C (2:25)

And the chances that what that center

B (2:27)

is doing in Israel is along the lines of what I've described is again, extraordinarily high. These are the things that Israel have

C (2:33)

done well in historically, and these are

B (2:34)

the things that we think Israel is going to continue to do well. And so we've built a practice, a vc, that really focuses on these domains with our entire team being ex engineers and we lead deals with basically our peers, our cohorts from the military and other areas that we just think these are the best people getting behind really exciting ideas.

C (2:55)

And then to your point, today we're

B (2:56)

seeing a really interesting conversion between what used to be two very different domains, cybersecurity and data AI in the form of whether it's anthropic or all these LLMs coming out.

C (3:10)

And what we're seeing really is that

B (3:11)

some of these AI companies are providing both cybersecurity attack verticals and defense verticals that the cyber companies themselves didn't imagine. And so what I think we're likely to see, and I'm projecting here the next 18 to 36 months, is we're going to see some cyber, both startups and large corporations get hurt very badly because they were built in an era where McAfee so your antivirus was the main way you needed to provide cybersecurity. And it's just becoming less and less relevant. I mean, I literally sat with the CTO of one of the biggest Israeli cyber companies earlier today. But at the same time, I think the potential for cyber solutions, if you can get that right, is only increasing because the attack surface of digitization is growing exponentially. So the potential is there, but we're going to need to see very, very different types of solutions.

A (4:05)

Right. And so how much sort of exists, we're all obviously now aware of Mythos and people are rightly freaked out about it. I think what we haven't heard about yet is, okay, that's, you know, there is the technology now that can kind of infiltrate a lot of systems. What's coming up on the other side that could stop it?

B (4:27)

So I think it's going to be something very different. I think the pivots that we're seeing, a bit like SaaS, companies who just turn around and say, yeah, yeah, we're doing AI as well, have got the wrong end of the stick. I think the ones that are building things inherently different from the bottom up are the likely to be the types of companies that succeed in an extraordinary way into the future.

C (4:49)

So I'll give an example.

B (4:52)

When you think about software historically, it's obviously been written by humans. And if you go to nearly any cybersecurity pitch deck, whether it's 25 years ago, when the cybersecurity vertical was starting, and even to five years ago, maybe even three years ago, I know this because I saw hundreds of pictures of cyber startups. They always used to say the number one issue is human error. Human error, human error, human error. Whether it's human error in the way they're connecting virtual machines, virtual sort of

C (5:24)

code or whatever it is, and today,

B (5:28)

when 50, then 60, then 80, and soon 90% of the code is going to be written by AI, saying that all the mistakes are going to come from human error.

A (5:38)

That doesn't make sense. Right.

C (5:39)

It's just wrong. So the types of solutions you're going to need, and I wrote a piece

B (5:46)

on this just over the weekend, but the types of solutions you're going to need are going to have to be inherently very, very, very different. Both on the KPIs, the way you measure success for these types of startups, are you just scaring the ciso, or is there an actual output closer to what we used to see in sort of cloud, where it's measurable and not just selling some doomsday scenario that they're willing to pay $100,000 to try and say that we've done advance for it. And I think by definition security will look very, very different in the future.

A (6:19)

So that makes sense to me. You're seeing like, I don't really see pitches from infrastructure cybersecurity startups because it's not what I invest in, but you do. So you see this all the time. Totally accept the notion that a company that's been doing this stuff for a long time and is trying to rejigger their system to. To be the AI company usually doesn't work. Of the new companies, are you seeing new ideas and new approaches and new tech that you do think can be effective in combating not human error, but sort of AI coded, you know, not even error anymore, but risk?

B (6:59)

Yes, 100%. So I'll give an example. So one of the biggest verticals in cybersecurity is endpoint management. If you go to any S&P 500 or literally any company with more than a thousand employees and maybe even less than that, they will have a product that does their endpoint management system. So it's the system that this chief security officer or whoever you have running the IT can say this person is able to access this database. I am providing you a administrative sort of. We all know that figure, right? In an agentic world, you probably have more agents and potentially tens or hundreds times more. So like different order of magnitude of agents running within your system and you need to manage their accesses, their databases, which of the agents can do what? So the endpoint management system by definition is going to be very different.

C (7:56)

If you think about the number of

B (7:57)

employees that Google has today, it's like hundreds of thousands. The number that antropic has is 2,500, maybe 3,000. But if you think about the number of agents Antropic has, that is in the potential millions.

C (8:14)

So when you think about like the

B (8:15)

endpoint management system that a company, a futuristic company that's built more sort of from the ground up like Antropic is, their cyber endpoint management system is going to have to be inherently different than the one that Google started and built into their system. And I pick Google as a relatively forward thinking company. Like if I was to pick Siemens or you pick yours, it's probably going to be even more extreme.

A (8:37)

So then if, if you're Google and your current approach to cybersecurity is outdated because it's based on a, what is now a fundamental misconception, what do you do? Are those the companies that you are investing in that you think have solutions that can credibly stop Mythos or whatever else?

C (9:00)

Yes.

B (9:01)

So what we have in our fund is we have roughly 130 executives, primarily from the S&P 500. So it's either chief security officers or chief data or AI officers that are part of our team that sort of help our portfolio companies. Then we are at least once a quarter speaking to every single one of them to understand like where they see the market, what are they currently spending on, where do they see the current gaps? And it's because we're very focused that we really leverage this network. And so if I was Google, and this is what companies like Google are going to be doing specifically in the endpoint management system, which is really just one vertical with inside. But what is likely to happen to a company like that is they will probably for a period of time be taking more than one endpoint management system. I mean, I'll tell you a secret, companies like Google anyway have more than one endpoint management system. It is so critical to their business. They typically take more than one and have an overlap. But what they will end up doing is taking an agentic one to start off with, meaning one that helps them specifically for agents. And that company will claim that they can also do humans at the beginning, but maybe they haven't got all the functionality that some of the old endpoint management systems do. But there'll be a hybrid period and then over time you can literally see like the startups of today and the future leaders of tomorrow potentially sort of eating the lunch of the current providers.

A (10:25)

So let's say a listener to this podcast who has a business and it's not a multi trillion dollar business like Google saying, I'm aware of the risk that exists right now, what should I be doing about it today? What's your advice to them?

B (10:39)

So yeah, so like specifically for cybersecurity, if I'm a business, the question I have to ask myself is like, what is my business? What is unique about my business? Because frankly, if I'm not using any

C (10:50)

agents and I'm not planning to put

B (10:53)

anything agentic into production, which maybe I should be asking the question of why not? But let's say I just decide I'm not doing that in my business at all. Having an agentic endpoint management system is not a wise solution for me. I may have other issues, which is some of the cybersecurity threats I might be receiving, I might be on the inbound side, might be from agents themselves. And then I need a sort of defense mechanism which is much more suited to agents targeting, trying to find loopholes, etc. And as you mentioned earlier that we've already seen that some of these agents are much better and much more powerful than humans. So maybe I would focus my dollars there then on a agentic endpoint management.

A (11:36)

Are there particular companies and systems right now that are available that you recommend people take a look at whether they're your companies or not?

B (11:44)

Yeah, I think it really depends on what business you are. So if you are a business that is forward thinking and you typically are amongst the first users of products like this, then yes, we've got a whole bunch of startups in the space and there's a variety of other companies that are good as well and cybersecurity is really just an enormous space, I would ask some follow up questions. Yeah, if you're typically a lagging company and you're just waiting to see what works, then probably I would be waiting 6 to 12 months and then making a decision.

A (12:17)

Got it from where we are today. So let's turn to the book and what's interesting is you take a particular concept in machine learning and then use it to critic framework for human decision making. So let's start with what is a local maximum?

B (12:33)

Yeah, so local maximum is something that any of the engineers or anybody who's on a first degree in maths, and it's not complicated maths, will be very familiar with the notion. The easy way to conceptualize it, and this is the way I initially stumbled across it, is imagine you are a paratrooper and you are walking in the desert, and your goal is to try and get, let's say, the Sahara desert. Your goal is to get to the peak of the tallest mountain. So I've given you a mission. You've got 10 hours. Get to the tallest mountain, peak of the mountain that you can, and suddenly you see a mountain that looks really, really tall. You start climbing up, you get to the top of the peak, only to realize that there is a taller peak somewhere else. You are now stuck in a local maximum. You are not in the global maximum. You're not in the best place you can be. But any step you take right now, given that you're at a peak, is going to necessarily have to be down. And this is a challenge that a lot of us run into, whether it's with our businesses, with our personal lives,

C (13:29)

and I see this with a lot

B (13:30)

of founders, where they end up running up mountains that looked very good to them, only to realize that there are better mountains, there are better positions for their companies, for whatever they're trying to reach. And the challenge for them is how do they gain the strength or the mental strength or the ability within their frameworks that they're working to both realize this as early as possible. And then obviously, you have to be able to go down off the mountain you're on to potentially go up one that is dramatically bigger. And the reason this is.

A (14:00)

Sorry, I was just like, in game theory, the. One of the things that, you know, I have found interesting is the notion of don't worry about sunk costs, right? So like, you're, you know, you're waiting for the subway. It's been a while. It's still not necessarily coming anytime soon from what you can tell. But you're like, oh, I've been waiting all this time already. It would now be silly for me to get out and hop in a cab because I've already invested all this time standing on the platform. But ultimately, if your goal is to get from point A to point B as quickly as possible, using that as a re justification to not change course, doesn't make sense. How would something like, I mean, if you're literally at the top of the wrong peak. Is there an issue where people tend to sort of insist on staying there because they already put in all this work and the only right answer is to start going down as quickly as possible? Or should they be thinking about it differently?

C (14:55)

Yeah, it's an excellent question.

B (14:57)

I'll make two quick points. The first one is the example you gave is great. It's one that I have a chapter on it in the book where you're not actually at a local maximum.

C (15:06)

Like when you're at the bus stop,

B (15:07)

you're not really at a local maximum. The reason you are trapping yourself, it's like a psychological local maximum because of the sunk cost, but you're not really at a local maximum. A local maximum that I would like just to give an example would be, I'm a marketing executive. I've gone into marketing at the age

C (15:23)

of 23, did 20 years at it,

B (15:26)

making a really great salary within marketing. But I realize that my dream is today is not to be in marketing anymore and it's to do something completely different. And my challenge is within the marketing space. I can get a fantastic salary and everybody treats me really well and all the rest of it. But for me to actually do what I want to do, I don't know, teaching or maybe just do something completely different, I have to actually be willing to take a hit to my ego, to my paycheck, to whatever it is to go down to, then go to some point which may make me much

C (15:54)

happier, more money, whatever I'm optimizing for.

B (15:57)

So, like, that would be a more classic local maximum. You have to go down, back up.

C (16:00)

Similar with startups and what I try

B (16:04)

to do in the book, and the reason I got sort of attached to this is because as somebody who does

C (16:07)

a lot of programming, this is something

B (16:09)

that machines, AI, machine learning come up against a lot.

C (16:15)

What is unique about machines is, is

B (16:18)

that they climb really, really quickly. If you're telling a machine to like optimize, do a B tests, meaning move up a mountain. Typically, even if it's a very big mountain, the machine can do it in a matter of like milliseconds.

C (16:33)

And the challenge that machines and engineers

B (16:35)

have is trying to decide which mountain to climb and when not to climb the mountain or to get off the mountain they're on. So to give a machine example is Walmart is constantly optimizing the search results that it's giving you when you click in the search, I'm looking for, I don't know, presents leading up to the holiday season for my 10 year old kid.

C (16:59)

And Walmart within their sort of roughly one second has to make up its

B (17:04)

mind on what is the best results to give you. And it has to start climbing a mountain. It has to say, okay, I'm going to give you these types of results in the toy section for 10 year olds, etc. And at some point, usually in the first third of that period, that 1 second or fraction of a second, it is already chosen a mountain and it will then be climbing and refining and refining.

C (17:28)

And what the engineers are trying to

B (17:29)

do the whole time is try and figure out what is the right ratio

C (17:33)

between choosing mountains, climbing mountains, maybe at

B (17:36)

the very end of the second it realizes it climbed the wrong mountain, should it be leaping to a different mountain?

C (17:41)

And so what I've tried to do in the book is take the techniques

B (17:44)

that the engineers at Walmart, at Amazon and the big tech companies are using to optimize their algorithms to avoid local

C (17:52)

macronoms and to try and translate it

B (17:54)

for humans who are trying to just live better lives, run more successful businesses.

A (18:00)

Right, so how, okay, so, so now let's assume that you know, you're not the algorithm for Walmart for holiday shopping, you're a human being. How do you one, know when you're stuck in a local maxima? And then two, you know, what should your decision making process be to realize that and then to switch course?

C (18:23)

Yeah, amazing. So I'll give an example here. So I really just organized the chapters as practical techniques which are all translated

B (18:30)

from the world of technology.

C (18:32)

So one of my favorite ones I called abx. And most of us will know what a B testing is. And that's when you take two ideas and you compare the two of them together. People know this a lot in marketing. I'm not sure which brand slogan, title campaign is better. So I come to a bunch of

B (18:47)

people, I try option A and I try option B.

C (18:50)

And usually these two options have to be very similar to each other, but with a small difference. And slowly, slowly I keep doing ABAB testing against each other and eventually I get to a marketing slogan or whatever I'm trying to produce where I can't improve anymore on the A B testing. And if you think about what I just said is I climbed a mountain. I started off at one point, I did A versus B. A was slightly better, so I took a step up, B was next better, I took another step up and eventually I can't move up any further. And what techies, be it like the Amazons, the Walmarts, etc in the world have learned to do is not a B testing but ABX testing, meaning they will constantly do ab.

B (19:31)

AB testing, okay, we know that works,

C (19:33)

but every so often purposely put into

B (19:36)

their algorithm a very, very different slogan, something left field in an extreme way.

C (19:44)

And what this is teaching them is are they on a good mountain? If I have done 50 AB tests and my first X scores close to to my best result from my AB testing, it tells me that I should spend more time searching for other mountains

B (20:02)

for completely different slogans.

C (20:04)

If my ex came back and it was really poor, it gives me more

B (20:08)

confidence to carry on with my current AB testing.

C (20:11)

And there's ratios of how many X's you should put into your abs and all the rest of that. But like a learning for all of us humans out there with our day

B (20:18)

to day lives is how many X's

C (20:21)

are we putting into our business. Things that are really left field where even if it doesn't work, it's teaching me something about the mountain range that I'm currently sort of operating within. And maybe it will give me way

B (20:34)

more confidence to continue or it will

C (20:37)

open up an insight that I never

B (20:39)

sort of heard before.

C (20:40)

Just by the way on that, I think the most famous marketing slogan in the history literally of the world, which is to the best of my knowledge,

B (20:49)

Nike's Just do it came about the

C (20:52)

covers because of an X. And yeah, there's a whole story.

A (20:57)

Yeah, so, so okay. But given that and this kind of what makes all this so fascinating is that we're taking a sort of rule that works in mathematics, works in coding and everything else, and now applying it to human decision making. Humans obviously are much more variable, right. As you think about like Kahneman's work, like we're not optimized to pick the most rational, efficient choice at all times because we have emotions, we have, you know, different brain chemistry, we have your microbiome effects. So there are so many different things that impact your thinking, your emotions, everything, your intellect, everything else. So I think about sort of the inserting more X's and they're sort of different personality types, right. So for me, and this explains why I work in Venture, I love the axes, right? Like that to me is what's interesting and exciting. And I actually would rather have lots of different ideas and tests and risks and see what works and what doesn't work and even live with the failures of it than just to stick with what I already know and just, you know, just have that. But there are other people who are the Opposite that more people are the opposite, where they don't really like, even if they might be able to learn something different or better, the amount of uncertainty creates so much anxiety for them that it's not optimal for them. So given the variability of human beings and that, you know, the normal notions of logic and efficiency don't always apply, how do you reconcile all that?

C (22:31)

Yes, it's a really fair question. And like, I'll give you, I'll up you one, which is, I haven't even said what we're optimizing for. Some people might be optimizing for money, others might be optimizing for influence. And some, someone might just say it's happiness. What I'm trying to do is say you pick what you want to optimize.

A (22:50)

Since we're now just debating, can I challenge that? Which is, yeah, I would argue that the people who think they're optimized. And again, I'm not talking about optimizing money so you can put food on the table. Right. And that's probably not the people, they've got bigger problems to worry about than sort of, you know, local maximums to determine their decision making process. People who would say I'm optimizing for money or influence. And then if you push them, you say, okay, well why do you want those things? And they would eventually land at because it will make me happy right. Now those may be totally misguided notions, but they believe that. So I actually don't think, unless we're talking about meeting basic needs, that there is variability in that. I think everyone is trying to get to the same place. And obviously people have different things that make them happy. And obviously people have misguided notions that I think, you know, as much as you and I literally have the title capitalist in our job description, capitalism oftentimes pushes the wrong notions around this of what actually creates happiness. But, but couldn't you actually sort of just make the assumption that people are optimizing for happiness and, and then try to use that to guide them?

C (24:03)

So I'm not sure I agree with your general assumption about people. But like for me, this could be a business that's optimizing for a certain outcome and should the business be adding risk or reducing risk? And how many businesses are currently trying X's within their product line versus AB testing? And a business obviously operates and under a very different matter. So if it's a for profit, it's probably optimizing, at least to an extent for profit. If it's a nonprofit, the KPI Might be something else. Yeah, I'll give you just an example that one of my pet peeves amongst some of these businesses are that people talk about their job in R and D research and development, and they sort of couple those things together, when in reality, there could not be a bigger difference between researchers and developers. Developers are running up the mountain. They literally talk about sprints. They're trying to crank out product as quick as possible. And if you go to a company that's doing real research, not like fake research, because we have a bunch of researchers, but real research, they're not cranking up product. They're looking for new mountains, higher mountains, ideally to potentially expand the horizon and help the company with pivoting or finding new land.

A (25:12)

So, so one, one quick. I know I keep interrupting, but just because I'm the reason why, fascinated by this, where I have, you know, one of the things that has really been discouraging to me is Washington's, you know, very much led by Trump, in this case, reduction of funding for research at higher education. And why, I think, you know, and again, I don't like the way a lot of higher ed institutions treat Israel and Jews. But putting that aside, like the companies that you and I invest in, most of the time they're developing, they're not researching. And venture capitalists were commercializing research. Right. We're not actually doing the research. And so, like, the innovation economy doesn't exist if the building blocks of research aren't there. And those aren't happening, by and large at companies. They're happening in academia.

B (26:07)

You're right.

C (26:08)

I think in the very large companies, they are spending.

A (26:10)

Yeah, yeah. I was, as I was saying it, I was thinking that. So. But let's put a handful of, you know, 50 companies aside.

C (26:16)

I mean, a good place in Israel, to give an example, is the Weizmann Institute, where you walk around there, they're really proud to be spending a huge portion of their both capital and best minds on what they call basic science that afterwards people will hopefully be able to commercialize. But they're trying to push the front of humanity just further.

A (26:35)

And they're specializing in the X's, which is what allows for discovery. Right. Whether it's the marketing slogan or, you know, think of how many different things over the course of history have been discovered by accident. Right? They just something, you know, they left the lab and something mutated in a weird way because someone forgot to put it in the fridge or whatever it is. And they realize, holy, we can cure penicillin, we can Create penicillin and cure polio or whatever, you know, so yeah, so, absolutely. So, all right, so we're solving for different things and okay, so let's assume that we're not in the individual context here, we're in the business context and, or professional. And so different types of businesses and different types of sectors have different things that they're solving for. What should people know to keep in front of mind as they're figuring things out so that they can take, take the underlying thesis you have and use it.

C (27:32)

So yeah, like, like an algorithm. The way I've written this, it's more like a toolbox rather than a simple sort of one. And done. So if you think about a toolbox, you have to see what the problem is. And like sometimes you use the spanner, sometimes you use the hammer. But I'll give you another example which I think is just as relevant to humans than businesses. And that is a question to ask yourself. Like I'm imagining now a 20 year old student in college. And a question to ask yourself is how much muscle am I building versus agility? And so what I mean by that is not necessarily when you go to the gym, even though that would be a good analogy, it's am I studying my. In spending my entire time picking one subject and digging deeper and deeper and deeper into this one subject where I've built a lot of muscle there. So imagine the guy who goes to the gym and just bench presses every day, just bench pressing, versus the one who's going and studying a little bit of this and expanding their horizons on something else, etc. You're building some agility. Now obviously the competition ends up being a bench pressing competition, the first person I described is likely to be better. But in the world that I think we're entering today, and we're seeing this by the way, with it, with AI more and more, the world we're entering today is one where predicting where the future will be, even five year outs, never mind like 15 or 20, is so hard. So if I tell those two people in the gym, like get ready for a competition and one of them just does a little bit of everything and one just does the bench pressing and some there come up to them and say, by your way, the competition is hurdle jumping, like nobody expected that, I can tell you who's going to win. It's the second one, it's the agile one.

A (29:08)

Right.

C (29:09)

And we're seeing this in businesses again and again, we're seeing this in algorithms. A good way to see this in Algorithm is like anybody listening to this can literally just like Google or GPT, like how many layers GPT1 had and how many nodes or variables and how many variables and nodes GPT2 had? 3, 4, 5. And just look at those numbers and typically layers correlates with muscle and nodes or parameters correlates with agility. And you always see both numbers going up because these models are just getting enormous.

A (29:39)

Right.

C (29:39)

But what you'll see is they have learned that nodes agility matters dramatically more than muscle, that is legs.

A (29:50)

Right, Right.

C (29:50)

And so I think it's not just in the algorithm world, but like for each of us. Prioritizing agility over muscle is just a question we should be asking ourselves. And the agile startups, and both of us will know it like from, from the world we're in, are nearly always outperforming the muscular ones. When Covid hits, it's the agile ones who are going to do well hit the agile ones are going to win.

A (30:13)

Yeah. The other approach only works if, if the TAM for the bench press startup is so massive that as long as you do well at that, it doesn't really matter. But that's a very narrow window of companies.

C (30:28)

You know at the age of 20 that you want to be a dentist and you are a thousand percent sure you don't want to do anything else.

A (30:35)

And you even know, Judah, what 20 year old does not want to be a dentist? Come on.

C (30:38)

There we go. There we go and do that. I'm not going to tell you not to do that. But, but yes. Another question I would ask myself and my business is have I got the right ratio between, call it Capex and Capex and opex or agility and muscle or whichever words you want to use.

A (30:55)

Yeah, yeah. So it's interesting because I, we've been talking about this on the podcast recently, which is kind of reform. We've just been exploring the notion of reforming higher education in the new world that we're living in. I mean, I think even before I was starting to become clear that at least the US approach was very broken in that, you know, we're telling people they need a four year liberal arts degree. There's almost $2 trillion in collective student debt in this country. It doesn't seem to translate into career opportunities. Universities are being driven by rankings that prioritize things that don't really necessarily help students. And at least one of the conclusions we've reached is at least today the thing that matters most is teaching and learning, critical thinking. And that translates as I would Say directly into your point around agility. And it is revamping both the way we teach existing classes and creating new classes. And even I went to law school. I didn't practice law. I will frequently get a young person say to me, should I go to law school? And I know that the question I ask in return is, well, do you want to be a lawyer? And I know the answer is no, because if the answer is yes, there's no reason to ask me, right? And so they say no. And then historically, I've always said, well, then don't go to law school. And then they get upset and say why you did it. And the reality is they either want to put off reality for three years or they want to make their parents happy or whatever it is. Right. I've started to rethink that. At least if you could go, in the case of US law schools, to a really good one that isn't just teaching you how to pass the bar exam, but teaching you critical things, thinking. Because in a world of AI and a world where we just don't really know what the job market is going to be in the next five or 10 years, perhaps that skill that teaches you, the mental agility is really the most important thing you could learn. How do you see that?

C (32:50)

So I think a variety. I think there's a lot of skills like just being able to not give up and sticking to a task and being able to power through. And I'm not even just talking to entrepreneurs. I think just to anybody who wants to succeed. Being able to fail and get back on your feet is another example of a skill which, whether you teach it as a formal part of education or you somehow are able to give this type of skill set to individuals, is immensely valuable. I'll point out another one which is, I think, particularly interesting in the local maximum context, which is, are we optimizing for the individual or for the community? So a great story for this is I. I remember in the military we had a tryout for like the elite military sort of units. And what the, what the tryout was is that they gave you like, I don't know, like an hour, an hour and a half. And imagine 10 guys in a circle in the middle of a desert where there's lots of sand. And you had these bags which you had to fill up with roughly 10 kilo of sand. And then each individual had to decide are they putting the sandbag near their own pile or in the center of the P of the group. And they told us that every bag that's near your own pile will get you two points at the end of the exercise. And every bag in the center will give everybody in the group one point. And they want to see who will end with the most points. And there's maybe 100 people doing this in 10 groups of 10. And what you see is a lot of individuals who start playing as part of the team because you want everybody else to put it in center. But slowly and especially towards the end, they start putting more in their own pile. But what the testers are looking for is do you think about your self interest? Are you optimizing for your individual score or are you thinking is your personality is, are your values a global maximum equilibrium for everybody?

A (34:42)

Yep.

C (34:42)

And you'll see completely different schools, these certain groups that will end up with scores like 520 because everybody has just worked really hard and put the, the sandbags in the middle. And then you'll have other groups that can't get above 100 because everybody's optimized for themselves. So I think one question that we have to ask ourselves when we're talking about education, it's not just higher education, is how do we make individuals perceive not just themselves as individuals, but part of a broader community?

A (35:11)

Yeah.

C (35:11)

Because by, by definition, like you talked about game theory, this will bring us as a humanity to, to a higher.

A (35:18)

Not, not just that I would argue, and I have argued that in a world of greater exist than ever. Right. So there's nuclear proliferation, there is the I think greater and greater ability either for someone to assemble a non state actor to assemble and release a bioweapon, or the higher risk of bioweapons leaking from a lab. I mean, Covid was terrible, but it was still quite frankly not nearly as bad as it could have been of what could have leaked from a lab. Whatever risk might come with AI plus climate that if we don't have global cooperation, we're not going to survive it. I mean, you could argue that the greatest accomplishment of humanity is that for 80 years now we've had nuclear weapons and we haven't used them. And the reason we haven't used them in part is because we're all at least around that one question bought into a larger hole of none of us want to see everyone die. And because countries have been able to work together on this issue, and if the zero sum mentality that I think has absolutely taken prominence in the US and quite frankly you're seeing a lot of it in Israel too, is allowed to continue and succeed, I don't think it's not just a question of, of humanity achieving a higher level. I think it's humanity existing in the first place.

C (36:39)

You're 100% right. The most classic case of this is when they talk about like the prisoner's dilemma.

A (36:44)

Right.

C (36:44)

And you can. Yes, you're 100% right.

A (36:48)

Yeah. So next question then. So you talked about resilience and you talked about the collective good over individual benefit. So in Israel, because everyone serves in the military, both of those things are taught and learned not through the education system, but ultimately before anyone ever hits the workforce, they've already developed those skills. And one of the reasons I like investing in Israeli startups is because every employee there has had that experience and learn those skills. So two questions. One, in the US we can't do that specifically because we don't need tens of millions of people in the military and we couldn't afford it. And in Israel, the rise of the religious right and their leveraging of their political power to get someone like Netanyahu to say let's exempt the Haredi from military service, I think is the death of Israel. Because what makes Israel a successful country, a resilient country and a happy country, and it does extremely well in the world happiness report support scores, is because of that sense of developed resilience and shared collective buy into the greater good. So given that Israel seems to be having kind of a zero sum rise, that is basically saying someone like Bibi, who just prioritizes himself over the good of the country itself, despite what he might say about security, and Trump certainly prioritizing his own good over and actually pitting Americans against each other. What do you do?

C (38:15)

Okay, so on the first question, I think there is no single solution for every society. I don't know the US intricacies as well as I know the British one where I have a lot of friends, and the Israeli one. But I'll give an example, which is a lot of people ask me, do you always pick the individuals, the startup CEOs and founders who come from the best tech units? And I say no. Above everything else, you're looking for a certain character and sometimes the guy or girl who at the age of 10 or 12 wasn't necessarily learning Python but was captain of the soccer team. And the individual I'm looking for is the one who was captain of the soccer team and when they were 30 down, was the one who walked back, picked the ball up from the goal, walked back to the center field, put the ball back in the middle, told his 10 or 12 year old friends, come on, let's do it and kick it off and start it again. Like the reality is, and we've been investing in startups, you and I, for a long time, you more than I have. But like there isn't a startup that just has an easy time.

A (39:21)

And if there is, I've never met it. If it exists. And by the way, as you want a maxim, if you never struggle, you didn't take enough risk and you weren't ambitious enough and you didn't achieve. Yeah, yeah.

C (39:33)

Like true leadership is the one I described. And whether it comes from sports or whether it comes from like the robotics championship or whatever it is, like there is certain underlying skills that you just want your society, your leaders, your business founders or whatever it is to have. And for me, one of them is what I've just described and it's that resilience. And with regards to your second question, I know that is a question that a lot of people asking specifically about Israel society over the last few years. I think it's fair to say this was pre October 7th, especially during the judiciary reform. I'd say on a personal note, one of the positives that I have seen, especially since October 7, is I would say the next generation. It's mostly my peers, or roughly like the age of, call them 20 to 40 year olds, I take it, maybe 25 to 40 year olds who are out of the military but came back to do hundreds of days of reserves. And I'd say, like you just don't feel it there. Like obviously there might be disagreements, but when you are in the tank together or you're in sort of up at 3 o' clock in the morning in intelligence or wherever it is, I think there is a tremendous wave. I'm only talking about Israel because that's the community. I know.

A (40:51)

Yeah.

C (40:51)

That the country. But I think we have a tremendous future wave which is waiting, which it's not easy to find your way into politics or leadership positions. There's a big sort of hurdle that needs to be overcome. But if a couple of years ago people were saying this is the Tik Tok generation, the Instagram generation, do they feel the same thing as like the founding fathers of Israel? I think if you just look at the number of reserve days that a lot of these people have done, or the commitment that their wives, children, it has been unprecedented. And yeah, I am very optimistic from that perspective.

A (41:33)

That is great to hear exactly. To say I have been feeling and maybe it's because of the Way things are going in the US right now, and just the lack of buy in, it feels like from all sides into a greater good. Incredibly pessimistic. So that is encouraging to hear.

C (41:51)

Well, yeah, for a venture capitalist, if you're pessimistic, that must.

A (41:54)

I'm not. It's funny, I'm not pessimistic about the development of technology. I'm not pessimistic about the commercialization of technology. I'm not pessimistic about my ability to find and invest in good founders and help them succeed. But I am pessimistic that we have, and this is for a separate conversation, but. And I'm trying to write something on this in the US and this sort of applies to Israel too, in a way. But we have a society that is multicultural and diverse, which I think is very, very good, because you need that diversity to really, you know, create an innovative society. We have a country now where pretty much everyone in every way has equal rights under the law. I know some people in the trans community might argue otherwise, but it's pretty niche at this point. But we don't have a buy in into the collective good that probably did exist in the US let's say in the 1950s, but in an era where a lot of people did not have equal rights. And one thing I'm trying to figure out is, is there an example of. And can you have a society where people are equal under the law? There is a diverse group of people in a country and yet there's, there's, you know, collective buy in, into the common good. And so far I have not been able to find any examples of that. And I certainly don't see that happening here in the US And I think that's the cause of my pessimism.

C (43:16)

Yeah, well, I mean, on the personal,

B (43:18)

I hope obviously that you're wrong, as I'm sure you do as well. Me too.

A (43:22)

Yeah. Well, I'm trying to figure out if, if I'm right, what are the answers and what can we do about it. And you know, there are things that I'm working on that I think will help. But I think that if we, you know, as, you know, the mobile voting stuff that I do, if we can move power away from the extremes and towards the middle, there will be more buy in to the collective good because politicians will have political incentives to represent the mainstream instead of the extremes. But anyway, this all argues for another podcast. So, you know, whenever you're free, we'll do it again. We didn't even get to like the whole second topic of just what's going on in Israel today. If you liked this episode, you will love Judah's book. It is called how to Move up when the Only Way Is Down. Judah, thank you so much for joining us. This was awesome.

B (44:04)

Thank you so much.

A (44:09)

Firewall is recorded at my bookstore, PNT netware, located at 180 Orchard street on the lower east side of Manhattan. We'd love to hear from you with questions, feedbacks, or idea for a guest. Just email me at First Bradley at Firewall Media or find me on LinkedIn. And to keep up with what's on my mind and my latest writing, please follow my new substack at bradleytust. Substack. Com. Thanks again for listening.