Podcast Host (3:16)

As you said, both of those individually are very exciting. And when you combine them, it's almost, and I'm sure you would agree it feels very difficult to even imagine where we could be a few years from now if they keep happening.

Salim Ismail (3:30)

That's right, and this is why I'm bald. I mean this is like an unbelievably full time job trying to figure out, okay, and by the way, each of those technologies like solar energy, biotech, neuroscience, drones, AI, robotics of them is doubling on their own, but where they, where you intersect them, that adds a whole other multiplier to the equation, right? So that's like crazy talk at that point. And so there's some staggering things happening. I'll give you a data point that's maybe useful for your viewers, which is when we were launching space shuttles like 20 years ago, it was $600 million per space shuttle launch. With all the employees and everything that NASA had. SpaceX comes along with the reusable rocket technology. They drop it by 10x to 60 million launch. Pretty amazing. And we're about to have a relativity space where Eric Schmidt has just joined the CEO. They're going to be able to do it for 6 million a launch. So that's 100 times drop in one generation of a, of crazy capability. And, and let's note that rockets is not some Silicon Valley digital social media play. It's like real physics trying to get out of the gravity well of Earth. And even there we've seen 100x drop, right? So this is like a staggering amount of demonetization, democratization that's happening across the world.

Podcast Host (4:49)

So let's stay on space tech for just a minute because this is an area that I hear, you know, people throw around space tech as, oh, it's going to be one of these next frontiers. I haven't heard that stat before though. About the $6 million rocket, which is, which is pretty astounding when you start to.

Salim Ismail (5:04)

Pretty incredible. And what can you do when you have 100 times more launch capacity?

Podcast Host (5:07)

Right, well that's, that's the exact question I wanted to ask you. What, what can you do if you have 100 times more launch capacity?

Salim Ismail (5:14)

So it turns out the thing that we'll be doing, we thought tourism would be one big use case for space. We thought maybe launching to Mars would be another great kit. But it turns out data centers in space seems to be the most likely thing to do now. Elon thinks like two, three years, we're good to go type of thing. I think it's going to take a little longer than that just because space is just really, really hard and, but it'll happen and it's going to happen over I would say a five to seven year period. But once you start doing data centers in space, like anything is possible at that point because right now there's an inner loop where we have energy generation turning into intelligence. Intelligence is turning into radical capability which the radical capability then feeds the energy loop and we've got this inner loop going that's kind of incredible that we've literally again never seen before.

Podcast Host (6:07)

So just staying on the Elon, talk about these data centers in space because again that's one that's sort of a hot topic right now. And I haven't gone deep here, but as I understand it, the energy is kind of built in. Does that kind of intersect with solar and how that's taking off and what happens here?

Salim Ismail (6:26)

Yeah, imagine you have a small nanosatellite, right? It launches into low Earth orbit, A fan of solar panels opens up, it's gathering solar energy to power the compute. It's beaming it down and then it's pointing the on the back and you have a bunch of fins radiating the heat from the chips out into space. And so the whole thing runs essentially for free once you have it up there. So that once you've got the CAPEX dealt with, there's almost zero opex, the energy is free, etc. Etc. So there's, the basic thesis is that we'll be able to do a huge amount of compute up in the sky over time. Now it turns out space based solar energy is about five times more efficient than on ground. But on ground that means you just have five times more solar panels and you're good to go. Right. So I think you'll see a combination of both. It's very clear the solar plus battery and Storage is going to be the major future for energy in the world.

Podcast Host (7:25)

I want to, I want to extend that a little bit because compute, to me, as you sort of said, like, compute unlocks all these new things you can do that, touch all these adjacent technologies. If you put your futurist cap on for a minute, where do you see the implications economically and I guess societally and you know, we don't need to talk about 50 years, but in the next handful of years, if we're able to start unlocking this kind of, you know, exponential energy, exponential computer.

Salim Ismail (7:58)

So the easiest framing is scarcity to abundance. So we've run the world on scarcity for the whole of history of humanity. Throughout history, water has been scarce, energy has been scarce for the huge history. Then we discovered hydrocarbons and fossil fuels and that radically dropped the cost of energy. But now we'll go to near zero because solar energy, once you start capturing it, is almost free. Once you hit that inner loop again of energy being free, desalinization becomes free. Right. If you have desalinization and clean water, it takes out half of the infectious disease in the world because that's covered by dirty water, creates most of the infectious diseases. So the ripple effects start growing very radically and you move from scarcity of lots of things to abundance of lots of things. And this is amazing at one level, very unnerving from a business industrial perspective, because for 10,000 years, if you didn't have scarcity, you didn't have a business. Right. And so how do you deal with that as we move forward? And that's one of the biggest tensions that I tend to explore with the future of exponential organizations. What are business models around all of this? How is this going to, how are we going to navigate this?

Podcast Host (9:09)

We.

Salim Ismail (9:10)

Do we end up in like a wall E scenario and overindulge? Right. Do we end up with a decent moderated thing? Will we fight wars if energy is free? Right now all the wars are being fought over oil. And if you can take out oil and you have abundant energy, well, what are we fighting over? Right. I'm sure we'll find reasons, but it'll make it a lot harder to do that work.

Podcast Host (9:32)

Well, that's sort of where I wanted to go next because I love the dialogue and I love the vision of abundance. It's so exciting and it's so compelling. And I think we as a society are in such dire need right now of something to look positive, to get excited about in the future. Right. And so what I wanted to ask Though Salim is like, how do you, how do you square that with like what you read in the paper and what's going on geopolitically and, you know, wars and. Because there's just, there's obviously this, you know, utopian vision that we're driving toward. But things on like a lot of people will say, yeah, right, you know, that's not my experience day to day.

Salim Ismail (10:16)

Yes. So we have actually seen this. Right. So if you go back 30 years ago, I'll give you a couple of small examples. So if you go back, say, 25, 30 years ago, before cell phones, we had no ability to. Communication was scarce. Right. A phone call cost a dollar a minute, wherever you wanted to make the phone call, etc. Roll forward 25, 30 years. I can video with my son for free anywhere on the planet. I can project empathy anywhere on the planet. So we've gone from scarcity to abundance in two very clear domains, communications and information transfer. Right. That's like amazing. Once you can have a couple of those things happen. And that's had huge ripple effects from an industrial business perspective. Let's take music for example. Used to have seven or eight music studios selling cassettes or CDs or DVDs. Selling the physical scarcity. Right. $20 an album for a CD or whatever. Then we digitize music and. And now you have itunes and Spotify. Delivering abundance on a subscription model. So we've made that full transition in music. We expect that transition to happen in healthcare, in education and transportation, in utilities pretty much everywhere. We've got this product scarcity model. We'll move to an abundance subscription model. So that's the transition we expect to see on that side. This causes lots of perturbation. Obviously those seven or eight music studios are essentially out to lunch at this point. Right. We sell, for example, 90 million new cars a year. Right now if you own a car, it sits empty 94% of the time. It's one of the worst purchases you can ever make. And it's depreciating while you own it. Right. If you have a house, at least house generally is appreciating when you own it. But a car is depreciating the second biggest purchase and, and we use it 6% of the time. Okay, so when we can use Uber, autonomous cars, robo taxis, whatever, that's completely going to make transportation 10x20x more efficient. But huge challenges for the car industry, what are they going to do, right? Do they turn into drone factories, robot factories? What are you going to do with Those. So there's, so there's the push and pull of this radical transformation, but it's happening across the board if you work in it.

Podcast Host (12:29)

InfoTech Research Group is a name you need to know no matter what your needs are. Infotech has you covered. AI strategy, covered. Disaster recovery, covered. Vendor negotiation, covered. Infotech supports you with the best practice research and a team of analysts standing by ready to help you tackle your toughest challenges. Check it out at the link below and don't forget to like and subscribe. I had a conversation like last year with Mo Gaudat about sort of this topic and big, big fan of his by the way, huge super smart guy. He talks in kind of broad terms as well about abundance. And one of the things he says though that we talked about was this idea of having what he called a messy decade or that things would get worse before they get better as we kind of, you know, wrestle with this disruption. Do you agree with that hypothesis? And I guess in your mind, like, what do we have to get right to make sure we don't screw things up too bad before things get really good?

Salim Ismail (13:34)

It's a great question and one of the most important conversations that we could be having in the world today. And Mo is leading charge on that conversation. We do have a very messy decade before you switch all our systems over to abundance based systems. Okay. It's a very ugly because our history around this is not great. When you have major technology challenges, you have Luddite revolts, you have people freaking out, you've got what I call the immune system response. In big companies, if you try anything disruptive, the antibodies attack you. Because all big companies are designed for efficiency and predictability, not for something crazy happening on the edge. And so this causes problems at the business level. But the biggest challenge I see is actually in our institutions. So it turns out there's about 50 institutions by which we run the world. Legal systems, healthcare systems, intellectual property, governance, et cetera, et cetera. And all of those are dissolving in front of our eyes in the face of technology. So one way we frame it is if you take these dozen technologies that are all intersecting and launching new classes or products and services like you cross drones and computation. You have flying cars, you have cryptography and computation, you have blockchains at the intersection of that. So you've got major application areas being unlocked with these intersections. Well, this is a huge challenge for all the institutions by which we run the world. They're all dissolving in front of us. Journalism is broken, education is broken. Democracy is breaking in front of our eyes. Right, right. Because with the pace of change, it's hard to run a democracy because you have to bring the whole population along to vote on things. And that takes a lot of time. And the world is changing so fast, we don't have time. So we're turning to autocracy so that we can get decisions made. Then you're at the mercy of the whims of one political leader. Bolsonaro, Trump, you name it. And yes, we can get decisions made, done quickly, but hopefully, are they the right decisions and are we making the right choices? So we've got some very big structural issues around our institutions. That's where I'm kind of spending a lot of my time. The way we view the world is we've got two futures in front of us. A Star Trek future and a Mad Max future. Right now. I used to think we'd have one or the other. And can we tilt from the Star Trek. The Mad Max future that all our politicians are pulling us to. Gaza, Ukraine, Middle east today is clearly Mad Max. Right. You can see right there we have the opportunity to move it to a Star Trek future, which would be Silicon Valley, Shenzhen, etc. Etc. Mochacino delivered by a drone wherever I want it. You know, all of that kind of bright future. I thought we'd be one or the other, but it seems per all the science fiction novels, we're doing both at the same time. So we're speedrunning pretty much every science fiction plot in real time in front of us. It's everything, all the time, all at once type of thing.

Podcast Host (16:32)

You mentioned the pushback that people have this kind of Luddite factor in the mix here.

Salim Ismail (16:37)

Yeah.

Podcast Host (16:38)

And it feels like it's growing. And, you know, AI seems to be like, you know, the kind of the. What people are rallying against. And I'm curious, Celine, because if you talk to a lot of the people that hold these views, they'll say we're going toward the Mad Max future. And I have no choice but to, you know, push back against that and rebel and say, like, no, I'm not using any of this stuff. Stuff. Is this backlash, in your view? Like, purely negative? Can it be a productive force? Where does it sit in the mix? And is there any way we can kind of, do we have to convert these people? Can we harness it for good? How do we use people's nervousness or tension to get somewhere better?

Salim Ismail (17:24)

Great questions. Okay, so let me first contextualize what's actually Happening. Okay? So there's. When I talk about immune system, it's the cultural resistance to breakthrough new ideas, okay? And it actually starts at the individual level. We all have at the back of our brains a little organ called the amygdala, right? It's an evolutionary mechanism that's scanning for danger. Because if you went back 10,000 years ago, if you heard a noise in the bushes, you ran. Why? Because bad news could kill you. Good news doesn't kill you. So if you missed a piece of good news, I might miss some fruit that I could eat. If I missed a piece of bad news, I literally died. So we're 10 times more likely to pay attention to the negative news than positive news because of that old survival factor, right? This is why Fox News does very well. If you watch Fox News, you're going to die this week, right? If you're lucky, you'll last till next week. But you're going to die. You'll die, but you're going to die. We call CNN the Crisis News Network, because when you can track every bank robbery in real time, high definitions tuned to 20 devices, you think the world is going to go to hell. And you vote on that. And then you end up with Brexit, Bolsonaro, Trump, whatever, Whatever, right? So that's the physiological challenge that we have to face. Where that intersects with technology is when we come across new technology. Take autonomous cars, for example. The first reaction when we see that autonomous car, we don't understand it. And then it provokes a fight or flight response. We basically go, oh, my God, that car might kill somebody. Let's ban the car until we figure this out, until it's safe, right? Because people don't want to be killed by robots. They'd much rather be killed by drunk people, which is what's happening today. Okay? So it takes time to habituate to that uncertainty, and it lights up our amygdala. It means you can kind of freak out about it. And that's the first freak out factor. When people hear about drones or autonomous cars or humanoid robots is always that initial factor. This is what Hollywood praise on. Hollywood always shows disruptive technology as dystopian, right? You have the Matrix, you have Skynet, you have the Terminator. If the robot overlords come and take over the world, that's always the plot. If we're lucky, we're pets, and if we're unlucky, we're food. You know, it kind of goes that way. But if you look at what's happening with technology, it's actually totally the opposite. We're actually radically augmenting the human experience with technology very quickly. Right, You've got spectacles, you've augmented your capability. The minute I get a vaccination as a child, I'm technically a cyborg. We're improving the human condition with technology in an amazing way. And now we have a dozen technologies all moving at that pace. So the umbrella thesis is really simple. Technology has always been a major driver of progress in the world. It might be the only major driver of progress in the world. Now we have a dozen technologies on an accelerated doubling pattern that should give people huge optimism. But because that uncertainty unknown triggers the amygdala, people freak out. So you have to get them over that hump. And we don't have a good way of doing that. You could use narrative and great stories about that, but people tend to listen to the negative stories. So if I could wave a wand, I would cut out the amygdala of every person alive. Because our chance of dying a physical death and worrying about a lion jumping out to steal our baby is almost zero compared to where it was 10,000 years ago. And yet we're governed by that freakout factor. We vote on it, we watch movies based on it, et cetera, et cetera. So that's the physiological challenge. This magnifies into all of our institutions, right? We freak out about the negative things that may come in scared. And you look at the political, the political discourse here. I'm Canadian, but I've been living in the US last 25 years. And it's kind of incredible to see that, oh my God, that Mexican is going to come and either steal your job, rape your kid or whatever, and you pick whatever ethnicity you want, et cetera. There's always this freak out factor. People create the other. And though Middle east is the oldest model for this ever, and you live on that paradigm, right? And it's very easy to stoke that type of tension. And then you have war, et cetera, et cetera. So hopefully what happens in the near term, and I'm talking like two, three, four years from now, is you get abundant energy. Abundant energy means we don't need to be fighting over energy, means we can have natural abundance. It'll ripple into all sorts of other areas. And we can have to. The amount of war in the world drops very radically because we just don't need to fight that much. And so that's the hope. Human beings will figure out ways of fighting, but that's the optimistic side of it. I totally agree with Mo that we have about a 10 year, very, very big danger point and if we navigate it well, we'll come to the other side of it and then go vertical from there. Have you ever seen the Gartner hype cycle? Yep. So, you know, you've got this technology and people get overinflated expectations, then you have this trough of disillusionment and everybody ignores it. Like 3D printing is there now. Everybody ignores it and then little by little it takes off and ends up way better than you ever imagined. Right. We've seen this, that curve throughout the world. AI is right now on the, in the peak of that inflated expectations, etc. Etc. It's going to do everything along with robots, etc. We have to get through that curve because civilization is going through that curve right now, going from a scarcity to an abundant paradigm. And we have to go through that trough of disillusionment and hopefully that trough isn't too deep or too long. And how do we minimize that? And that's a lot of the work that MO is doing I've been doing. I've built a whole ecosystem community around that, trying to do exactly that, to minimize that trough so that we can come out the other side a little bit more elegantly. Otherwise you end up with like several hundred years of the dark ages.

Podcast Host (23:10)

There's a lens on that I want to push on a little bit, which is we've talked about scarcity and abundance from sort of like a resource angle broadly. Right. I want to talk about it from, I guess an individual economic angle. Because to me, so much of this fear, this concern, this push for authoritarianism comes from this sense that this sense of scarcity economically that I'm in some way losing out, that somebody is taking something from me. And one of the narratives that is creating a lot of traction around AI is oh, I'm going to lose my job to AI. Right. That technology is actually going to leave me economically worse off than better off. And so how do we get to a world where there's more, I guess, economic abundance, where an individual, whether they're working on their own or whether they're working for a corporation that is thriving, how do we get to that economic thriving point so that people can let go of that tension?

Salim Ismail (24:19)

So there's about 10 questions buried in there, but, but an important set of questions. Let me try to address that a little bit. And it's probably the biggest question we're facing from a social perspective today is how do we navigate this from an economic point of view, we run the world on money today. Economics, success, prosperity, business. The main mode of discourse in the world today, where people get stuck is they see this awesome promise of AI, they see cool stuff, et cetera. They think it's going to replace human work. And this is an old conversation. We've had it since the Industrial revolution where people freaked out and said, oh my God, that's where you had the Luddite revolt. They were actually breaking the cotton mills, the cotton machines, because they thought they'd be out of work, et cetera, et cetera. Let me give you a story. In the 1970s, we had ATM machines get created and you could automate banking. And there was all this hand wringing that happened, going, oh my God, millions of bank tellers will be aimlessly wandering the streets. What will we do with them? There'll be social unrest, et cetera, et cetera. We have to figure out something to do with these millions of bank tellers. Do we stop the machines? What do we do? Et cetera, et cetera. What actually happened was that the banks, the cost of running a bank branch dropped about 10 times. The banks created, created 10 times more branches and the number of bank tellers didn't change at all. And this is something we see repeatedly in technology. I'll give you one more data point just to triangulate. The highest penetration of robotics in the world is in South Korea, Sweden and Germany. The lowest unemployment rates in the world are Sweden, South Korea, Germany. Why? Because as we automate, there is so much more work for human beings to do. An increased efficiency, design thinking, problem solving, et cetera. You can lift the human being up the stack, okay? And so that's what we've seen historically 100 times out of 100 when we've seen this. So when we automate with AI, I'm actually writing a third kind of a book paper right now called the Organizational Singularity. As we have AI coming along, it'll radically change how we organize the world. Right now, all of our workflows and companies are human to human based, human centric, right? You do something, you get approval from your manager, it goes up to the next level. A report gets rolled up by a fourth person, goes up to senior management, whatever, whatever. Well, over as we automate with AI, we'll be doing AI centric workflows. Well, where's the role for a human being? And I think what's going to happen is the human being does oversight and exception handling where the AI agents do a lot of the knowledge work, I think we'll be able to run a typical company with about 20 to 30% of the workforce that we had before. And the first people go is, oh my God, that means all over 80% on unemployment. No, because we'll create five times more companies. And so that's what I expect to see happen. And that's what we've seen, by the way. Every single time in history there's been a technology injection. We see a temporary drop and then we just create so many more work, so much more jobs than we could possibly imagine. So if that holds true and everybody goes, this time is different because of AI, I don't buy it. We've never seen this in history, why should we see it now? So we're actually mapping that out right now to see where this goes. Now you have to take into account I'm a bit of a radical optimist around some of this stuff. So I could be wrong because I've got rose colored glasses on. People are like, yeah, you're always on that positive side. It's going to be so much worse. But we've never seen it before. So the evidence is not there to support that freak out feeling. And so that's why that's where I'm hoping we'll go.

Podcast Host (28:05)

I want to follow that thread that you were talking about around kind of the size and shape of organizations and that, you know, organizations don't need to be as large maybe as they are today to do the same things. And one of the questions I've been wrestling with is, you know, there's this kind of dangerous question to ask, which is how many people should an organization have? And one of my hypotheses is that pretty soon it's going to seem silly to have these enterprises that are like 100,000 people and that we may end up with a lot more. There's like the story now about when do we get to a billion dollar company with one person? Is it going to be more entrepreneurial than before? Is there going to be just a more robust mid market? How do you see the shape of the economy changing in terms of the role that these companies of different sizes play?

Salim Ismail (29:01)

So there's about two major vectors to talk about here. The number one is what happens to people and what's the future of the firm? I think what's happened is the comet has struck and now we're going to have a Cambrian explosion of little furry mammals running around the big dinosaurs, meaning typically big operating companies or toast. One way to put this as my colleague on the podcast, Alex Wisner Gross talks about is that knowledge work is cooked. Okay? So this is equivalent to what happened say 100 years ago. We had all these factory lines where you'd be stamping out widgets and then you created a machine to do it and you were out of work, right? And so one level people are like, oh my God, what are we going to do? But as I said, we've got so much other work to be done. We shifted from manual labor to knowledge work. And from knowledge work we'll move to more creative thinking. Like for example, you can get a perfect lab grown diamond today that's absolutely perfect with no flaws in it. And you know what? People prefer the natural diamonds that have some flaw in it because it shows it's actually authentic and real and not lab grown. And so that's the kind of thing we expect to see happen. When you, when you study societies that have reached abundance in history, the Moguls taking over India, the Romans taking over Europe, the Mongols taking over most of Europe, et cetera, you typically end up with four activities. Food, art, music and sex. Okay, not in that order. So we expect to see this explosion of the creative class where people are now feeling a lot more self expressed. If you're building a company, you're building a company in a domain that you have deep passion for and it doesn't feel like working. And so there's this incredible self fulfilling of reaching a level, the highest level of Maso's hierarchy, which is self actualization. As robots and AI start filling in the lower levels, we think that's where we get to. Again, keep in mind, I have my rose tinted glasses, but I can see a very clear line to that. Many people today are doing the work they're doing because they want to do it, not because they have to do it. If you went back to just even two generations ago to say any of our grandparents, they were doing the work they were doing because they had to do it. They had almost no choice to doing that kind of work, to raise that kind of a family, to provide for their kids. And today we have a lot more choice. And I think that amount of choice is exploding every year that goes by.

Podcast Host (31:33)

I'm still chuckling at the four pillars of a thriving or abundant kingdom. But one of the things that comes to mind is most of those pillars seem to align with some of these panics around AI that AI is ruining music, AI is ruining art, AI is maybe in some ways ruining sex. I think food might be safe do you buy that or do you think that these are still areas where humans have something of value to contribute? As long as we're a species based on know that human demand, we pay for that humanity.

Salim Ismail (32:13)

Okay, so let me explain what I mean by that. If you take a professional team, soccer team or football team or hockey team today, right, Pretty much everybody has the same training, the same access to the right nutrition, the same physical therapy to be operating at the position, their game. They have psychologists for every tennis player and every hockey player, et cetera. And then what do we do? We pay to see them on the ice and can they make the big shot at the right time? Right? And we pay to watch that human flawedness to see can people perform under the crunch, et cetera? Can a music artist deliver a great performance on stage live in front of a lot of people? We pay for that humanity and we'll be doing a lot more of that. And we actually really appreciate and we value that human experience in a very profound but very subtle way. And I think as AI comes along, as robotics comes along, we'll be doing just that much more of that. Look at the success of like America's Got Talent, right? We love watching people strive and go through that hoop. And AI may do like today, an AI can beat most players at chess, but chess is more popular than ever because you kind of go for that human factor. Can that grandmaster find the right move in a pressure crunch situation? And we live for that. We live for that kind of angle of it. The drama of that. I don't think that's going anywhere. So that gives me huge optimism that as we automate lots and lots of other things, that human feeling, that human factor is something we appreciate more and more.

Podcast Host (33:52)

I want to bounce a hypothesis off you that I've been playing with for a bit about the role of, you know, this human factor. Because there's a lot of, again, there's a lot of talk that so many of these knowledge driven roles are just going to, you know, go extinct because they'll be replaced by AI. My hypothesis is that they're actually going to drive up demand for humans at sort of the premium level. So if you think about, you know, doctors, if you think about lawyers, if you think about consultants, that if you're willing to, if you really, you can get this long tail for cheap or free with AI, right, suddenly there's this tier one that never existed before that gives people access, but it's, I don't know, my hypothesis is that that's not going to remove demand for the best of the best. There's still going to be demand there for the things that really matter. And you really want the human touch for versus this complete shift. I'm curious if you buy that or if you have a different, you know, a different prediction.

Salim Ismail (34:52)

I totally buy it. And I totally buy it from the couple of following reasons. And let's take education and let's take healthcare as two examples. Okay, it's very clear that we'll automate education. It's going to happen. Having a teacher delivering a lesson repeatedly makes no sense in today's world. You may be familiar with the Alpha School where they're delivering all the academics in a 90 minute to two hour timeframe with a customized AI delivering a lesson plan to a kid so they can get a day's worth of academics in 90 minutes and then the kid is spending the rest of the time doing workshops on their biggest passions. So that's the future of where academics is going to go. It means though, that the teacher becomes even more valuable as the guide, as the mentor, as the coach. Not delivering the lesson, but giving the encouragement, giving guidance, being with a person as they're learning, helping them evaluate options, et cetera, et cetera. And that's really the most profound and highest level use of a teacher than simply standing up there giving out grade seven math, right? So that's one area. The same thing is going to apply to doctors. If you're a doctor today, you get like 10 minutes per patient, right? And you can't really diagnose this. You don't have time to spend with the patients, you have so many more to see, et cetera, et cetera. And there's some staggering numbers, by the way, in healthcare, if you've seen this or not. If a doctor does a diagnosis today, they get it right about 74% of the time. That's for human doctors. If you give the doctor An AI, you get it right 76% of the time better, but, you know, not that much better. If you have an AI alone doing the diagnosis, the AI gets it right 88% of the time. What does that tell you? That tells you doctors should not be doing any diagnosis of any kind. They're bad at it, they have their biases, they've had a bad day, they've hadn't read the latest papers on this particular condition, et cetera, et cetera. So if an AI does the diagnosis and then the doctor can spend way more time with the patient doing the treatment, figuring out regime, ensuring compliance, whatever the treatment protocol is, and everybody wins in that model. And that's the part that people don't get, is how freeing this will be for many, many people in many, many roles and many, many occupations.

Podcast Host (37:15)

Tying that back to organizations and business models and broader kind of sector impact on AI. There seems to be a view right now, and again you can see it play out in the stock market that there's just entire sectors that are going to kind of crumble into the sea because they're threatened by it. AI?

Salim Ismail (37:33)

Yes.

Podcast Host (37:34)

Do you buy that? Do you see entire business models that are threatened here? Or do you think there's just going to be some, I guess, call it delivery realignment in terms of the same things that people have always wanted.

Salim Ismail (37:47)

So there is going to be a huge amount of creative destruction that happens because the business model is changing in a number of areas. Let me give you a couple of examples. In say consulting used to hire a really smart person at McKinsey's or BCG or Deloitte Consulting or whatever and then you would sell them out, rent them out to a client at typically three times their salary and then that covered your training and the number of people you had on the bench and that provided a viable business model. And you, and you pay and you rent that amount by the hour, $1,000 an hour for a good senior consultant. In this domain, that model is breaking because in terms of where we're going, we're now doing consulting, advisory on much more of a shared value thing. We'll help you build that new division, but we want 20% of the revenues. And so then incentives are much more aligned. The upfront costs are low and the risk is shared in terms of the consultant and the client. So that type of business model shift is what's happening across the board. This is going to happen in many, many areas. And the difficulty is people aren't used to that transition or have been sitting pretty and comfortable. SaaS is the obvious example where you're selling per seat, right? Well, in an age of AI, you don't sell per seat. What do you do now? Now does that mean the end of SAS companies and people whack the stock? Because the multiples, we don't know what the multiples will look like, which is a valid point. And there's a structural change coming. But if you take say Salesforce, Salesforce is so wired into a lot of the their customers workflows that it's very hard. Switching costs are very high. And so people, it's much easier for people to Go, let's just have Salesforce automate and use AI and everything they do, and that'll seep into everything we do by default rather than trying to change everything about how we do things. So I think there's a reasonable future for existing organizations. They just have to pivot very, very quickly to whatever the new model is. And then you've got the horde of startups trying on the new model that will force that change. I think that's all great and it's going to force a transformation of lots of business models, SaaS being one of them. For me, the biggest issue is not so much private sector as public sector, because we need to reevaluate and rethink how we do everything in the public sector. Government policy, implementation, all of that stuff needs to be rethought because our institutions, the guardrails and the scaffolding that we use to create civilization safe, those are dissolving in front of us. And that's for me, the biggest concerning area.

Podcast Host (40:25)

I'm asking an unfairly difficult question, but how do you do that? Because in the commercial sector there's this competitive force, there's creative destruction and the survival of the fittest. In the public sector, you've got a monopoly on providing services, typically within a country or in a state. And there's just, I'm sure you've seen it firsthand. There's so much calcification, there's so much, as you said, these antibodies that are change resistant. How do you innovate in that environment?

Salim Ismail (40:58)

There's a model for absorbing disruptive innovation that we know now how to do it. And it works as follows. What you do is you do disruptive things at the edge of. Buckminster Fuller first identified this. He said, you can never fix a system from the inside. You have to create a new system at the edge and slowly move things to it and let that become the new gravity center. Okay? This is why the car industry could never create a Tesla. It had to be a startup on the edge. And little by little, everybody's moving to the electric car model because it's just a much more, much better technology. That's the kind of model that we're going to see evaluate. And as you said very correctly, private sector can navigate this. You have creative destruction. If you're not earning money, you, you get penalized and you go out of business, whatever. But public sector is a very different beast. But public sector has all sorts of advantages where they can do these types of experiments. So for example, in the uae, Dubai has A whole bunch of economic free zones where you can do things in media or in television or whatever without the typical regulatory constraints and try things out there. And then as things work, then you move it in the mainstream. Right? So that type of a model will start to emerge where we can try different policy ideas in different ways. We're huge fans of what's called UBI Universal Basic Income because we think that's the model for which the future of government will move to. But how to get there is the big challenge. Because to move from a taxation, employment, labor union model to that is such a big leap. We have very little confidence in public sector getting us there. Except that it's looking increasingly clear they won't have a choice. They have to move to something like that because it's too difficult to navigate the current system and figure out where things are moving. The way things are going is the model.

Podcast Host (42:48)

Then, if I'm hearing you correctly, it sounds like the model is actually similar in the public sector to the private sector, but maybe slower. Like it's still pockets where there's organizations operating at the edge, they're innovating effectively. And you know, you hope and pray that eventually everybody else says, oh, that looks like that is a better way to do it and they adopt it, or they have no choice but to adopt it. Is that still the model?

Salim Ismail (43:16)

That's somewhat the model. It's just more difficult in public sector because you have all these other requirements like audit trails and transparency. And public sector does not like to take risk. And taking risk is job one of any leader going forward. And so this is going to be very tricky. However, when you can do it and you can do sandbox experiments to try things out like stablecoins or whatever the experiment is, and then you can see if it works. If it works, you just roll it out as fast as you can, right? And you've got that luxury also. But I think the. If you go back to my point that I made at the very beginning, the cost of technology is dropping radically. Magical things are starting to happen. Let me give you a specific example. A few. A few years ago, if you were trying to get a wind turbine erected in Colorado in the U.S. okay, it was taking about two years to get the approval for that wind turbine. They had to check flight paths and are there any water mains or are there power lines? Da da da da da. And it took about two years to get approval. Okay, that was like a nightmare because people are, we need energy really badly. Why the hell does it take two years, somebody, one of the, in fact one of the applicants finally says, look, can I just help you out with this? Creates a Google map where they laid all the power lines and whatever, whatever. And you basically say that's where I'm putting my wind turbine. And the system looks it up and in 30 seconds evaluates whether it's okay or not and gives you your approval. Right. So a two year process shrank to 30 seconds. That's the opportunity in public sector. We can do so much more in a really radical way. I'm incredibly optimistic as to what the potential is, but you have to embrace it and you have to go for it and attack it. And what's incredibly powerful is that the cost will drop. And this is the part that most people don't get. A few years ago in the U.S. the Republican Party asked me to give a talk about at their national Leadership Conference. They do an annual get together with all the leaders. And I worked, was going back and forth with Eric Cantor was the speaker of the House at the time. And the talk would be the title of my talk was how would you drop the cost of government 10x within 10 years? Which you could do. And amazing what happens when you can drop the cost of government by 10x. That's amazing. Right now it failed because they said, well I said in order to do this you have to embrace technology. And he said, well we're Republicans, we can't do that. And so that was the end of that discussion at that time. But now people have no choice but to embrace the technology. And I think the potential is going to become huge. Let me give you a macro example. If you're the U.S. federal Reserve and you're trying to figure out okay, how do I drop inflation by 1%? Your reports that you're getting in about the economy are typically equal quarter or more late. So the data's out of date. B, you've got certain levers you can knobs and levers you can pull to achieve that. And honestly what you're doing is guessing. You're just, just completely guessing that we think this would work. We'll print this much more money, we'll try and do this. Da, da da, da. Okay, imagine you have an AI looking at all the financial transactions in the US in real time and has that many more levers and can say if you want to drop, here's the 10 things you would do. A human scans those 10 things says let's do those eight, those two, a little bit off the wall and boom, you have your Answer. Right. That's the kind of thing I expect to see that I think is going to be very, very exciting if you adopt that into government. And I think people will have no choice but to do it just because the pace of change will necessitate it. So I tend to be more optimistic than, say, mo on the future. But there's no question we're going through this very difficult, turbulent time, and then we'll see where we get to on the other end.

Podcast Host (46:58)

Let's talk about that a little bit more structurally, back to this notion of exponential organizations, which, you know, you've invested so much time and so much brainpower in. So maybe just like, again, to zoom out here, what does a successful exponential organization look like? Like, what are the key attributes there? And, you know, do you have any examples you can point to that help, you know, serve as a North Star?

Salim Ismail (47:23)

Oh, hundreds of them. Okay, so I was running Singularity University, right, 15 years ago, which was we were bringing the leading experts in the world in the fastest moving technologies and teaching the future. So we'd get CEOs, investors, executives, government leaders, and we'd say, hey, here's what's happening in biotech in the next three, five, seven years. Were these technologies intersecting? Which labs and companies are doing the most interesting work? What inflection points should you be tracking, et cetera, et cetera? And I realized that, wow, if you take all these technologies and the intersections, we have 20 Gutenberg moments hitting us at the same time because the printing press changed the world. And we call that a Gutenberg moment because it changed the world so radically. Well, AI changes the world. Drones change the world. Blockchain change the world. Life Extension will change the world. And so this is a huge tsunami of disruptive innovation hitting us all at the same time. How are we going to navigate this? The book came as a response to a question we started to get. So we'd have these executives for a week in the classroom. We give them this download at the end of the week, they'd say, okay, get all the disruption. My mind is blown. What the hell do I do on Monday? Like, how do I apply this? And the book was a response to that question. If the world is changing so quickly, how do you adapt to it? How do you organize for this new world? And the specific question we answered in the book is, how do you build an organization that can scale at the same pace that technology scales? We looked at 150 of the fastest growing unicorns in the world at the time. Spent a year and A half doing research and said, here's what they're doing, and teased out a model. We didn't invent anything. We just labeled what was actually going on. For example, TED was a single conference happening in Monterey. Thousand people a year, going great. Chris Anderson takes it over. He does three things. He establishes what we call a massive transformative purpose, which is ideas worth spreading. Then he puts all the TED tops on YouTube, leveraging rich media, and then he allows anybody in the TED community to go create a TEDx event in their local city. And then, boom, in like five years, he's created a global media brand. So he went from a single conference to global phenomena at a cost of near zero. Kind of incredible that you can do that, right? And so the definition of what we call an exponential organization is you created an organization that is scaling 10 times better, faster, cheaper than your peers in the same space. Okay? Tesla, for example, the number of moving parts in a combustion engine car is over 2,000 moving parts just in the drivetrain. A Tesla has 17 moving parts in the drivetrain. So in terms of maintenance, design, reliability, it's just three light years ahead. Okay? So that's the kind of thing we look for. A classic example that we use today is Airbnb. Okay? So if you think of the marginal cost of adding a room to an Airbnb is near zero for Airbnb. Some new user signs up, says, my bedroom's available, Boom, you're off to go. If you're Hyatt, you have to build a hotel, right? So what we figured out when the Internet came along, it allowed a whole host of companies to generate demand exponentially. You had the rise of Facebook, YouTube, YouTube, Instagram, et cetera, using network effects to drive user growth, right? What exponential organizations have figured out how to do is rather than dropping the cost of demand, dropping the cost of supply. And that's where Uber comes in. The cost of adding a car to their fleet is near zero. Same with Waze, same with Airbnb, et cetera. So you have this breed of startups now that are emerging that the cost of supply is dropping, collapsing. Now, if you drop your marginal cost of supply, your denominator reduces, your market cap explodes. That's why the valuations are so high. So from an economic thesis, this is very, very powerful. There was an economist in the 1930s, his name was Ronald Coase, and he C O A S E, if anybody wants to look at him. He posited the big companies existed because transaction costs inside a big company are cheaper than outside. A company company. And therefore firms are just becoming bigger, bigger and bigger. He won the Nobel Prize for this paper that he put together. It's called the Nature of the Firm. Well, in the second edition of the Exponential Organizations book, which came out three years ago, we declare cosa's law dead. Because the transaction costs outside a company are often now cheaper than inside a company. Right? In any company, try and get a web page built, it'll take you like months and approvals and costs where you can spend 50 bucks and get it done outside or zero and get Vercel to do it in five minutes today. So the cost of doing anything is dropping so radically means that we're changing the nature of organizations. And this is profoundly amazing because you can build new organizations for near zero cost. And AI is driving that curve very, very aggressively. So I've titled this new paper. I hope it's not going to become a book because I don't think I have it in me for another one. But the paper is going to be titled the Organizational Singularity. And the comet is when workflow, agentic AI workflows become self improving, which we have now. And the minute you have that, you can just let agents go, improve workflows, and you basically can sit back and do very little. And that's. I think the future of organizations is going to be built doing that.

Podcast Host (52:44)

It's. I'm, I'm a fan of that angle as well. And I'm just, I'm reflecting there. You know, there was so much meat in that answer. I feel like we covered like just an insane amount of ground in that, like, short.

Salim Ismail (52:56)

It's a mini MBA right there.

Podcast Host (52:58)

There you go, There you go. There was a question you asked right at the beginning, though, that's still rattling around in my brain, which is, what the hell do I do on Monday as a leader at one of these companies? And like, with all of this at play, whether it's agentic AI within your own organization, whether it's tiny startups that are able to break Coase's law and challenge your business model, what the hell do you do on Monday? Is it an innovation mindset that you need to focus on? Is it structural? How do you avoid, you know, being for a crassword like a loser in this space and make sure that you're actually a winner?

Salim Ismail (53:44)

Okay, so let me talk about this theoretically and then practically, okay, theoretically, you have to rethink the value of how things are getting done inside an organization. As I mentioned earlier, right now we have human centric Workflows, right? And that will fade away in the pace of AI, because AI will do things much more quickly and much more seamlessly than human beings can do it. And then so the human being becomes oversight, exception, handling, etc. Etc. We can see that today in, say, customer service, where you typically have three levels of support in customer service. Level one is easy questions that can be answered. Level two is they need some help resetting a password, whatever. Level 3 is really tough problem. Something's really not working. You need expert to bring in help. And AI is doing now level one and level two very well, means that you can focus your customer service on level three and deliver much higher care. Healthcare would be another example of where the doctor is spending more time with the patient or the teacher spending more time with the student. So the question becomes then, if that's the future, how do you navigate to that? And practically, this is where things become interesting. I've only, I've been in the boardroom of probably 150 of the Fortune 500, mostly yelling at the board, saying, wake up and smell the roses, et cetera. And you have to imagine that when this book, when my book first came out in 2014, it was a difficult conversation to tell people that they were going to be disrupted. Like, I talked to the CEO of a car company and they go, yeah, Tesla is never going to make it. Da, da, da, da, da. And I want to just point out something important here. We have a very, very difficult time with this concept of exponential change, okay? Mostly because we don't see it. Our brains cognitively do not understand this pattern. If I take 30 linear steps, I'll go to the other end of my house and you can all predict very well where I am a third or two thirds of the way, going a meter at a time for 30 meters. We're very good at linear extrapolation. All of our education, training, intuition about the world is linear. Okay? If you have exponential growth and I'm taking two steps, then four steps, then eight steps, and I'm doubling in every step. At step 30, I've gone a billion meters. I'm 26 times around the world, which is a bit further than the other end of my house, right? And very hard to gauge where I'm a third or two thirds of the way in that progression, okay? So we have a cognitive problem in seeing this curve. People don't see it, and then they get surprised. And that's the challenge. So job one is be aware of this pattern and get used to it. Let Me give you another example, just to triangulate. If I take a piece of 8.5 by 11 paper, it's like 0.1 millimeters thick, right? If I fold it, it's 0.2 millimeters thick. Fold it again, 0.4 millimeters thick. Here's a question for everybody. How thick is that piece of paper? If I fold it 50 times right now, you can run in a calculator on it, you can calculate that out. It's not intuitive at all, okay? Some people go, maybe it's this thick. Some people think it's the size of the room, et cetera, et cetera. It turns out at 13 steps, I'm a meter thick. At 20 steps, I'm a football field. At 38 steps, I'm gone around the world, and the 50th step, I've gone to the sun, okay? Now AI is doubling at that pattern, okay? So now hard to fold that 50th time granted, right? But the point is that most people will get to that distance, nobody will get to that distance, because you just can't get your head around that concept. And the reality of the world is it's moving at that pace. And so how do we adapt to this? So there's one way of doing this. That's the theoretical what do you do? What you do, if you're an existing company, there's only one path I've ever, ever, ever seen work. And now with AI, that becomes amplified. What you do is, is you take a. You create at the edge of your organization in a native digital twin, and you start building that organization at the edge of your existing company. And then you start moving workflows over to it, little by little. If you try and do it inside the mothership, you get the immune system response, the antibodies attack you, and you'll never again, you'll be stuck in a political fight. And you don't have time today to do, to make that assumption. So you do two things at the same time, and you, little by little, grow this other thing at the edge, move your most capable people over, and then deprecate the old as naturally and as elegantly as possible. And what will happen is some people will leave, go create their own startups, and they'll do very well. So the only real job description of the future is going to be entrepreneur.

Podcast Host (58:30)

Let's unpack that for a second. And I actually, I tend to agree with you on that one. When you say entrepreneur, there's the extreme way you can interpret that, which is companies of one or there's more of an entrepreneurial mindset where it's thinking like an owner. Is it the more extreme one? Is it first one, then the other? Can you unpack a little bit what

Salim Ismail (58:55)

looks like and where there's this idea of how having one person build a billion dollar startup with just a bunch of AIs is all very sexy. It's just not fun. You, you have the most fun when you're with a few people and you're figuring things out. You'll accelerate problem solving very quickly. Yeah, you could build a billion dollar company with one person or horde of AIs. It's just going to be a very lonely person. Right. So I think we'll end up with Companies in the 5 to 10 person range building billion dollar companies and they're going to do it very quickly and it'll just do more of them. Okay. I think that's more likely that we're going to see that. Now the way that occurs is you have end up with two classes of AI agents. Okay. Class one is the strategic agents that are doing market testing, market research testing, running experiments in the external world as sense making and figuring out strategy as a real time mechanism. And then a second class is a class of execution agents that are tuning product strategy, pricing models, delivery models, et cetera, to match that strategy. And then human beings doing oversight, as I said, exception handling, etc. Etc. That's what I think is going to be the future of the firm. And I think we'll just create a lot more companies doing things. And I think what going to happen is the cost is going to drop. Now what's really magical about this, and this is the hard part, is that we've been trained, you know, we've been doing education. I'll just switch to education for a minute here because we're going to see a massive collapse in education now. Right? Because higher education almost exclusively globally is job schooling. And it's all supply side. Like you go become an engineer, a doctor, a lawyer, a web designer, an accountant and you become, you get credentialed in that domain, say an accountant. And, and then you go to the job marketplace and you try and sell those skills in the job marketplace. That's essentially the paradigm in the world today, globally, bar none. We're moving for education from the demand side, supply side to the demand side. Today kids are saying what problem do I want to solve? And then getting the acquiring the skills, techniques, technologies that they need to solve that problem. Elon is the most easy, exemplified here. He's got three massive purposes. Get us to Mars, solve transportation, solve climate change, and he's building companies in each of them. Note he has no experience in the car sector, space sector, energy sector. He goes into each one with the beginner's mind, uses first principles and totally disrupts it. That's what's possible today. So the entrepreneur of the very near future is going to be a 16 year old kid who has a deep passion for some particular area and they're going to use AI to build a company with a few friends and they're going to go execute that and fulfill and scratch that itch. And that's what's going to be the predominant modality, I think going forward for business building going forward, find that passion and then go use AI to help solve that problem. And when you're doing that, it's not work, it's fun. And that's why I think people are going to be so much enjoying themselves doing this because it'll be fun doing this.

Podcast Host (62:09)

I think so too. And I'm a big advocate of that model and I think we're, I'm excited frankly to see some really amazing things come out of that shape of organization. If you're, you know, I don't want to call like just every company out there a legacy company, but it's starting to feel like we might be in that world right now. But if you're, you know, a traditional enterprise is your best bet then to basically replicate that model in house. You talked about the digital twin. Do you need to just find your own in house? Elon, Find five to ten of your best people and just say go out and build the next disruptor of our organization in house. So it's us instead of somebody else.

Salim Ismail (62:52)

By in house, I mean at the edge of the company.

Podcast Host (62:54)

Yeah, yeah, yeah, yeah, yeah. I just mean owned by the organization.

Salim Ismail (62:58)

But you're correct. So every organization has about 5% of the people that are super smart, super loyal, but completely unmanageable. Very creative, very loyal, great people. But you never know if they're going to get fired or they're just going to quit. Take those people, put them in the edge of a company and say your job is to red team this. Build an AI native organization that's doing exactly what we're doing and disrupt us. In fact, set two teams up. That way you distribute the risk even and then little by little move workflows over to that move clients, move product lines over there little by little and then you'll be there. And so that's I think the path that companies are going to have to follow. I really don't see any other alternative. If you try and do that in the mothership, it fails.

Podcast Host (63:46)

Yeah, I'm. No, I agree with you and I've got a background running, you know, edge of business innovation teams and we've had decent success with that. I wanted to slightly change the angle here and explore a phrase I've heard you use in the past, which is corporate metabolism. And basically how you can just as an organization, whether it's public sector, whether it's private sector, just again, to be crass, just get more shit done. And how with the, with the technology taking off exponentially, just being a slow moving business just feels like it's going to be more and more painful. And the reason I bring it up as a separate issue, Salim, is it feels like the building something truly innovative at the edge will pay off in a handful of years. And when it does, maybe your metabolism will be increased. But is there another angle you can take to try and increase it along the way?

Salim Ismail (64:46)

Yeah, to just.

Podcast Host (64:47)

Or to even get the idea approved in the first place?

Salim Ismail (64:50)

Yeah. So two things here. One is you have to do a bunch of things at the same time. Okay. One is you have to stop. If you do something at the edge, you have to stop the mothership from attacking it because that's the first thing that's going to happen, Right. People go, and I have this firsthand experience. I built Yahoo's incubator and ran it and I spent all my time fighting the company because the more disruptive an idea we came up with, the less the company could handle it, Right. I was like, wait, you hired us to do disruptive things and we deliver something to you and you can't cope what's going on here, Right? And Yahoo's. You could expect this at a bank or a telco, right? But Yahoo is a Silicon Valley company less than 10 years old. Why should it have this problem? And finally I met the CTO of Autodesk who said, oh, you have an immune system problem. Try anything disruptive, the antibodies are going to attack you. And it was like just visual gripping metaphor that was like. So we actually have solved that problem when? In 2015, soon after the book came out, I got a call from the CIO of Procter and Gamble who said, hey, we want to work with you. Book is required reading across the company, et cetera. And I said, I want to try this immune system thing. So they gave me a division and said, let's See, and we ran a 10 week engagement and we found a way of hacking culture at scale and increasing the metabolism in a legacy environment. We've done it 100 times now with big companies around the world, Black and Decker, hp, Visa, you name it. And we've reliably found a way of moving leadership, culture, management, thinking three years ahead in 10 weeks. We have an advisory capability around that. So as we started building this, I was like, okay, this is starting to smell like a consulting firm and God help me, I don't want to be doing that. So we built a community around it and we do all the consulting from the community. So we're like a distributed McKinsey is what we've built. So we now have 44,000 people in 150 countries. We train them in the methodologies, they use these tools and then they go transform companies, government, whatever, whatever. Essentially what we've been trying to build is like a Peace Corps. So as the world goes through this huge transition, you have a layer of people trained to how to navigate that, which is what we've done. So if you're a company, you have to do three things. One, run this immune system to block the attacking, B, set this thing up on the edge secret as possible, okay? And the third thing that people really struggle with, but it's really true, is that in the past when you wanted to do something, you had to do what was called make a big bet, had to put a lot of resources on the edge, the CFO hated you and would defund you as fast as they could, etc. But today people forget the cost of innovation is near zero. It just requires a crazy mindset to do something. So you do a portfolio approach of this at the edge. You don't even tell the cfo. And if something starts to succeed, you double down on it, right? So you use almost a venture approach to this. The third level of sophistication is once you see start something starting to succeed, don't bring it back in if it's really disruptive. It's not going to fit anywhere neatly. You're going to end up with a lot of politics. Spin it off and I'll give you a live example. Nestle created Nespresso and for several years tried to run it as a line of business, but it didn't fit in the main food processing behemoth. It's a different brand, different delivery, different customer proposition. Everything. They finally, after reading the book, they put the brick wall put in a separate building, said the Nespresso guys Go over there and boom, every hotel room in the world has a Nespresso machine now. And so that's the model that we see this occurring where you have companies like Facebook buying Instagram, buying WhatsApp, but leaving them on the edge, leaving the separate entity. And so you're building an ecosystem then rather than building one big operating company, which is what the future is going to be. That's why Google is breaking up into Alphabet and Google X spins off those startups. And so that's the model. But these now going forward will be AI native startups.

Podcast Host (68:45)

Tell me if this is an appropriate comparison or not. But it almost feels like when you use the Google example, when you use some of these examples, that CEOs need to start thinking about basically every company as though they're a vc, right? That it's a venture capital approach and they need to be making several bets at once and allocating resources among these bets. And not just we do one monolithic thing, don't touch it, and we're crossing our fingers. Is that fair?

Salim Ismail (69:12)

Yes, 100%. Let me give you a real example there. Cisco in the 1990s exploded and did unbelievably well for 10, 15, 20 years. The way they would do it is when a new technology emerged, like say Bluetooth, okay, What Cisco would do is they would set up two product teams inside the company to build Bluetooth products. Then they would fund Sequoia, which is their venture fund, and they'd say, you go fund two Bluetooth startups and we get to buy them at a pretty agreed price if one of them succeeds. So they get four cracks at the pinata. And so de risking it that way allowed them to essentially do what you're suggesting, which is kind of create a venture approach where you have multiple firms trying to the same technology. You can't lose, right? Because you've got a clear winner out of one of those four. And at the worst case, you've at least learned so much, you know what to acquire. If none of the work, and they explain exploded for like 15, 20 years, they stopped doing that and then they stalled.

Podcast Host (70:13)

Yeah, it's. I'm a big fan of that model. I'm just what I'm reflecting on. Salim. And you go ahead.

Salim Ismail (70:22)

There's a final piece that's really important to note here. What's really, really important to note is that as the world moves forward in this next few quarters and couple of years, you don't have a choice. The reason you don't have a choice is the markets are going to be changing so dramatically that you have to do something like this. It's not a luxury, it's not a thing, it's not even an option. You have to do it. And so we're going to be using our coaches to guide a lot of these companies as to how to go about doing this, because you don't really have a choice. Because as the world shifts into this AI modality, you'd better be right there, otherwise you're going to be left behind. Very quickly.

Podcast Host (70:59)

There's, as I was saying, I'm going to ask you a difficult question, so I'll preface it with that. One of the things that struck me about this, and I know you do and have worked with a lot of CEOs, I used to do a keynote about 10 or 12 years ago, probably around the time you were writing your first book, that talked a lot about Moneyball, if you know the story of the Oakland days and implications of that. But one of the things that struck me as I was giving that keynote and telling that story and got really familiar with it, is that Moneyball, and for those who don't know, Moneyball is the story of the Oakland A's and how their head coach basically took a very data driven, analytical approach to baseball, found out a new key metric and completely changed the way the team performed by exploiting something that other teams didn't know about. But what struck me about that is that the guy at the top was super bought in to doing something that was considered radical at the time. And one of the challenges I tend to hear a lot about is from people who are, call it C suite, call it executives, call it board members who they're bought into doing something radical. But there's one person in the way, whether it's the CEO or the cfo. What do you do when there's one person in the way? How can you be part of a change movement if you're not the one that ultimately is holding the pen or in the big chair?

Salim Ismail (72:25)

This is the hard problem of leadership today, is to take that big leap into the unknown and pull your team with you. If not, get rid of the team, you don't really have an option today. You can see the CEOs that are able to do that big shift, like Jeff Bezos does this, Elon Musk does this. Right? Eric Schmidt did this at Google. If you're, if you don't have people, if you've got anybody slowing you down, you have to get rid of them. Because anybody slowing you down today is a death knell for the entire company. Okay? The, the idea of kind of being cautious and safe is gone. In a world that is moving this quickly, you either have to be. There's only two modalities of companies in the future. You're either the disruptor or you're disrupted. There's no middle ground. So you pick. And by not picking, you're the disrupted. So it's a pretty easy discussion now. So these days when I talk to CEOs, it's not a question of oh my God, should I change, it's okay, what do I do? Right, because they can see now and it's easier, it's an easier conversation with doors, etc. I'll give you a funny story. A few years ago when we were doing these week long singularity events, there was a huge French company and the CEO said to me, hey, I'm going to send one of my board, my more forward thinking board members to your week long thing one by one and get them all juiced up. And I said, listen, don't send the most forward thinking ones and the most backward thinking board members because they're the ones that are slowing you down. So every quarter we had the guy from the French, this one of the board members, French company, he didn't know, you know, explicitly he was the post Pacquiao, but hopefully he's not listening to this. And we would break their brains and then they go back and stop being the most sluggish of the board members. That's not a great example. In that case you just need a new board. But this is a huge problem. Look at the board of any big company in the world and they should be put out faster for the most part. Right. They cannot get it. And this is an unbelievable shift that's happening. Great level of again, creative destruction, excited that's going to happen. But as I said before though, I think the bigger need is in the public sector because the scaffolding for civilization is really creaking right now and we really need our institutions to be rebuilt. And that's the work that we probably all have to do more than anything else.

Podcast Host (74:55)

That's, yeah, it's an interesting pivot and it's, it comes back to the problem of, sorry, I shouldn't call democracy a problem, but the, the fact that you've got the electorate and you know, they're the ones holding the, you know, the quote, board members of the public sector accountable. Right. All the, all the politicians and.

Salim Ismail (75:14)

Yeah, so the solution there by the way, is to Go to small democracies. We run the world. You know, in the lastly, if you went back from 500 years ago, it was really useful to be a big country. You could throw your weight around. You had critical mass in multiple industries, loss of natural resources, et cetera, et cetera. What we saw during the pandemic was all the smaller countries dealt very, very well with it, and all the big countries failed miserably at it because it was too difficult to bring the whole population around, have them pivot on a dime, do lockdowns, quickly, whatever. I think the future, what we see playing out in the world today is not so much nation states, but city states. So you take Brexit. Brexit was really about London versus the rest of the country. It wasn't left versus right. The big tension in the world today is urban versus rural. But if you have people, just imagine a future. You take a city like Montreal, if you have solar energy, vertical farming, satellite Internet, why do you need Quebec as a province or even the country really? The infrastructure, they're not delivering anything to you. And if you have energy abundance, then you really don't need to fight over anything. And cities kind of naturally think so we think the future of the world is going to be city states rather than nation states. We're just going through a final purge of the nation state construct in the last death throes of the oil economy right now to navigate that. So we see the future coming. The technologies are all there now. It's just a question of how quickly can we move to that model and move to that kind of Star Trek type of model.

Podcast Host (76:49)

There's an old quote that comes to mind when you say that which is the future is here, but just not evenly distributed.

Salim Ismail (76:54)

I don't even distribute William Gibson, of course.

Podcast Host (76:56)

Yeah. Which I think is a great quote, by the way. I'm not smart enough to have come up with it myself, but it makes me think about these kind of cities versus more rural regions and the tension that again, coming back to something we were talking about earlier in the conversation of being left behind economically. And there's regions that are getting. If the pace of change and the pace of growth is faster in a city, then you're just getting farther and farther ahead in a city than out in the country. And so in this model does it just. Is the implication, if you just play the clock forward, that everything needs to be cities? We, you know, as much as possible need to, you know, centralize on these nodes or how do you see that shaking out?

Salim Ismail (77:44)

I Guess geographically, so it can go a number of ways. But I'll give you a couple of the determining factors here that I see happening. Number one, we have autonomous air taxis, flying cars coming along, right? Which means I can live in a remote suburb and come into town in ten minutes. Okay, that makes it really interesting to live anywhere I want and then just pop in when I need to pop into the city. So I think that'll be one major drive for suburban areas because you have a very comfortable life and yet you can pop into the big city when you want to. Major innovation always happens where you cluster people together, just because the exchange of ideas, the friction between finding out things is much lower, etc. Etc. But you can actually have the best of both worlds where you can have live wherever you want and pop in very quickly for some major event. So when you can do that, you actually get the best of both worlds. The city is not taxed by having huge numbers of people living right at the center, which causes a mess, et cetera, et cetera. So I think there's going to be some wonderful things happening around those lines also with more and more electronic communication, like my CFOs in Spain, my CTOs in Silicon Valley, my head of community for ex. My exo world is in South Africa. We live a totally distributed model and it works. Everybody lives where they want and they. We get work done, sometimes own messiness, but other than that we get a lot more done. And everybody has a much better lifestyle by living where they choose or where they have family, et cetera, etc. Etc. So this is, I think, a pattern that's going to be much more pervasive than we thought. I do a lot of flying around, traveling. I wish that we could do everything virtually because it would save me a huge amount of time. But you know, conversations like this that I'm having with you, which we can reach a reasonable number of people, is such a profound opportunity compared to say 10 years ago, when you'd have to fly to an event and do it in person, etc. That I think I'm very optimistic about that world as it comes along. So I don't think you need to be in the. Have the density of a huge city for the future. You just have to have meeting places and events and gathering places for human beings. Because I think that won't, that won't deprecate anywhere near as we think quickly, as we think it will. We just love getting together. Just look at the buzz instead. Any stadium, right? People love being in Groups, There's a collective energy that comes about that's very powerful and very creative, that will drive a lot of innovation.

Podcast Host (80:20)

There's a tension in your answer that I want to tease out a little bit, which is people, wherever they choose to be, that's comfortable for them. And on that side, remote work and us talking like this versus you flying around the world, the stadium effect, the city effect, this has become kind of a hot button issue. And there's so much return to office right now and these proclamations that remote work is dead versus oh no, it's the wave of the future. How do you structure your thinking around that? Is it just somewhere in the middle and we need to address every decision thoughtfully, or are there some sort of framing devices around the best way to decide when it's worth flying somewhere, having a team, local versus, you know, I

Salim Ismail (81:12)

have a very clear framing. I've looked at this and we've looked at this as a company as well as in the ecosystem, et cetera, and I think I have a pretty clear way of looking at this. And the way I look at it is if you're doing breakthrough product design and product building and you're doing deep innovation, say you're SpaceX trying to figure out how to build rocket engines, right? You need to be in person. The speed of idea exchange and throwing up a whiteboard and discussing things as a group, solving problems together in real time is 10 to 50 times faster than if you did it in a distributed model. So if you're doing anything that's groundbreaking or at the edge, which is increasingly what we're doing, then you need to be in person. Person. However, the minute you move away from that, and now you're doing mission control for the launches of SpaceX, which is a standardized thing now, you can do that from anywhere, right? And AI will start to do a lot of that. So I think what's going to happen is for people doing really radical, groundbreaking stuff, research, etcetera, they'll do it together. And people that are doing more standardized roles, that are predictable and prescriptive, you'll do it, you can do it more remotely. And all of this comes with the AI wave that's coming, because that's going to automate a lot more of that, that prescriptive work, anything that can be described prescriptively, an AI robot is going to do it much better than you, right? And so, by the way, I think we're much further away from humanoid robots than most people think. Elon is there going, yeah, we'll have it within a couple of years. And I'm like, no, the liability issues and the edge cases are so ridiculous, it's going to take a lot longer.

Podcast Host (82:57)

I think that's true. And I mean, the other question I have for you around that is, to me, like, humanoid robots seem like a classic case of just, like, build what? You know, like, why. Why would the best robots be humanoids? You know, certainly, like, can there be a better form factor than just rebuilding humans? But they're robots.

Salim Ismail (83:15)

Have you seen our Moonshots podcast?

Podcast Host (83:19)

I've seen the podcast. I don't know that I've seen the episode you're talking about, but go ahead.

Salim Ismail (83:22)

So this comes up quite a lot where Peter is like, whoa. Well, I have humanoid robots. And I'm like, why in God's name do they have to be humanoids? You know, at least give me a socket for forearms rather than two. Anybody filling a garage bag almost wants. Always wants that third arm to hold the bag open. And why does it have to be bipedal? It can have wheels. You'll have so much better flexibility. So we have this ongoing debate on why do the robots need to be humanoid? They should be. I want to be doc octopus, right? Dr. Octopus. I want to have eight arms doing cool stuff. It seems absurd to me to lose that opportunity to do stuff, especially when you have the coordination capability that AI will give you. Anyway, just go for it. So we'll see where it ends up. It turns out the human form factor is really agile and flexible and can be used for many, many things. So that's one rationale for it. So if I'm a humanoid, I can do gardening and I can do cooking and I can wash dishes and I can fold clothes, right? Whereas if I have a robot gardener, it may not be able to wash full clothes in the same way. So that's one thesis for it. I think that's just limited thinking. It's very clear. We'll have very specialized robots doing window cleaning. You're not going to have a humanoid dangling off thing. And then we'll just keep doing more and more of those as the use cases need.

Podcast Host (84:40)

It got me thinking about, you know, AI and robotics. And, you know, to me, the first real test of that was drones, right? Which are, you know, it's functionally a type of robot that, you know, is often run by AI right now. And it's such like a. It's in some ways just like simple and primitive form factor, but there's. You can just do a million things with it. And things that I'm sure people never conceived of. And to me, it makes me believe that there's probably form factors a lot more simplistic than a full built out human body that can do an unbelievable number of things.

Salim Ismail (85:15)

100%. This is why we have such diversity in biological species. Right. Just in spiders, we have like 8 million different spiders doing evolved for different use cases. Why do we think we'll end up with one type of humanoid robot? That just makes no sense to me. So I think we'll end up with a lot more flavors and varieties than we think. The inner loop will come when we have robots building the energy, and with the energy, we'll build more robots, and that's that inner loop. Once you get that going, then the flywheel starts and we'll move things along very, very quickly from there. Yeah.

Podcast Host (85:48)

Now then you've got the real kind of technological singularity that's on the horizon,

Salim Ismail (85:53)

which, by the way, we believe we are in the middle of. In the podcast that we've been doing, we're pretty clear we're in the middle of it. It's not a point that's happening at some point. It started already. And it's just, look at the acceleration today. Today, by the time you have AI breakthroughs, by the time you've understood them, that's out of date. So we can't even absorb what's happening in AI. So that's the definition of a singularity, is that every model that you had broke for what's coming? And our models are now broken along those lines?

Podcast Host (86:25)

You know, one of the topics, you've talked about it on the podcast, it's, you know, pretty hot in general, is AGI. And I think about it relative to a singularity, because to me there's this struggle about is AGI a singular event or definition, is it a time period, or is the whole thing just sort of bullshit? And we've reached a level of technological sophistication where the phrase doesn't even matter anymore.

Salim Ismail (86:50)

So I call BS on the whole thing. And the reason I say that is multiple fold one. You know, the concept of the singularity is when human intelligence overtakes machine intelligence overtakes human intelligence. Right? And I've had a huge beef from that, from the beginning notwithstanding, I helped build Singularity University. We actually had one of the world's top physicists come and visit us for a day. And he spent a day with us and he said, sleem, I got to ask you a question. And he goes, I totally love what you're doing here. Amazing, fascinating, incredible, great stuff. But you're not really a university. You're not about the singularity. So what are you? And I was like, damn, we have a branding problem. The issue I have with what's called the technological singularity is first, we don't know what intelligence is. It turns out there's about a dozen facets of intelligence, like emotional intelligence, spatial intelligence, some people have linguistic or musical intelligence. You have the Eastern concept of presence or awareness. That's not in the equation anywhere. The intelligence as we have it right now is measured by the IQ test. People talk about AI is reaching an iq, whatever. The IQ test measures two things, the speed of thought processing and the ability to match concepts across frameworks. But if I'm a business leader, I'm using a huge amount of emotional intelligence when I make judgment calls about things. So that's just that you have a big problem in navigating this. Right. And as I said before, the minute something becomes prescriptive in AI or robots going to do it better than me anyway. So what are we talking about here? So I have a huge beef. It blows my mind that Silicon Valley has raised hundreds of billions of dollars for a concept that cannot be identified, tested or defined. That's. You got to have them for the salesmanship.

Podcast Host (88:36)

You're preaching to the choir here. And I mean, my view on it is that like, you know, in the 70s and 80s, you know, sure, when AI is some future thing way on the horizon and that level of reasoning doesn't exist, it's something to aspire to, but it feels like based on where we are right now, it's marketing hype and there's like, we're not going to cross some sort of magical line and people will be like, oh, that's AGI.

Salim Ismail (88:59)

No, we're not going to, because we. At last count, there were 14 different definitions of it. At last count. So I'll give you two just to. So people get their heads around it. Paul Graham has what he calls the IKEA test. That's his definition of AGI. Can you give a robot an IKEA box and have it build a shelf from. Right. Well, that doesn't feel like AGI to me. That just means you're putting Lego together. Okay, so that doesn't fit for me. Steve Wozniak, the co founder of Apple, these are high level thinkers, right? He has what's called the coffee machine test. Can a robot or an AI grind some coffee and make me an espresso? Okay, I'M like, well, that can be done already. I don't understand how you, how you equate that to AGI. That makes no sense to me. So I don't think people have any sense what they're talking about. And we have a huge. I think what'll happen is we're going to just expand what we mean by AGI. The goalposts just keep moving and eventually we'll go, oh, AGI is consciousness or sentience. And then we have a huge other bucket of definition problem because we don't know what consciousness is. We don't have a definition, we don't have a test. Right. I joke that if you talk to philosophers, a subset of consciousness is considered to be self awareness. Right. You look like you're self aware, so I attribute self awareness to you. I feel like I'm self aware, but my wife disagrees. And so it's hard to even have the conversation around these topics. I asked Ray Kurzweil this exact question and he had the most wonderful answer, which was, you know, language is a very thin pipe to discuss topics as complex as this. Which I thought was a brilliant answer, but totally sidesteps the. So I don't think we have any idea what we're talking about. We don't know what AGI is. We have no idea what we mean by it. I actually have a diagram. I'm just going to flash this up here. Okay. And I'll just. So I've kind of come up with like five or six branches of a tree, and the roots and the soil would be our cognitive ability to sense the world. But you have signal to noise detection, which would be what we call AI today. It's mostly detecting patterns in a mass of data. You have evolution, which is a whole other pattern of increased complexity and increased sophistication. You have collective intelligence. Like markets have collective collective intelligence. That's a whole different form of intelligence. You've got real world manipulation. So it turns out a lot of the reason we have brains is to physically move around in the world and adapt to it. Okay. I don't know if you've ever heard of this thing called the sea squirt. There's an organism called a sea squirt which runs around feeding on other things when it's a larval form. And then it becomes an adult. Plants itself on a rock and filter feeds from then on. Okay. And the first thing it does when it becomes an adult, it eats its own brain because it'll never need a brain again because it's never moving around anymore. Okay. And you look at trees or plants, they don't have brains in the way we do, it turns out. Squirrels, birds, mammals, etc. Brains mostly just to physically move around and adapt to our physical environments. So there's that whole other vector of it, and then you've got the whole consciousness qualia, the hard problem of subjective consciousness. So all of those four or five things are very, very distinct things, and we lump them all in this bundle called AGI. And I have huge issues with all of that.

Podcast Host (92:23)

Yeah, sorry. And to your point, to your point, it feels like maybe the next step is actually understanding what intelligence is before we try and, like, compare something else to intelligence.

Salim Ismail (92:33)

Yes. We have no idea what we're talking about.

Podcast Host (92:36)

Yeah, yeah. The human brain is deeply, deeply fascinating. And we could probably spend a long time talking about that. But I am enjoying this segment that's kind of become, you know, Salim calls bullshit on things. And I'm curious, you've picked a few now that you've dunked on, which I'm really enjoying. You've talked about humanoid robots, you've talked about, you know, AGI, AGI. Is there anything else kind of in your rotation right now that tends to get a lot of attention that you think is worth calling BS on?

Salim Ismail (93:05)

I think people, geopolitics, I call a lot of BS on. I think we're a lot of the geopolitics that we have right now, like fighting wars in Iran or whatever, are all operating on this old paradigm. And that keeps us in a. In a Mad Max mentality rather than moving to a Star Trek mentality. And I think we have the opportunity to move past it, notwithstanding, if somebody's out there saying, I'm going to kill you, you need to defend yourself and go get them. So totally understand that part of it. But I think my father. My massive purpose is to transform civilization. I did a TEDx talk in Toronto a few years ago titled how do you fix civilization? I was doing a workshop with my dad, who is 90 years old at the time, says, puts his hand up and goes, totally disagree with your fixing civilization idea. I'm like, wow, okay, do you not need to think we need to fix civilization? He says, yeah, of course we do. But I have an issue with the fixing part. My issue is when you say civilization. He said, we've not civilized the world, we've materialized the world. We're apes with tools operating in tribal paradigms, killing each other with more and more powerful weapons. We still have to do the work of Civilizing the world. I was like, wow, Wisdom Bob from the elders. You know, I think this is the part that we have to figure out now is moving past our tribal patterns that have ruled us for tens of thousands of years and moving to a more collective, harmonious model like you would see in Star Trek. There's a great little acronym in Star Trek when you talk to the Roddenberry family. It's called idic, which is the basic thesis inside story Star Trek, which is infinite diversity, infinite creativity. It's just a wonderful framing for how they viewed the world, episodes exploring the universe, et cetera, et cetera. And I think we would do well as a human species to kind of move towards that model.

Podcast Host (95:05)

I think we would too. I'm curious, you made an offhand comment earlier about running around and just ripping out people's amygdalas, which, you know, I, this probably be good for us as a species. But, but short of doing that, you know, there's, there's the age old problem, as you said, of you know, like monkey brain and you know, the most, you know, nuclear weapons and drones.

Salim Ismail (95:25)

Two, two, two quick points to that. E. O. Wilson, famous biologist, used to study ants, et cetera. He said, the problem with humanity is our emotions are paleolithic, our institutions are medieval and our technology is godlike. And that's the problem with humanity. I think that's such a great summary of. And all the problems come in the gaps in those layers. Right. Douglas Adams, who wrote Hitchhiker's Guide to the Galaxy, put it in, in his way. He said, anything in the world when you're born, we call that normal. Anything that's invented when you're young, we call that a career. Anything invented after you're 35 years old is just blanket bad for the world, just bad. I think that's such a great summary of things in terms of how we're able to cope with new patterns and new ideas and new technologies. Like you have to be below 25 years old to program open claw today, otherwise you're too freaked out by the potential of it and you don't touch it because you're too safety oriented. So there's all sorts of things happening, but it's clear we have to go past our, our human constrained mindsets that we've had for tens of thousands of years. Cutting out the amygdala is a bit of a dramatic one, but psychedelics is probably the next best thing to it that, to take a dose of mushrooms or whatever and just cleanse yourself out from all the cred and see your own divinity and then operate from there.

Podcast Host (96:54)

Believe it or not, that was the road I was going to take us down. And I know you've talked in the past about your at some point, year of psychedelics and the potential there. And when we talk about changing people's minds and opening their minds, you know, psychedelics are certainly one avenue, and for a lot of people, a pretty controversial avenue. I'm curious, based on your lived experience and on your research, is that something you would just broadly recommend as part of the solution or like what?

Salim Ismail (97:24)

Yeah, 100%. Let me explain why. So I go at this with a very data, scientific evidentiary basis. Okay? So about 10 years ago, we had a researcher come to singularity and what he was doing is putting giving people different dosages of different psychedelics and putting them through an FMRI machine so you can actually spot a single neuron in your brain firing in real time with what's called functional magnetic resonance imaging. Okay, so you put them through an FMR machine. See? Exactly. Oh, this dosage of mushrooms amplifies this neural circuitry or depresses this neural circuitry, whatever. So now you have a feedback loop, now you have data, which we did in the 70s when Timothy Leary and Steve Jobs and all these guys were doing psychedelics. They're just chucking it down, you know, they had anecdotal sense of what was, but no real data. But now we have real data. And then I came across the key molecule in ayahuasca, which is dmt. DMT stands for dimethyltryptamine. Okay, okay, this is amazing. Our brain naturally produces DMT and it's released by your brain twice in your life. Okay. Once at birth, once at death. Go figure that one out. Right? So it turns out we, our brain releases it to ease the transition into the world and out of the world. Okay. What they happen, what happened in the Amazon is they found if you take this root and this vine and pound it into a broth, you get a big hunk of DMT which goes through your blood brain barrier and therefore you get this near death experience. And what DMT does chemically and physiologically, it suppresses your parietal lobe, which is where your three dimensional sense of self sits. So when you suppress the parietal lobe, you have freer access to your higher divinity, which you don't have normally. You have to do a truckload of meditation, prayer, whirling, dervishes, whatever, kundalini to get to that. But you take a hunk of this and it gets you that instantly. It's a shortcut and you have a profound divine experience. And so when I look at the evidentiary nature of that, plus the prescriptive aspect of it is you're like, oh, this is fantastic. So my wife and I did like a year of this together, exploring this, these different modalities, and I'm a huge fan just because it's very clear what's happening inside the brain around it.

Podcast Host (99:41)

Well, it sounds like, you know, I'll just play some of this back to you and like, try and like slightly clean it up, but clean it up for like, I mean, people who are like thinking they're listening to a corporate podcast right now. But, like, what I heard is, you know, just like, you know, throwing pills down or tabs or what have you is not necessarily what's going to work. What's going to happen is frontier research and investments in the space to better understand how to leverage these tools that help us unlock this other side of ourselves.

Salim Ismail (100:21)

Yeah, and you can look at a simple example would be you take coffee in the morning and it makes you a bit more alert. This is just a 10x or 100x on the same basic pattern. It's just lighting up different parts of your brain in different ways. And now we have the medical imagery to show exactly what's happening. It's not a mystery anymore. We know exactly what's happening. For example, we now know that a single dose of DMT can cure depression. Like 100% cure it, you'll never have it again. So that's like magical stuff. So we just need the evidence will show up. I tend to be there's one modality to operate in the world that is coming, which is to be an erudite evidentiary, free thinker. So erudite, so you can articulate clearly evidentiary, so you're data driven, otherwise you're lost and you're guessing all the time. And free thinker, so you're not trapped by other paradigms of thought or other constraints. Like if you. The further you go in your education, your brain calcifies that much more. And you can't be that creative. Right. You have to undo all that type of learning. And so if you look at those three things and you have all the data and evidentiary basis for psychedelics, it's clear that this was worth looking at and testing out in different ways.

Podcast Host (101:34)

I really like freethinker as kind of the final word in that clause. And it was reminding me. I had a guest recently I think it was Chip Conley, who'd done some work with Airbnb, and he was quoting, there's this great Pablo Picasso quote, which is, computers are useless. All they do is give you answers. And he was saying, AI is useless. All it does is give you answers. But to me, the free thinking part feels like it's still really important, that being able to open your mind, ask interesting questions and have hypotheses and think about coming back to your whole theme about building something new.

Salim Ismail (102:12)

Right.

Podcast Host (102:12)

Groundbreaking in some way that feels like something that we're not ready, anywhere near ready to give the reins over to technology on.

Salim Ismail (102:25)

You can guide it in powerful ways,

Podcast Host (102:29)

but it needs a guide, right?

Salim Ismail (102:30)

It needs a guide, but less and less, by the way. Okay, so right now there are what are called dark labs, dark research labs. So imagine a biotech lab where you've got jars of different compounds and test tubes, and a robot can fill up different amounts of compounds in a test tube, right? So you've got dark labs where robots and AI are coming up with different compounds, testing them in a lab, coming up with the experiment, running the experiment, observing the results, reaching conclusions, and doing the whole thing autonomously on a 24. 7 cycle. Okay, that's happening right now. So that means the entire R and D cycle of coming up with a hypothesis, et cetera, is actually being done by AI now. And we're solving major challenges. So it's pretty clear this is part of the reason why people are getting so freaked out and correspondingly excited by AI. It's pretty clear we'll be able to solve all mathematics with AI in the next little while. When you solve mathematics, you also solve material science and you solve physics and you solve all sorts of other stuff. So we're coming into an era very rapidly. And I'm not talking years, I'm talking months now where pretty much everything gets solved. And so that's kind of like this crazy situation where we'll be able to do pretty much anything we want in a very quick, quick way. Energy generation. Yeah, done. A climate change solution. Yeah, done. So the people are getting very, very excited by that capability. What's called that inner loop will start to expand and hit more and more areas very quickly. Nobody would have believed you could automate all software development two years ago, right? And here we are. Boom. And so it's kind of amazing to watch what's happening as we automate R and D and automate. We thought, okay, you may be able to run an experiment with an AI, but the hypothesis still has to come from a human being. But no, now the AIs are generating hypotheses just as well. And so it's kind of an amazing time to be alive. You can freak out about it or you can go, this is incredible. Which mindset do you choose to pick? When we wrote the second edition of Exponential Organizations, we close it with a chapter on mindset, and we identify our five or six mindsets that we think are absolutely critical for the future. For example, a curiosity mindset. One is a gratitude mindset because the world is infinitely better than we've ever had it. You just wouldn't know that if you watch the news. But really, we're in an amazing place overall,

Podcast Host (105:08)

tying a few different themes together that we just talked about. What's one view on the future that you've changed your mind on in the past year or so?

Salim Ismail (105:22)

It would. It would. You'll laugh to hear me say this, but it's about humanoid robots. And I say that because. What. There are two things I'll tell you. One is that you have robots when you have a hive mind and a collective AI across all the robots. When one robot learns how to do a task like a complex knee surgery, all other surgical robots now know how to do that task because they'll have that shared collective learning. And that blows your mind when you think about it and you see that with the full self driving cars today, one Tesla learned something, and they're sharing that with all the other Teslas out there. And we're getting to a very rapid collective hive mind acceleration. I didn't think that was going to happen for a very long time. And it looks like we're there now. And in a related fashion, one thing that really changed for me over the last, say, six months or so in the discussions we've been having amongst ourselves on the podcast and others, is the fact that you have this concern that, okay, if robots start doing all the work, then economic activity drops. GDP collapses. And what changed my mind around that was the fact that even though you have so much work being done automatically and autonomously, yeah, GDP collapses, but so much more work enters the economic realm. So, for example, right now I drive my kid to work school every day, and that's not measured by economic activity. Okay? But once I have a robo taxi taking him to school every day, God bless the day, when that appears, then that activity becomes part of the economic GDP fabric and you have what? What's called Javon's Paradox. When you automate something, you just do a Ton more of it. And therefore the over. That's where you have an over capacity and over or a much bigger demand than you thought. And what I'm realizing is it's the same thing with AI. We'll apply AI to so many areas that we can't even think of right now that you won't, you won't lack from a need for more AI. Therefore, the demand for data centers, energy, etc. Just keeps going. There's no such thing as an AI bubble. We're using every little last circuitry of anything we can find to throw at different LLMs and different problems. So those are two areas that have really changed for me over the last few years.

Podcast Host (107:49)

Nice. And I, yeah, the point there about consumption and demand changing in the presence of abundance. I mean, it feels like that's been proven out again and again. I mean, I'm sure it has in decades and centuries past, but you know, we've watched even the last handful of years in terms of how people consume things. You know, did you use the example of music about like.

Salim Ismail (108:09)

No.

Podcast Host (108:09)

Now you can just listen to music, have all the music you want all the time, and you know, access to data, access to media, like, it just completely changes the curve. And so it's exciting in some ways to wonder what we'll be able to do in an abundance world.

Salim Ismail (108:26)

I have a son who's 14 years old and I'm so, so jealous. I mean, when I was 14, you didn't have any of this stuff. To be able to just build anything you want and just dream up something and have it emerge and the kids today have such unbelievable potential in front of them. It's super incredible to see.

Podcast Host (108:49)

Are there any, you know, just because your son is that age, are there any things that you use as guideposts in terms of what we need to do, in terms of how we raise our own kids today to make sure that, you know, the future really is better for them. And we don't let unlimited unmetered access to everything, you know, guide them astray or prevent them from living a fulfilling life.

Salim Ismail (109:16)

Two things that I found are incredibly important if you have kids today. Number one, we've been telling kids for hundreds of years, go get a deep skill, as I mentioned before, an accountant, a doctor, a lawyer, and then find the job marketplace and go get trained up in some career that will be in demand when you enter the workforce. Right? And that's gone now. Right now we're seeing parents tell their kids, pick what problem is most exciting to you and Go attack that problem. And so we're shifting education from a push based education systems to pull based learning systems where they'll pick a problem and go get the training techniques to solve that problem. And I think that's one very, very big shift. The second big dynamic I would see is the. And it's actually quite an incredible damage being done by digital devices. We tried to put technology in classrooms and have smart boards and have kids have tablets. It turns out that's ruining the kids because the kids need much more tactile learning when they're young than just a digital device that you can engage with. It's so Jonathan Haidt has written books about this. Lots of people are seeing this problem. We're seeing this in real time where kids should not have devices until they're 16, 17, 18 years old and their neocortex is formed. It goes back to that model I mentioned earlier, where a lot of the reason for growing up is be out in the physical world learning how to be in the physical environment. And when kids are stuck in a device. We sent our son to go hang out with some of his friends, like last weekend, and he sends us back this photograph of a basement in the dark with six kids sitting with their phones individually, not talking to each other, and all you could see is the glow of six cell phones. You were like, wow, they're not even talking to each other. Right? And so this is the antithesis of what it is to really what they really need to be growing up. They need to be out there throwing sticks around and beating each other with stones and like moving that physical body around to understand how to navigate the world. And they're losing that in today's world. So that part of it, the physicality and being out in the world in a, in a tactile way, and then the, the whole moving to the demand side would be the two big things I would say for any parent today.

Podcast Host (111:45)

I really like both of those. And as you painted the picture about like the kids in the basement as well. It's just like one of the things I see, and I've got a couple of kids younger than yours. One of the things I'm hoping that we can sort out sooner rather than later is this kind of degradation of what I'll call shared experience, that you've got a shared space, but everybody's having a unique experience. And these damn phones now just feed you what they think you want, which may be different from what your partner on the other side of the couch wants or what your friends. And so everybody is having these unique consumption habits and what that foundationally does to us as a society, if we lose the ability to relate to each other and if we lose the desire to relate to each other because people aren't feeding that as much as the device is.

Salim Ismail (112:34)

Yes, very great points. Right. And this is why people are starting to date AIs and get married to AIs, et cetera, et cetera. I mentioned on a podcast with I was on a podcast with Tony Robbins recently and the point came up that when you get married, for example, a lot of the reason for being married and being a long term intimate relationship is to take your deepest issues and work them out in the context of that relationship. So if you're dating an AI, where's the opportunity for any of that? Coming up. So it's a very surreal world we're entering right now.

Podcast Host (113:07)

Yeah. Yeah. It's going to be an awfully interesting time. Hey, I'm just being mindful of the clock as well. Salim. This has been a really interesting conversation. I've had a ton of fun. We've covered so much ground here. I really appreciate you joining us.

Salim Ismail (113:24)

It's great to have been here. Great questions.

Podcast Host (113:27)

If you work in it, Infotech Research Group is a name you need to know no matter what your needs are. Infotech has you covered AI strategy, covered disaster recovery, covered vendor negotiation, covered. Infotech supports you with the best practice research and a team of analysts standing by ready to help you tackle your toughest challenges. Check it out at the link below. And don't forget to like and subscribe.