context with the world shifts a bit, and it really matters to like how you experience things and how your perception— yeah, yeah, yeah, totally. Um, and you can feel it. I have friends who, you know, like might have gone into ketosis for the first time and they say, Jonathan, all the noise stopped in my head. Like, because they just like the, the constant, uh, glucose, uh, fueling of their brain was— it had a certain feeling to it, and the ketogenic had a certain different feeling to it. And not to say that either one is better or worse, but there's there's like metabolic context.

This is different.

Yeah.

Different quality.

And one thing that's really interesting about pharmaceutical studies is they don't use metabolic context in order to like contextualize the hypothesis of a randomized control trial, for instance. Yeah.

Yeah.

There's no science that sort of relates this functional medicine of metabolism to the, you know, the world of like clinical outcomes. Completely missing information, completely missed opportunity to understand the human condition.

Totally. In a very real way in the longevity health space, a lot of my female friends say this, is there's almost no research done on nutrition for women or fitness for women. And that's just like the tip of the iceberg is just gender differences. But what about all the other differences, right? All the other heuristics that make up what a human subjective being is, like down to their beliefs, their cultural context, how they were raised around food, sleep, exercise. Sleep, diet, exercise. And that, just for the pod, what we're really outlining is there's so much to modeling the human and subject experience, subjective experience that then ties to metabolic health, that then ties to, you know, whatever personalized solution that you might have that helps you, I should say, helps you be agentic in your own problem solving. Like if I were a woman, and I only had the information of today's, well, how do I operate to maximize my longevity? I would have to go to the cutting edge because there isn't that much information out there relative to the corpus of information we have about our particular archetype of males. It's not even like all males that we've like sampled. It's like, you know, particular archetype of males, right? Like white males. So yeah, just adding on that sort of example to show the strata of where we're at versus where we could really be in the future with this fifth layer vision, where we get down to those very subjective heuristics at the human level all the way down to the cellular level.

Yeah. Yeah, I think it gives an interface for that personalization and for that subjectivity. Yeah. Yeah.

What would you say to the people that would be like naysayers to this though? What would you say to those people? Because I'm sure they're listening.

Yeah. I think in one very real sense, the world's changing no matter what. Yeah. So like, it's just a question of like, what's the world, the future that we want to create? Um, it's as simple as that. And I think that a lot of the future is things that we as individuals may not have control over, such as like the way that the military is going to adopt AI for warfare, right?

Like these things, or surveillance, like whatever, right?

Right. Yeah. It's— I mean, these things like we're not going to have short-term control over these, and we might have longer-term control through cultural change and through the propagating ideas that help, you know, regulate our governments. But yeah, um, you know, so I, I say first is like, think about like what's the most ideal future that you want to create given that change is, is a given.

Yeah.

Um, and then I think like, think about, uh, what sort of advantages Like when you see somebody else who can read 100 times faster than you, are you gonna be able to want— are you gonna wanna read like them? When you see someone that can run 500 miles, are you gonna be able to wanna do that?

Yeah. Yeah. That sounds great.

That was for you. Yeah.

Climb like V20 or whatever. Yeah. Yeah, for sure.

And I think like, that's the way to think of this is like, you know, like what are the— when I— when it— when we co-evolve as like groups or cultures of humans and we have extended productivity or efficiency in a certain direction that gives us economic viability or better lifespans or whatever, um, that like, you don't want that to be Elysium. You don't want that to be like a group of people— Elysium, uh, a science fiction movie where like all the wealthy tech-enabled people live in space and they circle around Earth and all the poor people like, you know, —like a caste system, basically.

Yeah.

Yeah. I think, you know, and this happens already, right? Like, you know, there are— it happens— something like 800,000 pygmy, you know, pygmies in the jungle in Uganda that have just opted not to join society. Like, they want to be hunter— they want to be foragers. Yeah. And there are lots of tribes who've opted, and there are levels of people who have not joined into— they live a rural lifestyle instead of a metropolitan lifestyle. Yeah, right. And that's always—

it lives on a spectrum of choice.

So I think the spectrum continues, and we'll, we'll see, like, you know, a group that increases agency, a group that, you know, they work on their adaptability. But, you know, what happens when, uh, all of a sudden the advantage that you get from adding this fifth layer, exoneural level, is there— is, is going to, um, give you 200 to 500 years plus of life?

And that's evolutionary fitness. Yeah, like, so then do you self-select out of evolutionary fitness?

I think that's where you're getting to, Yeah, I'd say, you know, like, I want to be on that high agency bandwagon. You know, if you, if you want to live a life where you live a more conservative viewpoint, you want to live and die as your ancestors sort of did, uh, you know, before, that's, that's your privilege. But guess what's going to win out over the long term? Yeah. Um, that's, it's evolutionary fitness. And if you think about like the, like the life cycle of a knowledge worker today, they might get 30 years in their trade to like get good at their work. Yeah. Think about somebody who has 300 years and all of computational power. Like, think about like what they're able to accomplish, right?

That makes so much sense to think about it that way. Yeah. Because to do anything worthwhile takes so much amount of time and effort and attention. Totally.

But if you, you know, it compounds. Yeah.

Like you don't become a master at your thing one year into it. Right. It's every year compounds and you skyrocket over time. Yeah. But you have to think about that. You could increase the amount of time and tenure you have to gain expertise. 'Cause I love mastering things, or I just like optimizing things. So that's like, sign me up. Waitlist, where is it? Great. Got 300 more years to learn things and be my best self? Hell yeah. Yeah.

And hopefully it's not like immediately self-selective, but you can imagine a battlefield with BCI-enabled soldiers versus a battlefield of not. I mean, and it's just really—

Is it because you're, Saying like we'll still be fighting wars because that this will be the scarce resource is like getting to that technology.

Um, I think like, well, there seems to be geopolitical situations that will create the need for warfare for the foreseeable future.

Okay, let's go down this rabbit hole for a bit.

Well, tell me more. Maybe this is in the decades of human future, maybe it's in the centuries of human future. It certainly has been a millennia of human past. Yeah, right. Um, and so like the military is obviously one of the people on the frontier of brain-computer interfaces because they can extend the OODA loop, the Observe, Orientate, Decide, and Act, onto a group of soldiers who can share information on the battlefield with drones and equipment. And you have like sort of a, uh, a fighting organism that is far— it has a combined arms warfare capability that's— that can adapt more quickly than in real time. Yeah, in real time.

Yeah, that's wild. Yeah, because the information maximization—

yeah, it makes a lot of sense. So that means the other side has to do it just to keep up, right? Yeah. And so you've got this kind of like, you know, warfare dynamic that's pretty typical in combined arms warfare. But, you know, like, there's, there's dystopian ways that could evolve, and then there's, there's ways where that, that technology gets propagated through a population so that they can gain the same benefits.

Tell me more about that. Because this is all about the positive side. Yeah, like what's your vision for that? Like at the geopolitical level, like then down to the individual level, we were just speaking about after we went down this rabbit hole, which thank you for entertaining, is how does one see themselves and what they want out of the future? Like if you know this, the way the world's going, no matter what you like or not, there's a billion people using ChatGPT that It's not going down for sure. Well, maybe temporarily with them selling us out to the government and more people going to cloud, but overall AI adoption's going up. That's not denied. So if I were a person trying to navigate life and let's say, I got a family, I got my 9-to-5 job, what advice do you give them to have hope for the future?

Right? Yeah, you mean short-term or long-term?

Short and long-term.

Yeah. Yeah. I mean, the world we live in, right? We try to stay at the front of the technology cycles. Yeah. Learn the new tools, the capabilities. It's about augmenting agency. So ultimately, you're going to want to adopt the tools that give you more choice, more optionality, you know, whatever your optionality is today. And I think that's like, you know, you don't want to be on the side where you have limited choice because you have less capability. Yeah. You're pigeonholing yourself and you're making yourself like subservient to a world that's evolving on top of you, you know?

Yeah, it's a lower agency play. Yeah, yeah. And that's a choice, you know?

It is a choice. Some people want to be in that.

That's totally fine. Yeah, not for us. That's not why we do this pod. We're about self-alignment, baby. Right, right. That's not self-aligned to my best self, for sure. Okay, so tell me more about your future best self. You personally? Like, imagine you now, then maybe like, let's say 30 years down the line, we unlock this fifth layer.

Yeah.

Like, what's this look like next 30 years for you?

Well, I think about existential threats and existential risk. Yeah. And how can I mitigate that? Like, you know, I'm a very healthy person. Yeah. Like, I do pretty much everything you can do.

Also real quick, for those that don't know how healthy Jonathan is, my co-founder is ranked ahead of Bryan Johnson on his own app. For his streak, his gamified streak for sleeping and dieting, or sleeping and exercising every day.

Yeah, sleep and exercise.

Yeah, how many days are you at? Like 400-something streak of like 530 days of perfect, for the app, in Bryan Johnson's app, perfect sleep and exercise and amount of steps or whatever. Yeah, yeah. Calories burned, amount of steps and sleep. Anyway, so yes, Jonathan's super into health. Go ahead, Jonathan.

Yeah, I'll say that, yeah, you put, you know, make sure I get good nights of sleep and make sure I get exercise. Like almost every day. Yeah, right. And that's, that's me. Like, then I can get my work done after that.

Yeah, yeah, yeah. That's the high agency play. Yeah.

And why am I doing that? You know, well, at my age, heart disease starts to become a risk. I think at any age, actually. But, you know, especially as you get older, um, I want to prevent heart disease, I want to prevent cancer. Like, these things are two things that are sort of under my control. I don't have full control over them, but like, I'm going to reduce risk as much as I can. Yeah. Now if I continue this and I continue to adapt technology, I adapt wearables, I adapt software technology, AI, you know, how can I continue to identify and reduce that potential risk? You know, and that's what I want to think about. Like, hey, I want to talk to every cell in my body, you know, make sure we're all on the same team here, play the same way. Yeah. Cancer happens when there's a cell that like loses its ability to communicate with the rest of the tissue system. It starts exercising primordial epigenetic signals and it starts to grow like crazy because I think that's what its job is. Yeah, right. Yeah, like, I want to make that cell feel safe.

Come over to the team.

Or, or gently give it its cell death signals. Yeah. You know, so that it can move on. And yeah, you know, uh, and so like from the bottom up, I think like that's how we think about extra risk. Like with the environment we put in, how am I exposing my body and what type of like, you know, stresses am I putting on it. Um, but then there's, there's like higher level risks that happen at the cultural level, the society level, humanity, at the Earth rotating around the sun level, you know, um, meteor coming out of nowhere level. Yeah. And I think there's going to be like, I think it's an exciting life to live to just keep working through one challenge after another and to say like, this is the next risk that I see, um, having computational models of it, like extending agency to be able to deal with it. Like, that's, that's the life I want to live. And to me, that, that's my interpretation of how, uh, evolution has acted so far and what, what I was born into. And I want to just continue that, that sort of path of evolution that I was born into, the ways that, that continue to sort of like propagate mitigation existential risk. Um, and I don't think it happens at an individual level. I think it has a population level. Yeah. And so like, you know, one of my, one of my goals with the paper was to get, um more people bought into the vision, and possibly even get like an X Prize specifically for metamorphosis. Yeah.

As a goal. Tell me more about metamorphosis as a goal.

Yeah, so that's, you know, the paper is called Longevity is Metamorphosis, and it counters the idea. I think it's important that, you know, like the repair arguments, arguments, technology, that whole wave, it makes sense.

Yeah, pharmaceuticals, repair things, make sure your eyes work. Repair makes sense.

Over the next 20 years, every engine needs repair. Organ replacements by, you know, organoids and bodyoids that can replace problems that are in pharmaceutical repair and surgeries. Like, there's gonna be a lot of people's lives saved by that. So this is a yes and on top of that, right? Um, and so the metamorphosis says that the way we really solve all these health problems, the way we really do repair, the way we really do replacement is to, to extend the embryonic cycle so that in addition to our body building the 4 layers it does now, it builds a 5th layer that we have direct access to, agency, and we can have this exonervous system that can repair anything else that's wrong with our body at any given time. And that's metamorphosis. We become like the version of humans that we're evolving into. We start to take control of that, and we, we accelerate what might be a function of natural selection and during a single lifetime we start to test adaptive fitness hypotheses during our lifetime. And that's, that's us metamorphosizing into the next, you know, for Homo sapiens into like Homo sapiens plus technology. Um, and, uh, there's one version of this where I think we're, we're engineering ourselves through nervous system engineering techniques, but there's another system where we, we are born and genetically engineered so that like when the neural crest forms and their embryonic process, a fifth layer also forms. Yeah. So this is a— it's a, it's a two-pronged approach where I definitely see, um, babies being born in the future that are an evolved version of humans. Yeah. And we've specifically genetically engineered them to be more robust, more resilient against the environment. Um, and it'll be a choice that, like, a lot of parents want to give their children.

I would. Yeah. Fast forward me to dad mode, I'm like, so you're telling me I could make my kid live 300 more years and give them all the tools I didn't have, like this iron and NEMA bullshit, they would never even have it. Right. You're saying there's a different option than yes? For me, that's where I'm at, bro. Yeah. Like, I'll give my kids a good life, at least what I believe would be, which is longer, happier, healthier. Hopefully they can do more for the world, you know, for the collective. In summary, what I'm, hearing is a couple things. The before metamorphosis was, who does Jonathan want to be? Jonathan wants to be building in the front lines, preventing existential risk. And it sounds like you see that as a recapitulation of your priors, essentially, like of everything that came before you. And towards that, you take the philosophy that if that's the case, I need to maximize my own longevity. To serve my own mission, right? And towards that, it's the basics. Sleep well, eat well, exercise every day. Is that how you stay so disciplined? Is like, you know, your eyes are on the prize because it directly ties to the purpose that you've chosen?

Yeah, I'm a little, I guess, a little atypical. Even since a young age, I've been driven by existential risk and threat. You know, for me what it's done is filtered out like what's signal and what's noise. Yeah. You know, um, things I've chosen to—

or who's, who's signal and who's noise, right? Because that's a lot of the world, right? Right. Yeah. Okay. No, that's helpful. And I bet the pod audience is thankful to learn more about you. So, um, on the Metamorphosis side, it sounds like, uh, You're painting an analogy of like, look, we're already kind of doing this. It's just going kind of poorly because we don't have the right tech. Yeah. So if you see this other world where it's like the right tech could come and we could bring it to fruition in an intentional way in society, like taking what Brain Corporation is doing and the Science Corporation is doing and saying, hey, if we as a society rally around this shit, we could all have better lives and we could all minimize existential risk. That seems to be the thesis that you're making.

Yeah, yeah, that's what I saw. Like, you know, my— I think what my history and Levin's work and the, the same number of collaborators around him primed me for is that when I saw Max Hodak's technology at ScienceCore with this synthetic drawn layer, I, I thought like, oh, if we adopt this as the vision we want to have, we can have hundreds of different companies that are experimenting with this and trying to augment the human body in this way. And like, that's really— and bring it to reality.

Yeah, yeah, yeah. So on this, let's connect it to personalization and what we do at Clarity. Because at the end of the day, like a lot of this is personal, right? Agency. Yeah. Subjective, like purpose. It's all, it's all like quite personalized. Like n equals 1 is how we think of it. So in this future that you're painting a vision of, where does this idea, like, is that where Clarity would live? It's just maintaining all the information from all the different sensors you might have? Is that how you kind of see it? Like, I'm curious. So, so for people that are listening the first time, we build Clarity, which is essentially digital twins or models of the self. From a very highly considered neuroscientific and philosophical point of view of subjective being. That's what we do. We make the best in the game structured data to provide the right context at the right time for any LLM product around your user based on a very highly considered mathematical model informed by academia. So I'm curious, like our thesis has been we're building these digital twins particularly for the enterprise, and we'll move into personal cases, to serve right context at the right time for downstream systems to get the right response for this user at the n 1 level. So where does that fit into this broader vision?

Yeah. Well, it wasn't too long ago ChatGPT released memory, right? And it had a model of how it sort of understood and remembered you. Right. And then we saw this OpenClaw moment happen and all of a sudden, like, you know, oh, well, it's better if I have all the memories of me on my own hardware. Makes sense. Yeah. And it can just like insert that context into every single interaction I have with either cloud API or the local, the cloud LLM or local LLM that I'm running. Right. Just keep doing that. Right. Keep on having those moments where the data becomes closer to you. Right. And it's just, obvious. Like, you just—

that makes sense.

Open Claw as a brain-computer interface with all my data, like, in this fast communication bandwidth with my nervous system as possible, my sensory organs as possible. I mean, um, it's just a natural continuation of what's already happening. And so the minute the BCIs afford that opportunity to, like, just bring Open Claw into your body, people are going to do that, and they're going to get benefits from it. And other people are going to say, hey, I want to do that too.

Yeah, yeah. There's clearly market demand for personalized agents. And what you're saying is personalized agents, how personalized something can be is dependent on the context you give it. How close is it to your data? How close is it to your making sense of the world and your subjective being? Our file system is as close as it gets right now with Open Claw, right? And, you know, hooks into my Gmail and Calendar. But the next level is this fifth layer down the line where it's like, We just have this API, boom. And it's talking to any digital stuff. Yeah. Yeah. It has all the context that your body can provide through, you know, any— Yeah, evidenced context, like down to, 'cause right now our observations that are evidenced are clicks and event streams. But this could be your pH level, your oxygenation in your blood, right? Lots of different things. Yeah.

And those are the examples we're talking about right now are like lower on the stack of homeostatic needs and like they are meaningful, but then higher on the stack is like, what's our, what's our stakeholder graph? Yeah. Who are the humans that we're socially connected with right now? Who are we overweighting, underweighting? Like, and yeah, having that context like automatically and mapped to every single query that I'm doing is going to be informative to the value that I get out of AI.

Take action, see how aligned or unaligned, or predict if action will be aligned or unaligned and how unaligned or aligned towards who I wanna become. Yeah. Right? That makes sense. I mean, like even right there, like what if we had these, like I had this fifth layer and at night I sleep at 8, 8:30, let's say at 7 I go and in my fridge is a donut and I like crave it. And then I see, sleep supplements on the right, right? Or like just go to fucking sleep on the right. And then right there I see unaligned donut, not aligned to what you want, bro. Super aligned, take the sleep pills, like go to sleep.

Yeah. And it can feel that way too.

Yeah. Like I think that's the quality of it.

Yeah, the qualia with when you bring, when you extend your OODA loop, You know, the approximately 900-millisecond OODA loop. We had 700.

Could you, uh, for other people that don't know OODA loop, say, could you describe that, the acronym?

Yeah, I use it, you know, like the sense-making loop of like, you know, having an experience, collecting sensory information. OODA loop itself stands for Observe, Orient, uh, Decide, and Act. So yeah, taking information for your environment, making a decision, acting, um, and it maps of course, to the active inference and free energy principle that we leverage from Karl Friston, which is a nervous system-derived mathematical model. But we— so like the OODA loop of, say, a primate, say like a chimpanzee or even more primate, an older primate, might be 150 to 300 milliseconds. They see the world faster than us because there's not as much information to process. Yeah, the overlaying of our neocortex and the prefrontal cortex takes time to process, so we have a slower processing time. It's something like 700-900 milliseconds to fully process, and our OODA loop can be short-circuited, right? Like, if you were to hear a— if something would explode right outside the window right now— oh yeah, your neck would turn in 15 milliseconds. Yeah, and it would shortcut your OODA loop. It would bring your attention to this, the danger, potential risk.

Yeah, right.

So yeah, it's, so it's layered. Yeah, there are shortcuts, uh, and, and I think one aspect of like metamorphosizing is that we're extending the, the duration of like how much information we can capture the world and take in. Yeah, you know, um, and I think at first it will get a little bit longer, and in fact it might get a little bit shorter later on, but I think that's like an engineering—

what do you mean shorter later on?

Um, I think the fifth layer has the like You know, when, when the neural crest forked from the, um, forked, yeah, from the other version of this shit.

I'm going to go do something else, guys.

Yeah, what happens is that like the, the nervous system has taken over some homeostatic responsibility from lower levels, right? Um, and that will happen when we have a fifth layer, is that it like it will take over responsibility from some of the things our nervous systems do today, and that actually So might happen at a faster pace because we can like, you know, use computational technology to sort of like regulate things that our nervous system is doing today. It's just sort of offload things that are burdensome to our nervous system. Got it.

Okay. Makes sense. Makes sense. So it's like necessary evil-ish. It's more like an engineering optimization, just like— Or I'm saying the current state is necessary evil of there's so much information the brain, and we have had to evolve this way to make meaning out of things. But I specifically say like that towards our slowness in reaction, but then towards, but then if we were so fast at reacting to everything, it's like, there's so much signal, like information, and that's how we can even make sense and action in the world is that our brain protects us from having to consume all this information. Right. Right. That's why the brains are prediction machines.

Yeah. Like you wouldn't want to go head to head with a hummingbird. No.

Like— No. Not at all. Yeah. Yeah.

When it comes to perception. Yeah.

But, you know. For sure. Yeah. Yeah. Bringing my human brain to hummingbird knife fight. Or whatever analogy you want to make there. Yeah, you're not going to find a connector I lose. No, I lose. Okay, interesting. So what haven't we covered from the fifth layer thesis that you have? Oh, wow. Yeah, there's a lot.

There's a lot.

Yeah. Curious if there are areas you want to dive into further. Yeah.

Let me talk a little bit about like the nervous system and evolution. Okay, say more about the nervous system.

Yeah.

'Cause like, I don't know, I've been just really dumbfounded in terms of like learning about how different traits evolved. You know, for instance, like warm-bloodedness in mammals, largely driven by nervous system adaptations. Hmm. Really? Really counterintuitive, right?

Yeah, that's interesting.

Yeah. What? Yeah. Oh. You can look it up, 'cause I mean like—

I wouldn't have intuited that at all.

Yeah. Like, so the nervous system has particular, like it has endocrine, communication channels that it can communicate across the germ layers to, and has other communication channels. I've written about this. I don't have them top of mind, but, um, yeah, the— so in finding adaptive fitness and finding like evolutionary, you know, characteristics, uh, the nervous system has an outweighed ability to control the lower-level germ levels in terms of like, you know, so it, it can basically find adaptations faster than like pure pure natural selection could if you were just to like, like do the meso— you know, the endomorphic and ectomorphic changes that would be needed to like find warm-bloodedness.

I see.

The nervous system accelerates that. All of human domestication is driven by, by the nervous system evolution. And so we have, um, we have our ectoderm, which is the technical third germ layer, forks off and it becomes the neural crest. So both our skin and our nervous system are formed from the same ectoderm. Okay, right. And one of the fascinating things in writing this paper I found is like how the face was formed. Um, and so like the fourth layer, the nervous system, has the ability to create these, these like embryonic adaptations that create your eyes, your nose, your ears, your mouth. And you'll see like a lot of, like, a lot of the, uh, sort of genetic mutations that you see today will heavily affect those areas because like, you know, that's the, the part of the body that's like undergoing the most rapid sort of mutation in, in the human, uh, you know, evolution experiment. Um, so yeah, this is really cool paper. Isn't it like maybe 1983 or something like that, that talked about like, uh, how the neural crest, uh, works to form the, uh, different parts of the face. And, um, so what I'm trying to do here is paint a picture of like how the fourth layer of the germ layers could, could pull on the strings and like puppeteer the embryonic process in order to find faster evolutionary fitness. Now think about if we can puppeteer from the fifth layer. Yeah, right. And the number of things— if we can directly control the nervous system, we can directly control the ectoderm, the mesoderm, and the endoderm, then we have like a rich, like a rich way to evolve and to metamorphosize. And we could really take on, you know, evolve forms that are maybe unimaginable. But get wings.

Like, what kind of rings? Get wings. Wings. Yeah, go fly. I'd love to fly. Like, shit.

I don't— it's hard to really fully imagine it now, but I think step by step through neural engineering, we will discover those things. Yeah.

You know, neural engineering just sounds like a very cool term. Yeah, that's so cool. Like, to think of it as an engineering problem down the line is just— I don't know, there's something that my brain likes about that.

Yeah, it's a little more empowering than just being able to control your hair and nails.

Yeah, yeah, yeah. No, it's way more empowering. For sure. And so you said you've shared this with other key people, and it's also given them some positivity in terms of view of the future. Say more on that. Like, particularly with Skye, you've told me about some of the conversations you have with him on that. He's a professor. Is that at OSU?

Oklahoma State. Yeah. Yeah. Yeah, I think I've known Skye for years. We've exchanged messages about this topic and about You know, a lot of things related to consciousness. We recently wrote a paper together about consciousness. Cool. When I shared the paper, he said, dude, this is the most ambitious thing I've ever heard from your brain, which I thought was cool. Yeah. Uh, cause I, I tend to be like a little too far ahead of my time and be a little too ambitious. And like, but this one hit him in a place where I thought like, this is getting excited, but it's meaningful because he actually works with communications in the military. And he's privy to like a lot of the brain computer interface, uh, research that's going on. Okay, so he's like knowledgeable.

Yeah, really on the cutting edge. Yeah, cool.

I sent him the paper, he sent me back a page long of research of military technologies of all BCIs and military plans. Wow. I put that into Claude and I'm writing a paper about like the future of war based on it. Yeah, whoa.

Yeah, what's this looking like? Like what's the future of war looking like? I know we talked about this a little bit earlier, but what are you thinking the future of war looks like with this, with BCIs and the fifth layer?

Yeah, it's— there's a lot of different vectors. I think one is like, uh, shared introspection amongst, uh, a battle unit. Um, it is very meaningful, like direct connection to the sensory systems like drones and having that shared amongst like a whole team of people that can all be on the same page. Together, uh, really powerful. Yeah. Um, I, I think you're looking at like the ability to train and accelerate training in, in, uh, you know, different situations that you might have to respond to in ways that like are ingrained. You can respond in as a unit in ways that were previously not possible. Yeah, yeah, totally. I think just like the, um, you know, the cells and bodies evolve, you'll have specializations that are closer to exteroceptive capabilities and then capabilities that are closer to interoceptive capabilities. You'll have specializations in units where you have people using their fifth layers, like, like, that might be deeply interoceptive and really like, like, meant to like just create cohesion and amongst the unit. And then you'll have people that are like more on the front lines and like, like deeply in tune with what's going on in the world. Yeah, what the cosmic patterns are. It makes of— yeah, those will be certain specializations. So I think you have, you know, a form of sort of superorganism that you have to picture, um, that's— that you're capable of, of building, um, where the fifth layer is just like the connective tissue, kind of like different cells.

Yeah, which is all specialized in different things, right? Right. Well, these be some, uh, stem cell equivalent that could, you know, swat in and do whatever, right? That makes sense.

Totally. Yeah. Like, that's what you are. You're, you're a stem cell. You're— that's it. Like, how do I differentiate in this unit? And you're gonna have all this, like, training history that, you know, you know what role you need to play and yeah, what you can take on. Um, that's interesting.

Yeah. You know, it makes me now think about what's the future of work with BCIs. Yeah. The fifth layer.

Have you thought about that? I mean, yes, as much as we're all thinking, because I think like, yeah, we think about where's AI going to impact work and BCIs These guys are just like an accelerant.

It is. It totally is. Yeah. They're on parallel tracks that sort of compound and accelerate each other. I would say. That's how I have it in my mind. Just for the pod audience, we were just talking about this on a prospect call earlier, but there was an engineering leader and a product leader and they're asking, hey, we got hundreds of engineers. So what's the, baseline and improvement from here with all these AI tools coming out. And so we mentioned, and in a conversation I had earlier with a founding engineer, since December 2025, 100+ PRs from somebody full-time and really into the tools, high quality is not unheard of, right? 100 pull requests. And I myself am not the best engineer. Jonathan knows that. He always checks me in the code base, which I am appreciative of. And that's why you're the CTO. But I ship too. And I recently shipped this product to make for myself, Clarity Growth. The amount of time it would've taken in the old days, last year or the year before and before that, probably a year of work by multiple engineers. And then there's so many dependencies of vendors for datasets or whatever. Then it's like, but it's like, nah, just throw all my API keys and the environment variables and then let Claude code go to town, draft up a really good spec and you'll make all the integrations. You'll figure out all the schema mappings. You'll like, basically if you just read the spec and you're good with it, it could nearly one-shot a product. That's right now in the software space. We're nowhere near this world that you laid out, this vision of BCIs and the fifth layer. So now I'm thinking, 'cause I've been thinking for us where we do digital twins for the enterprise, like we have these self models. Jonathan and I have our own self models and our self models, he uses my self, like you'll use my self model, I'll use your self model to review my code or whatever, right? Because we want every pull request, every code change in the code base, which we see as our software asset that we go to market with, that's our asset that we sell. So everything, all our effort to go there must be aligned with the business outcomes of making money, right? Or decreasing the cost that it takes to make more money. So your CAC or cost of acquiring customers. So we realized actually, and every, I think every founding team, will align with this, which is just there's a level of alignment debt with your founding team where you're just like, you guys are on it. You're working 100 PRs and you're shipping, you're dealing with pipeline, you're making sense of the world, especially in this crazy world of AI native landscape. And then you get together and then me and you, when we get together, we're like, dude, I just, in the past 24 hours, there's a whole new direction we could take the company. We should be doing that all the time. But the fact that we can get information like that and it feels as if it's been months since we've talked, there's such a problem with normal founding teams of alignment debt to just get on the same page. So we're rowing the boat the right way. Every PR is shipping towards the business outcomes. But today it's like now it's, we could each get 100 PRs in every week. We can't even keep up with reading all the code. We have to read the English. And so there's this alignment problem that has happened because the throughput is way higher, right? And so that's just right now. And so I think that's an information maximization problem. We haven't flattened the information flow between the minimal context windows our human brains have versus our agents being able to produce so much, right? So then the idea being, can my agent be one of one with me? And can our agents be one of one with each other, right? That is yet to be proven, but I think the direction there is clear is if you can solve that, you could generate a lot more business or asset value accrual. Like that just makes sense, right? Generate 100 PRs. Let's say 80 of them are aligned with making more money, more get to $10 million AR would be the first milestone for us. Well, what if you can get it to 100? What if you get to 1,000 of 1,000 PRs? What if you get to 100,000 of 100,000 viewers, right? And then you ask, there are other actions than pull requests. Like how do we make sure every action is like that action where donut versus make more money for the business and we make more money for the business. So anyway, I set that up to say our world of work is changing quite a bit. And you're a forward, very deep thinker. And I'm, you know, you've obviously thought about our world of work 'cause we work in it every day. Well, what about everybody else? Like their lives, people's lives are changing. Industries are kind of gonna get nuked. Like we're seeing that with cloud taking over a lot of like, you know, like security, cloud for security or cloud for legal tech or whatever. And they're just clearing out the stock market, right? So what is the future of work with the fifth layer? Is it kind of what we're doing but at an accelerated, pace where we are all in alignment at all times, like that military model you mentioned, but just in a different domain that is not military?

Yeah. I guess with the way software's been built, you remember the Lean Startup days? Yeah. Even the idea that you're testing hypotheses, you go interview customers and then you iterate on permutations. And one of the problems you found on teams then is you have a customer interview and you all talk about it and everybody would have a different idea what the customer just said. Exactly. Yeah, yeah.

And it happens all the time.

Yeah, sort of these super low nuanced ways of looking at user understanding. And then you get things like that are a slight evolution of that, like Jobs to Be Done lenses. They give you like a little bit more particular language and it's like, you know, more structured, more structured.

A team can get onto that language and ontology and way of working.

Yeah, there's still some debate necessary as to like what means what, You know, but it's still kind of low nuanced. Still. Yeah. Yeah. So, but imagine you have a deeper understanding of who your users are, who's the software that you're building, right? And I think now you're taking the value hypotheses of a product onto a much more precise understanding of the subjective representation of the user and where their opportunities are to provide value. And it gives you more opportunities, more surface area to like build bespoke software that meets those value sets. And so, you know, I think like if you're, if you're, if you're spinning out 10,000 PRs a week or something like that, it gives you the opportunity to sort of like meet a whole new value gradient of human.

That's what I think too, is like, Right now with current tooling, you could nearly one-shot a product. Generating demand ends up, distribution ends up being your bottleneck nowadays with software businesses, which is a hard problem. That's half of the problem. So I'm not saying that's not hard. Yeah. But if you accelerate the timeline, then it's like, if I could generate 1,000 products in a week and I make sure all of them can make money, then that's what we're talking about here. Then, I mean, there's another view of the world, which is like, will software as a service even be a viable business model if everybody can make bespoke software? There's a different, whole different can of worms we can go into in a different pod. But let's just say software as a business is still a thing, or at least maybe services implemented with software augmentation, maybe that's still a thing. YC seems to think that's the way things are going is outcading outcome-based AI-native services companies seem to be like the way to go. I kind of look at that as like a stopgap in between the wave 1 and wave 3. Like we're in wave 2 of transitioning. So I think that take is very much like a, obviously go do services to learn more while nothing is certain. That seems to be my take of like that. Outside of software. I wanna talk outside of software. Let's look at like, I don't know, what's a profession that's so not us? Like, I don't know. I imagine doctors, nurses. Yeah, yeah. Let's take an orthopedic surgeon. What's their, I mean, I guess now we, okay. There could be orthopedic surgeon robots in the future, but anyway, I digress. Without blabbing more, what do you think that profession looks like in the future? We talked about what our profession probably looks like We don't even know the next few years, but we do know things are, it's changing. But what about for that? Like the people that are not in software listening to the pod, like what's their life gonna look like?

I picked that one because I think it's a good candidate for augmented reality.

Oh yeah. Tell me more. Yeah.

Well, so then you can imagine that you're a surgeon with glasses on that then you can generate bespoke software software onto like whatever they see, you know. Um, and so like you're actively modeling what the surgeon's seeing in their world. You're creating an active world model of like, you know, the, the human body that they're currently operating on or whatever. And there's going to be like a large corpus of information of like best practices. Maybe something unique comes up during the surgery and you need to like problem solve or something like that. Um, the The opportunity for the software to meet that is that like, you know, sort of every context that the doc, that the surgeon is in becomes a unique, like, this is their pain point right now. Right? Oh, interesting. So you have like a lot of these micro pain points that you're generating bespoke user interfaces to solve the micro pain points. Yeah. And you've probably heard of Yann LeCun's AMI Labs that just launched building world models. Yeah, like there's the, uh, Fei-Fei Lin, I think, is another one. Hugely funded companies building models of the world. And, um, another essay I'm writing on this Fifth Layer page is like, where are the world models going to stop, like, being— where are they going to need self-models? Yeah, that provide utility, right? Totally. And so if you think about that orthopedic surgeon, um, you know, where, where's the point where you really need to like have a full interoceptive awareness of everything that they have going on and what they're their experiential history is in dealing with problems so that you can recommend a step that they're trained on, right, that they're familiar with. Um, and of course, like, we always talk about the dinner problem, like, what am I gonna have for dinner?

Yeah, yeah, um, classic.

Yeah, the, the demo of, um, AMI Labs was actually like, uh, somebody wearing augmented reality glasses. They're making spaghetti or something like that. You look at the tomatoes, you look at the pasta, it identifies those things. It says like Okay, now grab a glass of water and put the spaghetti in the pasta, and it's telling you how to make the dinner. Yeah. Um, but it needs a world model to understand what, what are all the value gradients that you're eating for. Do you need a low glycemic index? Do you need something that tastes— has a certain taste profile? Yeah. Do you need to eat something that has iron?

Yeah, yeah.

It's something that absorbs iron. Yeah, yeah. Um, yeah. And so everything that it knows about you is gonna like help structure like how you can actuate your environment, put the food in your mouth that's going to serve your interoceptive well-being. That's wild. So think about all the, the dimensions of value that you meet there, no matter what the role.

Yeah, I see. Interesting. So, so here's another question, like, what is economic value in the future? This is the thing. What does that even mean anymore, right? Like, what is economically valuable? Because look, here's an example. Software engineering was supposed to be untouchable. And then now it's like, it's really hard for a lot of people I know to get software engineering jobs, right? So there's this question of, similarly to like, hey, Uber and Waymo are trying to automate out taxi cab driver jobs. What do they do now? Like, if that's kind of solved and it's commoditized, that's the, it's now a race to the bottom to provide the cheapest ride car services at highest margin automated, then what is economic value in the future? If it's sort of like the direction is things are just flattening or maybe a different view of it, I'm kind of checking myself is like, if we solve what today's world is of what we view as scarce, software engineering being scarce, no longer scarce, it's abundant, then maybe it changes what we value clearly. Like, actually, as we say that, as I say that, that, that's actually the answer. We don't know. So our, our value perception will be different because the world will be different. And if we still live in capitalism and it's still, you know, capital-based, then, then yeah, it, it— I guess it's, we solve certain problems, certain things become more scarce or more abundant, and it's still probably driven the same way, supply and demand.

Yeah, we've talked about Adam Smith Smith's The Wealth of Nations before?

Oh yeah, yeah, we have. Yeah.

So do you want to say more on that? Well, what the Enlightenment brought on was like a rethinking of marketplaces and of representation systems. Yeah, that's why the United States of America is market efficiencies. Market efficiencies, right? Um, so like if we think that— because there's a lot of like value captured in economics today at information arbitrage levels, I think. Definitely market makers, definitely.

And insurance, like any facilitator kind of middleman kind of thing, right?

Yeah. And there, there is a hypothesis that all of knowledge work is just sort of an information arbitrage play. Every software engineer— it makes a lot of sense— and that like the more these models can capture those values, the less it goes away. Yeah, the less it's captured by the human economics. It just— the value goes to zero. Yeah. And what does that leave behind? Right? So I think if you, like, take Adam Smith's representation systems and marketplaces and you make them fully efficient with, like, the end realization of the enlightenment— yeah, with AI, take it to its infinity— yeah, then you're left with basic homeostatic needs. You're left with what I think of starts with, like, metabolic needs and the value chains that supply them. Okay. And that's about it. And that's really like, you know, that's the economic game that life's been playing this whole time. So it's not like really, it's kind of like a reversion to the mean situation.

No, that makes sense. No, yeah, I get that because information age has been, I mean, actually all of life, but then especially magnified in the information age, especially with the explosion of the internet, is just this arbitrage of knowledge. It makes sense, right? Like even if you just take it for the pod audience, follow this line of thinking so you get what I'm saying. Take a kid that's born rich. What are they exposed to? More knowledge, probably from knowledge workers, probably from people that are privy information or knowledge or skill sets. And they get that, they grow up, they go work in tech or whatever, or like, you know, like some other lucrative job. and that's it. And so that even that is a sort of inherited arbitrage, right? So, um, so I think what you're trying to say, Jonathan, is let's just say all of knowledge work is a byproduct of inefficient markets. Yeah. Where this arbitrage is existing through generations because there's a necessary skill that replaces because of market efficiencies. Because we're human. That's kind of, okay, if we look at it from that philosophical lens and we hold that truth, then yeah, then all of it's gonna be commoditized. And then yeah, I agree with you. Then logically it does mean like a regression to the mean. And we were just talking about this earlier, but all the data centers being built, the electrical engineers and the plumbers and stuff are gonna get made, right? Like they're gonna make lots of money 'cause who's gonna make 'em? Software engineers? No. Like, right? It's people building the physical infra. So yeah, that's actually a very clear tangible thing where it's like software engineering demand lower, demand for a 1000x AI engineer different. That's a different, it's a different role now. You can't just take a normal software engineer and you take like a founder level builder, like 1000x PRs, like that's just, they're different classes entirely. And so I think I just wanted to say there's that nuance. But the demand for software engineering as a profession has clearly gone down. And the bet I think a lot of companies are making is, well, we could do way more with limited people. The demand to meet that compute, it's funny, regresses to the mean of like, okay, well, like, we need people that can actually build the physical stuff to then capture economic lev— value at the digital level. So it's just, I guess, pendulum swings, right? Like, it's swung in the 19th and 20th century, 21st century, 20th, 21st century, and then it's kind of swinging back now that we've put intelligence in a rock. Yeah, yeah. It's a funny way to think about it.

Yeah. I'm curious, have you heard Elon Musk's plans for the Optimus, self-replicating Optimus? No, what's that? Well, okay, so this is relatively recent, but he made— he merged SpaceX and XAI. Oh, okay. Yeah, cool. Merging them. And he plans to build this Optimus robot. And one of the key characteristics of it is that about 30 Optimus robots can build new Optimus robots.

Oh, hey, shit.

Right? Yeah. And he's basically looking at it as like a race to capture energy from the sun. He says right now we like capture less than 1%, 1 millionth of 1% of the sun's energy. Okay, so we need to build Optimus self-replicating space robots that are going to build, um, you know, uh, AI data centers in space that can capture solar power. And like, that's where it's the Kardashev scale, right? I see. Um, we're just working towards the Kardashev scale. I see.

So that's like super endgame. Interesting.

If that's the reality you want. Yeah, yeah. This is— and this is why I wrote The Fifth Layer. I don't like that reality. Yeah, yeah, I don't. I don't like mega corporation.

Tell me more.

Harder than the sun.

Go hard at that. Yeah, tell me why.

Tell me why. Well, it's, it's sort of like one path of evolution, right? Like the, the two major hypotheses of biomass on this planet were chordates, which of which were one bilateral animals that, that grew, uh, central nervous systems and then grew into humans and then like kind of like started domesticating everything. Yeah. And the other one was eusocial insects. Um, eusocial insects account for the second most amount of body mass and maybe the most biomass on the planet, uh, if you don't include humans that domesticated cattle. Yeah, yeah, yeah. So like there's there's ways we can choose to evolve. And like self-replicating Optimus robots on the moon that go build the data centers and then send profits back to like an oligarchy back in the US or on Earth is one reality we can build. Yeah. I don't think we want to do that. I think, I think we want to elevate the subjective human experience. Yeah, I think it's what we want to use technology for, elevate the quality of human experience, the quality of lives that we can live, um, as directly as possible. And, you know, look, uh, Musk also built Neuralink, so I think he sees this, and I think he saw that, like, we need humans that can compete against these Optimus robots. But, like, what are the economic incentives that we're going to give to, like, create one reality or another? Um, and ultimately, will, will, like, a population full of fifth-layer humans compete against a population full of Optimus robots to like harness the solar system energy, and which will find its ecological niches and what will it look like? Yeah, I think that's kind of like—

and then that's like a simulation that you could sort of think through.

Yeah, and amongst that simulation of the superorganism we're going to build, we're just like a little cell and superorganism, and what is it like to be us in that situation? What would we want it to be like?

Yeah, um, that my little like ion channel down the road, it's Like, do I go down on positive or negative? Like, what am I differentiating today?

You know, like, um, yeah, so I think, I think we want to like, uh, prioritize like the elevated human experience first and then like, you know, use robotics as an extension or a tool of the human superorganism to, to like mine planets and space and stuff like that. Yeah. Um, But yeah, I mean, look, we're at a juncture now where the race is on. The race is on for sure.

The race is on for sure. It's wild to think about. And to what you say, I'll add that totally agreement. Don't think that the end game for humanity should be we build robots that build more robots that then go capitalize value in our solar system and beyond to bring back money to some oligarch on our planet. Like that, I'm like that. Doesn't seem, if I gave that to everybody in 3 slides on Earth and said yes, no, like I'm pretty sure there's a lot of no. Like I know Elon's gonna be like, yeah, but like pretty sure a lot of people can say no. That would be my forecast. My take is I think I just have a deep desire and value around everybody becoming their best self. Because I think that things are in arbitrage right now. There's so much— like, I didn't grow up with a ton, right? And I know how painful that is. And so when you are— when you don't have much, you're only worried about getting anything at all. When you have your home, your shelter, your food, you think about purpose. You think about, let me serve the collective, let me serve my brothers and sisters around me. But when you don't have that, you're not even there to figure out what is my subjective purpose and how do I give back.

Yeah, right.

And so there's a question I've always had, which is, do 8 billion people in the world, would they say that they are really living towards their subjective chosen purpose? No, I'd say that's like a very small number of people that could say that. So there's a question there is, if we love life, we love humanity, and we love each other, wouldn't we want to see what our potential is like? What if 8 billion people on the planet did say yes and actually felt that, that I am in alignment with my best self, I'm actualizing every day, the world would look so different. I think the world would look completely different. I think we would've made or innovated on so many things that we probably haven't, right? And that, at its core, like, here's a tangible example, but look at Einstein. Everybody thought he was a dumb fuck, actually. Like, he was dumb in school, and then he went and got a know-nothing job, and was a nobody. And he chose to just do this know-nothing job to pay the bills so that he could spend all his time doing his obsession, which was thought experiments. And he came up with theory of general relativity, like, which has informed lots of innovation and our, you know, bettering of lives and technology and stuff like that. Imagine if we had 8 billion of those. You know, it's kind of a stretch, but like What if, and he went through more adversity, right? But what if people didn't have to? And we had a higher chance of producing thinking like that, that moves all of us forward. That like gets me excited. That gets me really excited, which is thinking about how do we empower everybody to become their best selves? And then if we did do that, like I feel positivity in that direction, in that ion channel, if you will. That just feels like the right way to go. So whatever we gotta do to do that, I'm game. So that's why, you know, we're always thinking about alignment. It's how do we align all the cells on the planet? Yeah. Right? How do we align all the people in this analogy on the planet? And towards that, it's, if we are all thinking like, well, I am a little stem cell, what do I do for this giant organism of humanity that I'm in? We gotta give those little stem cells the ability to figure it out. Right. That's how I see it.