Resource Abundance Will Transform Macro... And Human Life

24:51I was at a conference last month and we were sitting around a table and we got into a conversation about artificial intelligence. And you can usually see the skeptics and the people that immediately start focusing on the security issues, the hallucinations issue. And like the going, I'd say, number one thing that gets brought up as an example is this case, this lawyer who used ChatGPT for his defense and came up with cases that didn't exist. What you're talking about is actually, I mean, and again, it's forcing students to use it and then to figure out where it's not good, which I guess in the end should help AI get better the more you're using it and kind of going through it and going through the models.

25:34but also not staying awake. Make sure you can get this point across, because I think there's still a lot of people that are focused on the negative side of it. And I think the security and the hallucination side is the focal point. Well, you have to remember that every single generative AI system that is generally available right now is basically trained by pointing at the internet. Well, if I want to have someone write a term, if I want to have somebody write a term paper, I'm not going to tell them to go to Twitter and ask for a treatise on the difference between socialism and communism. You're going to get garbage, right?

26:09And the problem with having a fundamentally internet-oriented training environment, which is what most of the current generation LLMs that are available use, is you end up training it on some really mediocre stuff. I'm waiting for somebody to say, here's my LLM that was trained on Roger's Thesaurus and the Oxford English Dictionary and insert five books on rhetoric that I am not a good enough writer to know about, right? And to create something that just speaks English and that knows how to go search for stuff. It's important to remember university is really about learning how to learn and then creating the pathways in the brain for critical thinking so that when you are presented with a problem, you can then analyze the problem thoughtfully and then analyze it and then attempt to solve it or at least create a framework that allows you to guide other people to solve it, right?

26:57A university is not just a piece of paper, right? And I think a lot of students right now and a lot of people kind of outside of academia have kind of lost that thread. And so you have to understand that you're using a hammer, but it's a flawed hammer. And that you as a computer science student like mine are, are going to be building the better hammers later. And so you have to critically analyze it. And so I'm a big fan of critical thinking in that regard. The other thing to remember is if you are thinking about a general use technology, it's not going to be amazing at any one thing. It's going to be okay at a million things.

27:30And so the best LLMs over the next couple of years are not going to be general use LLMs. They're going to be targeted at specific problem sets. They're going to say, okay, I want to have an LLM that just does clinical genetic analysis for specific cancer types. Great. That's going to be a good LLM because it's going to know how to speak English. It's going to know the right data. It's going to have a highly curated learning environment. And it's going to be, in essence, governed in a way that allows it to give you thoughtful information or say, I don't know. Just like if you're talking to an expert, right?

28:02If you ever talk to an expert about clinical genetics, you're never going to get a yes or a no answer. You're going to get, here's four papers. Here's what I think. Here's what the data is saying, but here's all the stuff we don't know, right? And that's the nature of humanity, right? All right. So let's, since you talked about your students. I think I just did a podcast on the importance of a different skill set for people in a world of AI, meaning I've already seen it where not everyone embraces using it. There seems to be a fear of it. I'm not sure what that is because it will reduce the amount of time you have to spend on mundane tasks and give you more time to think and connect dots.

28:43Are there skills in the people you're looking to hire for your own business right now that maybe are different than before AI? I know you're still in the tech side, so these people are going to have some kind of expertise. But for people outside of technology that are looking to improve the way their employees, their efficiency and productivity of their employees, what types of skills, soft skills do you think are more important now than before the release of ChatGPT? Oh, that's an interesting question. I wish you'd put it in the preview. I would have thought. I got to go off-roading every now and then.

29:18So here, I'll answer this in two parts. You'll love this. So when a student asks me that, I'll say there are really three things I think are going to make somebody entering the workforce today very effective. And when you're leaving university, I hope that you can, to one degree or another, answer, kind of show how you're qualified in the following three areas. The first is your communication skills, written, verbal, presentation, the ability to stand in front of a crowd, even if it's five people and say something. The second is the technical mastery of being able to sit down with a clean laptop and go from zero to creating something, whether it's a piece of art or a video game or some code or whatever.

29:57We are no longer, that is no longer an optional skill set for someone entering the workforce. And then the third is some degree of domain knowledge. There's an area you care about. You care about health. You care about finance. You care about entertainment. You care about something, right? If you can combine those three things, you're going to be in a great spot. I think for people already in the workforce, where they probably already have good communication skills because you will have picked it up over 5, 10, 15, 20 years, and you will already know a domain incredibly well. You'll know something.

30:25You'll know, you know, I don't know, quantitative analysis of public equities or whatever that is, right? The thing that you might be missing is that hand-on-keyboard activity. And, Jordy, you are the perfect test case of someone that I would hire because I can hand you a laptop and say, install Python and write Hello World and then install this open-source thing and do this, or install OpenBB and tell me why it's not quite as good as a Bloomberg terminal. Which, by the way, for those who don't know what OpenBB is, it's an open-source project. You should absolutely look at it. We've name dropped AutoGPT so far today, which is also open source, and we've name dropped OpenBB.

31:05Highly recommend you look at both of those. They're absolutely fantastic tools. But you have the ability to do that. And so when I'm hiring right now, I am hiring for communication skills, good culture, and I'm hiring for domain experience. I do not need them to be, for the most part, PhDs in computer science who are just guys who sit in a room and code. That is what this enabling tooling that I talked about at the beginning allows you to do. We took a person who joined our organization not too long ago who fundamentally had no technical skills, none whatsoever, and went from zero to auto GPT on their laptop, having to build stuff, having to do them, basically duplicated my content creation pipeline, which Jordi, you know quite well, in the exact same way.

31:53And it took him two days, basically. so you know what's funny about your answer so since you didn't hear the podcast that and i forget the exact title of the podcast and by the way i keep referencing this podcast that sultan has done with me my podcast podcast is in search of green marbles and we just did one on these soft skills so you hit communication you hit creativity for your domain experience i called it competitiveness which is literally the ability to sit down and realize that if you do this, you will get better because that's what happens with AI. It's literally like having the smartest person I've ever met next to me who makes mistakes, but is incredibly efficient.

32:31And so it collaborates with me, which is the other one, collaboration skills and understanding how you work with another person. Because really, ChatGPT to me is another person. It helps me think. It connects dots. I ask it for analogies. I ask it for titles. That's the way I like to think about things. So when I come up with something, I'll just say, hey, give me five analogies that would work really well with these two disparate things. And then critical thinking. And those are the Cs. And the whole argument was that the softer skills, you're breaking apart IQ, where the commoditized portion of IQ is the memorization that we got in school.

33:05And now these CQ, this whole thing of competitiveness, communication, critical thinking, creativity, all of the things that you mentioned in there are important. So I completely agree. It's funny. You went off-roading, but you gave me the answers that I think are definitely showing up. I wanted to take it a different direction here in terms of breakthroughs. And one of the most fascinating podcasts I listened to last year, way before ChatGPT came out, was with, I believe his name is Dave Johnson, who's the head of artificial intelligence for Moderna. And I've had this belief that when we're connecting back to economics and people are arguing about inflation and this and that, the inflation to me was caused by the breakthrough of artificial intelligence to come with an answer to the pandemic far faster than the government ever thought possible.

34:07So when you print a certain amount of money, you're doing it because you believe that the pandemic is going to last for a long time. And I think most forecasts on when we'd have a vaccine were in the three to four years, maybe at the fastest we could get to two, but that would be a record. And the blueprint for the vaccine from everything that you've heard from Moderna now was actually there a weekend later after they got the sequenced genome. And so that was a breakthrough in AI that clearly has gone on. Since then, we've had alpha fold with protein folding. We've had a fusion breakthrough where AI has been involved.

34:41We have all this excitement and disbelief over LK99. And I asked you about it, and you obviously know a lot about these, all of these, but it seems that these breakthroughs are happening more and more because of AI, and it seems like it's speeding up. And these breakthroughs, obviously, like the vaccine example, will have impacts on our modern day, kind of the way we make decisions long term. Can you talk about how many more break, you know, are we in an age of massive breakthroughs and the impact that this is going to have on long term assumptions? I think it's going to be difficult to say that we're not about to see some degree of exponential discovery, because between the enabling tools, the amount of data that's been digitized already, and how so much of the infrastructure is now digitally native, that once you start gluing these things together, it's very easy to point an AI at it, right?

35:33You know, there are entire segments, entire market segments where, you know, you might need 500 databases to analyze something, and that might take years to glue together. So I commercialized some clinical genetics technology like 10 years ago, and it was like 400 databases. I'm going to team of like 20 developers, and it just took like three years, basically, to go from zero to something useful that you could use in a clinical setting, maybe a little less. and you could replicate that entire thing in about a day now with the enabling technology and so now it is no longer just automating a process or digitizing a process it's now the point where you can say like with alpha fold point the technology at this data give it a problem set orchestrated appropriately and it will just spit out a thousand different answers and then you can create a different system that will analyze it those different answers and prioritize what might make the most sense for you or not.

36:28And this ability for us to automate the innovation, to automate the research, to automate the discovery of prioritized things that go from the science bucket to the engineering bucket, which has traditionally been, especially in places like biotech, one of the biggest failure points and risk factors and costs, by the way, you're now at a point where that all of a sudden, that economics changes. And so in as much as I would compare building an AI company to building a biotech, there are places where you can build biotechs from zero to one and get to a point where you're in a phase two trial in a very small number of years instead of five, six, seven years.

37:09There have been a couple of companies that will take orphan drugs or something like that and then try out some alternative use and then spit them through an IPO, make a bunch of money, sell them off to a Moderna or Pfizer or someone like that and move on. This is different than that. I'm sure there are people listening who know that kind of SPAC-based biotech phase two process. This is not that. This is, okay, I have a database of every rare genetic disorder, and I have a database of every drug we've won through phase two for drug company X. You can hand two people a reasonable laptop and an internet connection, and they can go through and look for all the things that all the other guys missed over the last 20 years.

37:47and you can rerun that every day. And so every morning you could wake up to an email saying, oh, by the way, this snip, there's this thing over here that you didn't notice. And oh, there's this research paper over here. Maybe you should read it. And all of a sudden the time for value for an individual, for those real honest to God experts, it's no time whatsoever. And with the automation being built in, you now are at a point where if you're in an environment like that and you aren't doing these kinds of things, somebody is going to come along that does do that. And they're either going to charge you a fortune for it, or they're just going to eat your lunch eventually.

38:20Yeah, what's really interesting about that, too, is, I mean, the largest industry in the United States right now is health care. And we obviously have a massive debt situation that involves a lot the future liabilities that we have for people getting older and being less healthy. So what you're describing and, you know, you were the one who who told me to look into protein folding and alpha fold. And as I do with, well, everything you kind of give me some insight in, but also my general thing, I love learning about what the future is going to look like. The thing that's different today from 10 years ago, when I used to look at things 10 years ago, these were just kind of guesses on what would happen and when we'd have autonomous driving.

39:03And if you go back and read the things from 2013, maybe 30 % of it is happening at the level. The difference with AI, we're getting to the problem. We're making daily breakthroughs in health. And also, where I want to kind of have you expand upon this, you and I did a podcast, and the energy side is so important. So if on the one side, we're impacting longevity and we're impacting health, which is a major thing. At the same time, we've had breakthroughs in fusion since, and again, it's not because of ChatGPT, but it's been over the last six months we've had some breakthroughs. And then you have this LK99, and for those who didn't see it or they don't know what it is, we're talking about something very important, and you can describe how important it would be.

39:53But the reason I want to spend time on it is not because it's guaranteed to happen. There's a lot of skepticism on this, But you and I had a conversation. And what I said was, regardless of whether it's true or not at this point, the beauty of human intelligence is if we're being creative and testing things and we're getting to the point where we've made it 93 % of the way there, AI should make the last 7 % much easier than it's been in the past. Because if you get that close, can you talk about that, but also LK99 and what it would mean if we did solve that problem? Well, we'll tie it back to what we were just talking about.

40:29Any discovery pipeline, any time you're going from a question to a solution, any time you go down that pipeline journey, there are going to be two places where AI is going to have the biggest fundamental impact. And it's at the edges. It's at the very beginning. It's the very end, right? So let's just say LK99, which is for those who, you know, you should Google it or whatever, but it's basically a very, very interesting magnet that fundamentally alters how we would potentially create energy. And as we think about fusion, which is really the next big step in energy for us, you know, and there are a bunch of really great new companies that are doing awesome work there.

41:07You know, it gives you the ability to potentially radically alter the cost, the amount of energy, the time to development, et cetera, et cetera. So I'm not going to go down the physics side of that part of it. There's a lot of skepticism over whether this invention is real or not. And to your point, it doesn't matter. if it's 96 % of the way there, if it's 80 % of the way there, it doesn't matter. It's shown a pathway. And that's the important thing. And AI may or may not get us there, but guess what? I think AI will absolutely participate in that. And so at the very beginning and the very end of the discovery process, AI is now at a point where it is taking away millions and millions of people hours per day work that then will optimize the people who have been doing that so they can get us through that middle sector, that kind of valley of death where so much fails.

42:00And then AI can get back involved and can systematize it, automate it, and run it down the path. And whether we're talking about biotech or energy or material science, which is another really interesting area, or how to how to create more reliable production-ready quantum computers, which is, I can't believe we've made it this far in the conversation without saying quantum. That's rare for us, frankly. You know, all of these next steps, they're companies in my mind that are going to be the ones that really create that real scaled value. You know, they create the quantum computer chip that goes in your phone in a couple of decades.

42:41They're the ones that create that silver bullet cancer drug that so many people get excited about, or frankly, they should be looking more at like childhood cancer. I mean, that's an area that has been ignored for decades that really needs some energy attached to it. But it would allow us to spend more time on diseases that have far more applicability, that aren't just based on how much someone can pay, which is one of the reasons why we have this massive debt structure in the healthcare system in this country, and move it to actually solving some of these bigger issues. And it doesn't matter, but the ability for us to automate and then use some degree of AI to solve that problem is inexorable.

43:21And frankly, if I am looking at a company that's operating one of these spaces and I don't hear them using AI to make themselves more efficient and don't, or if they're a new company and they're not kind of building it into their operations from day one, then I'm going to have some pause over what I think their long-term model, their long-term success might look like. Well, you've hit on something that breaks directly into kind of where I want to finish this little last part on in connection with the economy and the disruptive breakthroughs and what it can mean. And this year for the markets, and we'll kind of segue a little bit into the markets here, the focus for about the last year and a half has been on recessions.

44:01Probably started around March to April of last year, where people became obsessed with the concept of recessions. And what you just described in terms of LK99, the longevity side, and the breakthroughs we're seeing in healthcare, we went from the early 1990s. You were at Netscape. It came out, I believe, went public in 94. We had a recession in the 1991 period, and then the next recession we had was in the 2000 to 2003 period, really, 1-2. So we went 10 years without it. Then we had, obviously, the great financial crisis in 08-09. And then it took a pandemic to get the next one. So if you kind of go through, we've really had a spacing out of about 10 years.

44:47And even though people are obsessed with a recession call right now, I don't see it happening and don't think it will. And one of the reasons is because I think AI, the most important impact, is not about earnings. It is about what efficiency and productivity will look like in the future. And even though that will mean almost assuredly some type of pressure on employees, the big thing is I have spent the last 14 years on this concept of abundance and then going in reverse and saying, well, what is a recession or a business cycle? And it's really about scarcity at the end of the day. And AI is getting us closer and closer to abundance.

45:27So do you think there's a chance, and again, this is kind of a bold statement, that the way recessions look like and how they act will be disrupted by the speed of AI, meaning five years from now, will we be talking about efficiencies and productivity, which make it very difficult to think of scarcity the same way? Oh, that's such a good question, Jordy. When you and I last talked about this, I hypothesized the notion of the GDP of a nation not necessarily being that valuable of a metric in the not-too-distant future. And it ties into this. And I want to start my answer by talking about that, because that implies that there's a single economy in the United States or in any one country.

46:09I am not convinced that 20 years from now, that will be the most effective. I don't think it's the most effective way now, but I think it'll be more in the mainstream of thought 10, 20 years from now. Because we are going to have so many little economies that interact with each other in very automated and I'll say buffered ways, that there are still going to be certain places where housing is a great example, energy is another example, where the emotional impact of that subset of the overall market could have bigger ripples. but that these individual markets are organized, kind of self-organized because of things like automation and artificial intelligence to create buffers.

46:47And so the ability for recessions to exist is harder because you can't just knock out one subset of a market and then all of a sudden everything else tumbles over. When we move away from human work time being the fundamental metric under which economic output is measured, which is still where we are today, I wonder if we're finally at that turning point. And this is like the kid who cried wolf. About every five to ten years, some guy writes a paper or a book, whether it's about the fourth industrial revolution or whatever, and they try to argue that economics point. I do wonder if we're kind of getting to a point where at least at the base layer, there won't be that ability for that to happen very easily, especially because that's kind of the positive case side of it.

47:35The negative case is, if that does happen, it means there's a fundamental structural flaw in the overall market system, right? And I think we've proven after a couple hundred years that's not the case. And I will go back to Fusion as kind of the last part of the example. You know, the most interesting fusion company out there, and I won't name drop them here, is really going to build two devices, right? One is a device that sits at the base layer of energy demand within a municipal infrastructure. So it's the bottom 25%. It's the 20 % that runs the water treatment plant. It's the 20 % that runs the hospitals.

48:08It's that very steady state, keeps the streetlights going. All the variability can be in the organic batteries that we currently use, i.e. fossil fuels. The second device they make is actually super interesting because it generates rare earth, which has obviously been in the news a bit recently with certain activities from the Chinese Communist Party. And so we are at this moment where I think you will start seeing AI do two things. It will create a base layer of economic buffer that insulates us from massive economic disruptions in certain areas, and that will continue to grow over the next few decades.

48:45And the second is it creates the black swans. It creates the company that is going to have 20 employees instead of 2 ,000. And those 20 employees as an average unit of work are orders of magnitude more efficient and more value generating than the ones in the other companies. And so, you know, there's a, I mean, I'll give you a great example. You know, there is a startup hedge fund that we're working with. And, you know, if you were to take the size of what that hedge fund should be in terms of human capital, et cetera, et cetera, it probably should be, you know, somewhere between 75 and 100 employees.

49:19You know, it's kind of if you look at their peer group, that's kind of where they should be. I'm not convinced they're going to have 15. Right. Think about what that means in terms of just, you know, the money that flows out the door on a weekly or monthly basis. Right. I mean, you can speak to that with a massive amount of credibility. Right. Just imagine, you know, what that ends up looking like to the space. And so the ability to build disruptive, differentially scaled businesses with differential human capital value creation was very much limited to specific categories of places where technology had a direct application.

49:52I think we're moving past that. I think it is now much broader across the overall market. We're actually getting near the end of time. We have a couple questions that have come in. So I'll break it here just because we only have a couple things left to talk about. from the list. One of them was directed, the first one was towards me, and it's a simple one. But the question was, do I think there's a possibility that long-duration stocks are performing in the face of yields going higher, the selling in treasuries, due to AI reducing future CapEx as opposed to the liquidity element? And my own thought on this, just real quickly, and then the other questions are really for you.

50:36Number one, I think the web 1.0 and 2.0 playbook is out, meaning I think from talking to people that everyone assumes with a new technology that it's going to be about buying tech stocks. And I think the mega cap names have gone up significantly, partly because of the AI enthusiasm, which is not only seen in what the NDX is doing, but even the startups, there's been some very large raises at much higher multiples. So there's a lot of enthusiasm in the private market, at least isolated to AI at this point. So I think bringing out the web 1.0 playbook that the winners will be the same. And we don't have to get into this, but I know you've got views on buying companies was part of web 2.0, just buy up everything that was competitor.

51:18And AI is not going to allow for that because these are more customized and they're really going to be completely different. So I do think smaller businesses will eventually win out. But for the time being, I think it's a combination of just cheap tech stocks, people diving in with the new technology. And I think eventually we'll get back into where higher rates will be worse for longer duration stocks. Another question which just came in for you, can you cross-fertilize Cesar Hidalgo's law of relatedness to this discussion in changing developing world opportunities? I hope you know how to answer that.

51:54I have no idea what's going to go. We do not have 20 minutes for me to talk about that. But I mean, the short answer is yes, but I think is how I would answer that right now. The interconnectedness of the world, and I actually would go back to James Burke to out-nerd the person who asked that question of connections and the day the universe changed, Vane. He actually nailed this, which is that fundamentally small improvements in technology generations later are so institutionalized that they fundamentally alter the culture. And I think that's what we're seeing now. I think the difference that matters now is that the timescale of those impacts is now measured in days, weeks, not years and decades.

52:37So that's my short answer. Well, in the last, and I'll take it, I know you have to jump right around noon. We'll finish it up in seven minutes. But I do want to talk a little bit about, not specifically the project we're working on, but some of the things with Inside the Hedge Fund World for people out there that might be thinking about this. Do you believe that what we're working on is an indication that the world of hedge funds, where there's been a very, very difficult barrier to entry for using artificial intelligence in your business, where it took a lot of money for the investment, and it took a lot of people, that this gives a chance for smaller hedge funds for much smaller costs to be able to compete with the larger ones.

53:22Absolutely, yes. So if you look at the mega cap hedge funds that have had a bit of a monopoly on using artificial intelligence because of just hundreds of millions of dollars, hundreds of people required to do so much of this work, it is now not that, not remotely. And the ability for, and I'll speak more broadly, not just on the hedge fund side, but what three years ago would take 100 or$200 million and a couple hundred people and a couple of years, you can now do with cut a zero off dollars, cut a zero off people, and cut 50%, if not more, off the time to delivery. And so I would say that hedge funds are going to probably be on the front end of this because there tend to be early adopters in a lot of ways of a lot of these advanced technologies.

54:06I think there's a lot of opportunity for mid-tier hedge funds to radically, radically compete in a way they haven't been able to compete before. Yeah. And I can just say to everyone listening, the reason that I listened to Sultan and I bought my own laptop and focused on downloading Python, taking Python 101, going through the process of of uploading auto GPT, understanding the power of it, using it, using code interpreter to the point where, you know, in no exaggeration, by the time the week is out, I've averaged generally more than four hours a day using artificial intelligence with part of everything I do, including data analysis and Bloomberg and everything that if you use it, there is a tremendous amount of benefits that come.

54:55And I think that led that belief led me to saying, hey, I think we got to work together because there's some things that I always wanted to build. And I think the market's been disrupted. And that's the final thing that I kind of wanted to speak about, which is, you know, with inside the hedge fund world, one of the most important things is risk management. And one of the most important things to making money is to make sure that you come up with a risk framework that can help you be in control of your portfolio. And behind that is a need for the stability of the relationship between correlations and volatility, because every risk model to some degree is assuming those.

55:35And it is my belief that because of AI, over the last, I'll say four years for sure, but let's say three in general, we've kind of left this world, and I don't think people have caught up on it, where risk models and these ubiquitous risk models that everyone uses now, you still have human inputs of figuring out what data you're going to use to say you're hedged. And nothing is worse in managing money than thinking you're hedged and you not being hedged. So the project we're working on is under that assumption. Can you just talk a little bit about your beliefs in that and how much you guys have worked on it already?

56:10Sure. I mean, and here's a place where my experience kind of leads me to have a slightly different view than many others, because I've actually run both financial services and biotechnology programs through federal regulatory bodies and then was also a federal regulator. And so understanding how to think about hybrid risk models and where risk models fail in a variety of different areas. I've had firsthand visibility to that on both sides of the table. And let me tell you, I think most risk models that are used, I think both in biotechnology and in financial services, which I have direct experience with, are frankly inaccurate in most cases.

56:47They aren't actually fully understanding where the risk is. There's far too much human interference with it. And there are a lot of places, and I cite some of the banking issues we've had in this nation over the last year and a half, where human assumption or it's always been that way or it hasn't, you know, whatever, whatever, gets you to a point where there's complacency amongst these environments and there's too much change in the market right now, whether it's real-time payments on the banking side or changes in Fed interest rates happening at a different rate than what the models can absorb, or simply understanding the impacts of what your hedging infrastructure looks like relative to what you actually think it is versus what it actually is.

57:27And so having the ability to build risk models where it's not just a risk model or five risk models, maybe it's 150 risk models with thousands of different factors that are all operating in a dynamic environment so that what it's actually providing is not a score saying, oh, we're fine. It's saying here are a bunch of different ways of looking at this. Here's some traditional methodologies. Here's some more advanced methodologies using AI and understanding how those live with each other. And then the discussion becomes, it's an evaluation of different risk models in parallel that allows someone like Jordi to be able to say, well, you know, if the market's kind of doing things like this, I'm comfortable with risk model A, but you know what, this is a different environment we're in.

58:09I'm looking at a very specific market that's acting in a very specific way. We need to think about it in a different way and understand what it means relative not just to the individual portfolio, but the overall organization. And the ability to do that in real time, the ability for that to just be sitting on Geordi's screen, I don't think too many organizations do it. I don't care what market you're in. And I know that if that existed in the banking infrastructure, I think a couple of banks that don't exist anymore would probably still exist. I think a number of drugs that have been pulled off the market would never have gotten there to begin with.

58:42And certainly the data shows in both cases that everything needed to make decisions in both those cases was publicly available already. And so it just they weren't thinking about risk the right way. All right. So we have 90 seconds. So I'm going to make one little statement here on what you just said and just give people a little bit more on this risk model concept. Risk models not only have to use historical relationships, but then the output that they're trying to measure is things like growth and value and quality, momentum and a whole bunch of different factors. The number one factor to generate performance this year has been AI.

59:15You can go back to pandemic. You can go back to reopening. In a world of AI, I think assuming that historical factors mean anything in the marketplace where change is happening so rapidly, I just think you're going to have instability in both the risk models and those relationships between correlations involved, but also in factors. So let's end it with one final question for you. If people want to read more about what you're talking about on a daily basis and see more of Soltan, where can I find you? Just spell my name right and put it in Google or go to soltanismyname.com. That's the marketing person's website.

59:50All right, Soltan, we've extended our relationship even further with this conversation. Thanks to Real Vision again. And I'm sure both of us will be seeing you again soon. Fantastic. Thanks for having me. What's up, revolutionaries? Thanks for tuning in to the Real Vision Daily Briefing. For more content like this, head over to realvision.com and get unfiltered access to the very best, brightest, and biggest names in finance.

1:00:36once your account is approved. And the great thing is that in addition to crypto, Plus500 gives you access to a wide range of instruments, S &P 500, NASDAQ, gas, and much more. Explore equity indices, energy, metals, Forex, and beyond. With a simple and intuitive platform, you could trade anytime, anywhere. Experience the fast, accessible futures trading you've been waiting for with Plus500. With over 20 years of experience, Plus500 is your gateway to the markets. Visit us.plus500.com to learn more. Trading in futures involves the risk of loss and is not suitable for everyone. Not all applicants will qualify.

1:01:10Plus 500. It's trading with a plus.