We Study Billionaires - The Investor’s Podcast Network - TIP575: The Future of AI w/ Bob Muglia

Starting point is 00:00:00 You're listening to TIP. On today's show, I'm joined by Bob Muglia. Bob spent 23 years working with Microsoft, and he held a number of executive positions with the company, the last of which being his role as the president of the server and tool business unit prior to resigning in 2011. In 2014, Bob became the CEO of Snowflake, and he helped lead the company from zero to $200 million in revenue before parting ways in 2019. Bob recently released a book called The Datapreneurs, which explores how technology got to where

Starting point is 00:00:32 it is today and where he foresees technology and AI going in the future. This episode touches on a lot of interesting topics, such as Bob's lessons from working with Bill Gates and Steve Balmer, what differentiates Snowflake from their competitors in the data warehouse space, whether Bob was surprised that Berkshire Hathaway invested into Snowflake's IPO, what the art of data innovation is, when Bob foresees artificial. General Intelligence becoming a reality, what industries he sees AI impacting the most, what he looks for when investing in technology companies, what the future of online search may be, and so much more. Without further ado, here is my chat with Bob Muglia. You are listening to The Investors Podcast,

Starting point is 00:01:19 where we study the financial markets and read the books that influence self-made billionaires the most. We keep you informed and prepared for the unexpected. Welcome to the Investors Podcast. I'm your host, Clay Fink. And today, it is such an honor to be joined by Bob Muglia. Bob, welcome to the show. Great to be here, Clay. Well, Bob, we got connected because of the book you recently published called The Databpreneurs, which I'm holding up here for those watching. And as I was looking into your background, I'm thinking, man, I just have to get Bob on the show. You are really one of the builders and the visionaries of all these different technologies that are developing and especially AI. And where all this is heading

Starting point is 00:02:07 with all the companies you're involved with and especially your experience working with Microsoft and Snowflakes. So I think a good place to start would just be to touch on some of your background. So could you please walk us through your journey of what brought you here today? Sure. So I mean, it's interesting because when I was in college, I was sort of studied, I was a computer science degree at University of Michigan. And I sort of was focused on networking and communications because this was the early 1980s and I felt like that would grow tremendously. And I joined my first, you know, when I was at Ann Arbor, was in Ann Arbor at the University of Michigan, my first technical job was at a company called Condor Computer, which is in the late

Starting point is 00:02:44 1970s. It was an actual true relational database. It ran on an incredibly early microprocessor system called a Crememco. It was this big box with large, if you can believe, at eight inch floppy discs that held a tiny amount of data. It had just a tiny amount of memory in the machine, and yet it was a real relational database, and that's where I started building business applications with data. And then when I joined Rome, which is a communications company back in the 1980s, I wound up building business applications there and working with databases, some of the early networking databases that ran on PCs back in the early 1980s. And that led me to Microsoft. My wife and I actually had a desire to move up to the Seattle area. And she was the one who really found

Starting point is 00:03:28 Microsoft. She's a Stanford MBA and a pretty smart cookie. And she was the one that spotted Microsoft in like 1985. And I joined Microsoft in early 88. My wife actually joined, you know, six months later and was there for five years. I spent 23 years at Microsoft. My first job was on SQL server. I was the first technical person on SQL server. We had a product manager and I was the technical program manager helping to work with Sybase who actually built the product down in the Bay Area and helping to deliver it on PCs running OS2 at the time initially was running OS2 back in these crazy days. And then I wound up working at Microsoft doing a series of real sort of startup kind of things inside Microsoft for really throughout the 1990s. Staying on a business for like two years

Starting point is 00:04:14 or so, I helped to build Windows NT and the Windows server business. I ran the early, early server products and actually ran the tools part of Microsoft, brought together Visual Studio that was in 1996 when we first created Visual Studio and I was running the division at the time. I did a whole bunch of things that were really solving problems at Microsoft, new problems, and then spent my last seven years running the server and tools group at Microsoft, which was Windows Server, SQL Server, System Center, and the developer tools, the developer tools products. So I brought that division to about 17 billion in revenue.

Starting point is 00:04:51 9 to 17 billion. It was interesting back then. We never got mentioned. The server business was growing 15% a year, constantly just 15%, 12%, 18%, you know, we would never get mentioned in the earnings report unless something bad was happening in Windows or the office, because they would always talk about those things. But you knew if servers was being talked about that, it meant that they wanted to highlight something back then. I left Microsoft in 2011, spent a couple of years at Juniper, which reintroduced me to the Bay Area, and then took the snowflake job in 2014, May of 2014, and ran the company for five years. Subsequently, since I left Snowflake at 2019, I've been working on boards.

Starting point is 00:05:30 I'm on five boards of small private companies. And I really act as an advisor to CEOs, all the small companies, helping them, helping them build their business model. We've got a lot of brilliant technical people out there. And yet they don't necessarily have the business side of things. And I picked all of that up. I mean, I was fortunate to learn at the hands of people like, Bill Gates and Steve Balmer, who were as brilliant a businessman as I've ever seen. And so I was able

Starting point is 00:05:56 to help a lot of small companies in their CEOs as they start to build their business. So that's what I'm focused on now. I mean, man, what a resume. 23 years with Microsoft ended up being a president of one of their divisions there when you left in 2011 and then CEO of Snowflake for five years, which we'll be talking about a bit here. I was curious if you could share some of your takeaways and experiences and key learnings from working with people like Bill Gates at Microsoft. There's a ton, an amazing amount I learned from Bill and Steve. Bill is technically one of the strongest people I've ever met. He's a brilliant guy. Also very good with people and recruiting technical people. Bill has a great ability to build strong technical relationships with people.

Starting point is 00:06:40 And that I think has really guided him on his career. And I think maybe if I want to say if there's one thing I learned from Bill, it's that, you know, building these relationships with people, the technical people, is so critical. If you look at my career, I've mostly assisted brilliant technical entrepreneurs to build their technology and their product. It's really where the idea the datapreneurs came from, this idea of data entrepreneurs. And even when I was at Microsoft, I had the recognition that I was working with data entrepreneurs while I was at Microsoft. And then, of course, subsequently have done so at places like Snowflake and the things I've done since then. But I learned an enormous amount about how to work with those technical people from Bill,

Starting point is 00:07:19 because I think Bill is about as good at that as anyone was, although he could sometimes be, Bill can be pretty aggressive sometimes and sometimes relatively dismissive of people. I try and be a little bit more respectful than that. Bill's, one of the things Bill is known for saying is that's the dumbest thing I've ever heard. And he used that line so often, it started to lose its effectiveness, because eventually you decide that once you've heard something's the dumbest thing you've ever heard a hundred times. It must not be that dumb. But from Steve, on the other hand, Steve is an incredible businessman, probably one of the most multi-dimensional, smart mathematical thinkers I've ever seen.

Starting point is 00:07:59 Steve could keep in his head a relative spreadsheet of the Microsoft revenues byproduct, by country, and have a really good understanding of that, a stunningly good understanding of that. Like literally to the point where, I mean, this is a notorious thing at Microsoft is there was this thing called the Rev Sum. It's a brilliant, brilliant idea. This idea that there's so much wallet, a share of wallet and you want to understand what you're getting across all of your different markets and things. And what Steve would do is he focused on key drivers like a socket. Like a PC is a socket. You know, it's a, it's a sell. It's something you can sell into. So when a PC is sold, it's something that you could then, you know, Microsoft would earn revenue on on Windows, but there was also opportunity for. office and all sorts of other tag on products. And Steve would build these spreadsheets to have the finance build these spreadsheets that had all of the products and all of the regions and countries. And it would literally be this giant sheet of paper, this 11 by 17 sheet of paper, which Microsoft loved. And there was 2,500 numbers on one piece of paper, I swear to God. And you would look at this thing. And Steve could, you know, Steve would get presented to this in a meeting. And he would

Starting point is 00:09:05 stare at it for about 30 seconds. And he would go, this is all wrong. I know it's wrong because that number is wrong. And he would point to a number in the middle of this page of numbers, which I'm like, I could, to me, it was just all a bunch of numbers. I could barely see the thing. And at that point, the finance guy would start, you know, those are these papers. There'd be a 10-minute period where there'd be a discussion about whether that number was wrong. And honest, I swear, eight out of ten times Steve was right.

Starting point is 00:09:30 And it'd be real the meeting. And maybe it wasn't counterproductive even. But it still created a set of expectations, if that makes sense, that everyone knew when you presented that spreadsheet to Steve that he was going to look at it. And if it was something wrong, he was going to find it. And it really drove the company in a lot of senses. My style is very different than that. Steve could be very aggressive and sometimes disrespectful in meanings. I always tried to be respectful and things like that. But I do think that pushing people to do more than they expect they can do is an important thing. And that's certainly something that I learned from both of them.

Starting point is 00:10:07 Yeah, it's fascinating. I think about what you said about Bill where, you know, you look at many these great companies. And I think a key attribute you look at Steve Jobs and Musk. They're just brutally honest people and they will just achieve what they want to achieve at all costs. And difficult, frankly, they're all difficult. Find one that isn't difficult. All these brilliant founders, Mark Zuckerberg, you think he's easy. Your jobs was difficult. Bombers of pain in the tail. I mean, they're all fairly difficult. But it's part of what makes them what they are, really. In your book, you talked about how data is one of a company's most critical assets. Given that you've been in this, we'll just call it the technology industry

Starting point is 00:10:47 for so long, when did you realize, you know, how critical data was going to be for companies? And how has your opinion on that changed over time? I think it really very early, you know, when I started working at Rolm, even, I was working on collecting data. You know, we were selling PBX systems, business telephone systems. Rolm built the first digital PBX, which was very innovative back in late 1970s and early 1980s. And at that point, even then, we were collecting information about customers that we were having to put into these systems. So I recognized how important it was. When I joined Microsoft, I was focusing on collecting data and working on information with SQL Server. And SQL Server really

Starting point is 00:11:25 did change the industry in that it brought business computing to businesses of all sizes. If you go back to the early 1990s, most small businesses kept their books on pencil and paper. And that all changed largely because of the work we did at Microsoft with the products that we built. that were specifically targeted at those industries, whereas most of the other companies were targeting larger businesses. So, you know, I recognize that data had an ability to impact businesses of all sizes, certainly when I was at Microsoft working on SQL server, but also because very early on Bill started an initiative called Information at Your Fingertips was really about using technology to gather business information and to be able to bring that to people. I mean, ultimately,

Starting point is 00:12:09 it has turned into, the actual implementation of it is turned into the internet and Google for all practical purposes of search. But, you know, that vision that Bill said in 1990, which I do talk about in the book, it was very much a big driver of all of the things that happened. It was a big driver in shaping my focus on the importance of data. I also wanted to touch on a little bit about your role at Snowflake. To my knowledge, they sort of were operating under stealth mode for some period of time while they developed these products and services, and then they took those products to market once they were ready to go. So talk about how you ended up joining Snowflake in 2014 and how your journey with them evolved over time as they sprung out of this stealth mode. When I left Juniper

Starting point is 00:12:57 in late 2013, I decided I wanted to work at a much smaller company than I had been working at. At Juniper, I had an amazing realization, which was that in Microsoft, I was building new business, all the time. And Microsoft was able to do that. And I realized that part of the reason it was able to do that is had this cash cow called Windows in Office. It was throwing off a lot of profitability. I joined Juniper and Juniper was had stagnated shortly after I joined. It wasn't growing. And I realized that it is almost impossible. I was trying to build a software business in a hardware company. And I learned that it was almost impossible to build a new business at a company that wasn't growing because there just simply isn't enough cash to feed both the primary products that are generating the revenue and to build

Starting point is 00:13:38 these new businesses. And I wanted, and I came to, and while many people love fixing broken things, and there's certainly a lot of joy that can come in that. I had seen a lot of that at Juniper. I was fixing a bunch of broken things. And I decided I wanted to build something new, which is what led me to small companies. And I started looking around. I was connected through, I had a contact at Sutter, at Sutter Hill, which was the founding VC of Snowflake. And that created the connection that when I met Benoit and Terry. And when I first met with them, they had this, What they would do is the way they would interview people is they would ask them to do a presentation in front of a group of them. So I came in as a CEO candidate and they essentially

Starting point is 00:14:17 asked me to talk about what I was doing at Juniper and some of the things I was trying to do from a pricing perspective there. And so I had a presentation that I essentially had to do in front of them. And then they talked to me about what they were building at Snowflake. And I recognize that although the product was early and it was still not fully functional, if it did though what they said it was going to do, that it would revolutionize the industry, that in fact, databases had always been limited in the number of users they can support and the size of data. And because of the architecture that Benoit and Terry had created with Snowflake, leveraging characteristics of the cloud, which were not available before, because of the way the cloud

Starting point is 00:14:57 works, it was possible to build a database that separated storage and compute and allowed those to be scaled independently, essentially letting you put in any amount of data you want and work with as many users as you want. I mean, remarkably, a single snowflake system can scale to essentially any size. And the way they described what they were doing made total sense to me. And so they seemed like pretty smart guys. And so I decided to take a bet on the fact that they would make it work. And fortunately, that was a good bet.

Starting point is 00:15:24 I recently had a guest on the show who is an investor and an incredibly intelligent guy. And I had asked him about what technologies today excite him. and the one thing you mentioned was just how much you loved Snowflake's products. So I'm curious if you could dive more into what sets Snowflake apart in the data warehouse space and what makes them so special and to allow them to create something that just isn't available in the market. I mean, I think it's multiple things, right? I mean, first of all, I think it's being at the right place at the right time.

Starting point is 00:15:57 I mean, you couldn't build Snowflake today. It's just a fact. You know, people are trying to introduce technologies that have characteristics like Snowflake. Good luck. You've got well-established players in the business, all of whom have a whole lot more money than you have if you're a small company. We were competing against companies that always had a thousand times more capital than we did. I mean, they had names like Amazon, Google, and Microsoft. And how do you compete with them?

Starting point is 00:16:19 Well, you know, the answer is you have a better product. I said many times that if we were just a little bit better, like 20% better than Amazon's competing product redshift, we would have been totally wiped off the face of the earth. But in fact, Snowflake was effectively infinitely better because it solved problems that Redshift couldn't solve. We were also fortunate in the sense that I mentioned Redshift. Redshift is the data warehouse that Amazon built. It's still available. It's a good product.

Starting point is 00:16:46 Always been a good product. But it doesn't scale. It was built using a more traditional database technology that was built in an on-premises environment. And it didn't have the cloud scalability characteristics that Snowflake had. And so what happened is that Amazon, in doing an incredible. incredible work establishing the cloud and frankly building a very good product in Redshift. What they did is they seeded the market for us. And customers who adopted Redshift, if they scaled and grew in size, they would need

Starting point is 00:17:15 an alternative solution because Redshift's ability to handle large amounts of users or data was much more limited. And Snowflake turned out to be the answer. So we were able to move a lot of customers over. The product was a lot better than anything else. It could solve problems that nothing else could solve. And I think the way we built it was in a way that was very friendly to customers. I mean, I give Ben Juan, Terry, almost all the credit for building a phenomenal product.

Starting point is 00:17:42 But I do take some credit in building a phenomenal company. And I feel very good about the values we put in place and the approach that we took. I always felt that in order for companies to adopt a product, they really want to like the company and want to like working with the company. I've dealt with so many difficult companies in my days that it really is it's such a breath of fresh air when you're working with companies that want to solve your problems. And we built into our values, this realization that helping the customer succeed. Our first value, which we put in place was we put our customers first. And I learned that value from honestly Jeff Bezos, who I was fortunate enough to spend a small amount of time with when I was at Microsoft.

Starting point is 00:18:27 And, you know, I realized how customer-centric Jeff made Amazon. And I wanted to make sure Snowflake was as customer-centric as Amazon. So, you know, we put our customers first. And, you know, in that value, the first line of that value is we only succeed when our customers succeed. Which turns out to be true from an actual revenue generation perspective, because Snowflake is a usage-based pricing model. And so essentially, we didn't get paid unless the customers used our product.

Starting point is 00:18:54 And so we had a lot of reason to want them to. to be successful in using it. We were motivated to do that. They were motivated to do that because they had a problem they wanted to solve. And we were in a good place to solve that. So, you know, in addition to an incredible product, we built a very customer focused company and a very values-based company. And that was important. Values were something I discussed in every team meeting I ever had. We put the values together about 12, 18 months after I got there. It was a bottoms up process. It started in the engineering team with some leaders in engineering. It included a process that touched on every group in the company and people had an opportunity

Starting point is 00:19:33 to contribute to that. When we had these values, we really embraced them and drove the company that. The other thing I'll say is that Snowflake didn't try and solve every problem ourselves. There was so many, the space is so large. So, you know, we put in place a company that was very partner-centric that works with partners across the industry and does so very openly. And I'm pleased that that culture is still in place today. I think that came from Microsoft. Myself and they had a product at Snowflake now, Christian Kleiman's X-Microsoft, bunch X-Microsoft DNA now and there. And Microsoft, in my opinion, is the most partner-centric company that the world has ever seen. So, you know, we learned there. Let's take a quick break

Starting point is 00:20:11 and hear from today's sponsors. All right. I want you guys to imagine spending three days in Oslo at the height of the summer. You've got long days of daylight, incredible food, floating saunas on the Oslo Fjord, and every conversation you have is with people who are actually shaping the future. That's what the Oslo Freedom Forum is. From June 1st through the 3rd, 2026, the Oslo Freedom Forum is entering its 18th year bringing together activists, technologists, journalists, investors, and builders from all over the world, many of them operating on the front lines of history. This is where you hear first, hand stories from people using Bitcoin to survive currency collapse, using AI to expose human rights abuses, and building technology under censorship and authoritarian pressures. These aren't abstract ideas.

Starting point is 00:20:59 These are tools real people are using right now. You'll be in the room with about 2,000 extraordinary individuals, dissidents, founders, philanthropists, policymakers, the kind of people you don't just listen to but end up having dinner with. Over three days, you'll experience powerful main stage talks, hands-on workshops on freedom tech and financial sovereignty, immersive art installations, and conversations that continue long after the sessions end. And it's all happening in Oslo in June. If this sounds like your kind of room, well, you're in luck because you can attend in person. Standard and patron passes are available at Osloof Freedom Forum.com with patron passes offering deep access, private events, and small group time with the speakers. The Oslo Freedom Forum isn't just a conference, it's a place where ideas meet reality and where the future is being built by people living it. If you run a business, you've probably had the same thought lately. How do we make AI useful in the real world?

Starting point is 00:21:59 Because the upside is huge, but guessing your way into it is a risky move. With NetSuite by Oracle, you can put AI to work today. NetSuite is the number one AI Cloud ERP, trusted by over 43,000 businesses. It pulls your financials, inventory, commerce, HR, and CRM into one unified system. And that connected data is what makes your AI smarter. It can automate routine work, surface actionable insights, and help you cut costs while making fast AI-powered decisions with confidence. And now with the Netsuite AI connector, you can use the AI of your choice to connect directly

Starting point is 00:22:36 to your real business data. This isn't some add-on, it's AI built into the system that runs your business. And whether your company does millions or even hundreds of millions, NetSuite helps you stay ahead. If your revenues are at least in the seven figures, get their free business guide, domesticifying AI at net suite.com slash study. The guide is free to you at net suite.com slash study. NetSuite.com slash study. When I started my own side business, it suddenly felt like I had to become 10 different

Starting point is 00:23:08 people overnight wearing many different hats. Starting something from scratch can feel exciting, but also incredibly overwhelming and lonely. That's why having the right tools matters. For millions of businesses, that tool is Shopify. Shopify is the commerce platform behind millions of businesses around the world and 10% of all e-commerce in the U.S. from brands just getting started to household names. It gives you everything you need in one place, from inventory to payments to analytics. So you're not juggling a bunch of different platforms.

Starting point is 00:23:40 You can build a beautiful online store with hundreds of ready-to-use templates, and Shopify is packed with helpful AI tools that write product descriptions and even enhance your product photography. Plus, if you ever get stuck, they've got award-winning 24-7 customer support. Start your business today with the industry's best business partner, Shopify, and start hearing... Sign up for your $1 per month trial today at Shopify.com slash WSB. Go to Shopify.com slash WSB. That's Shopify.com slash WSB. All right. Back to the show.

Starting point is 00:24:22 It's quite amazing to me how you and your team at Snowflake managed to convince the big tech players, the Microsofts, the Amazon, the Googles, convince these guys to work with you instead of heavily competing with you. I'm not so sure about that. They always competed at the same time. I mean, you know, it was always, and it varied. I mean, I had a horrible relationship with Amazon while I was running Snowflake, largely because of my past history with Andy. Andy and I had our first tangle when I was at Microsoft, Andy Jassy, who now runs Amazon.

Starting point is 00:24:54 So Andy never forgave me for the years ago when he was building AWS and he wanted to license Windows Server. And I was happy to license him Windows Server. I just wanted to license it the way I licensed it to all my other customers. And he wasn't happy about that. So I don't think, I think that was there. for a long time. And I was able to build a reasonable relationship with the Microsoft team when I was running Snowflake. And then as soon as I left, it totally reversed. And, you know, Frank built a good

Starting point is 00:25:20 relationship with the Amazon team. And Microsoft, the Microsoft thing soured. But the cloud guys were always a little bit of a co-opetition. And a lot of it was competition, to be honest. To my understanding, you still have a steak in Snowflake today. I just read that you unloaded some of your steak at the IPO. Would you consider there to be any major competitors to Snowflake today? Or what's your view on the competitive landscape? Oh, it's very competitive right now. You know, there are five, I always say this. There are five major data providers, data platform providers. You know, this is often called the modern data stack. You know, it's really this idea that you take, you know, you can consume data from any source. It goes into a cloud service. The cloud is, it provides you with the

Starting point is 00:26:06 scalability and the flexibility that. that's there. And SQL is used to help cut, you know, to help work with data. So the modern data stack has become very pervasive. In addition to Snowflake is a platform provider, data bricks is another independent provider that is done quite well. They focus more on the machine learning side, less on the data warehousing side. But they're building data warehousing and Snowflakes building machine learning. So they're competing very fiercely with each other. And then there's the three cloud providers, all of which have viable products these days. Amazon has Redshift in their data product line. Google has big query, and Microsoft has built what they now call fabric, which is an integrated set of data products. So they all have products that compete with Snowflake. I still think Snowflake's the best product in the market, and Snowflake's ahead in a variety of ways, but it's a highly competitive space, which I think is great for the industry. I think that's a super good thing.

Starting point is 00:26:58 And I think one of your jobs with Snowflake was to create that successful business model within the company. And one of the things I found interesting in that development. is you didn't go the subscription model route. You ended up essentially the clients pay for what they use with the products. So how critical was this commitment? And why was this the appropriate route for you to go? Well, it was the only route at some level because when you move to a cloud platform like Snowflake is,

Starting point is 00:27:27 you're paying, you have real cogs underneath you. So you have to have a model that is somewhat driven by usage. And the more directly correlated, those two are, the better you are. you are. So usage-based pricing had been in place previously. And in fact, Amazon probably should be given credit for driving that with AWS, which is almost entirely a usage-based pricing model. Now, Amazon is an infrastructure as a service provider. And so their pricing model is very physical in its nature. You are literally paying per computer you buy or the amount of storage you're using or the amount of data you're flowing across the network.

Starting point is 00:28:08 All of those are measured and you're charged for your usage associated with it. The biggest thing that I did at Snowflake when I put the pricing model in place was move away from a very physical-based model to a logical model. So instead of saying, you know, a customer gets a warehouse that has four nodes in it, you know, we said that was a medium warehouse, I think is what that is. And so I went to T-shirt sizing on it where each warehouse. size doubles the previous size, just like, you know, it's a small, medium, large, extra large. I think they go up to 6xL now.

Starting point is 00:28:42 So it's, you know, gets to very large clusters of servers that can work together. And we created this idea of a credit, which is a effectively, you know, a credit is an hour of usage of one of these nodes. But by calling it and creating it as an object, a virtualized object, it allowed us, it made it easier for us to discount it, easier for customers to consume it. And I very much wanted customers to think about Snowflake as a value-based service, not a physical service that you're paying for the rental of these hardware, because it isn't that. It very much is an application service, a platform service for customers.

Starting point is 00:29:20 And so by moving to credits and this t-shirt sizing, it moved to a much more logical approach to describing things, which allowed us to apply a discount. The biggest thing about a credit is it's a vehicle, it's something to discount that customers, you know, based on how much they're purchasing, et cetera, you could give a customer an appropriate price. And so that's what put the whole thing together. It is now pretty widely accepted in the industry as a model as a variation of the usage-based pricing model. There are a lot of details in there, too, by the way. Like, for example, a customer buys $100,000 in capacity. They can use that capacity any way they want. They can use it for compute. They can use it for storage. They can use it in any

Starting point is 00:29:59 region in the world they want. And when they run out of capacity, they go back to essentially book pricing, list pricing. They lose their discount. So the customer and the company are incented to do another deal and purchase more capacity. And that model has worked very, very effectively. I had mentioned that you had unloaded some of your shares at the Snowflake IPO. And to my surprise, I was reading on a, according to Forbes anyways, they said that it was Berkshire Hathaway that I'd purchased shares at the IPO from you. And I think people like to say that Warren Buffett purchased shares in Snowflake, but I think it's safe to assume that some of his colleagues did the research on that one and made that decision. And Buffett, he's generally had a bit of a distaste for IPOs

Starting point is 00:30:46 and technology in general, as many in the audience know. You know, he thinks about things like the Wall Street incentives of trying to get the highest price at the IPO. So I'm curious if Berkshire Hathaway investing in Snowflake at the IPO, if that surprised you, I didn't expect it by any means, and it wasn't something I was directly driving. I did know that was happening. I think it was great for the stock. I mean, it helped to really drive the stock up. I mean, I do think I feel a little bit bad.

Starting point is 00:31:11 No, I feel a lot bad, not a little bit bad. I feel a lot bad for public investors of Snowflake because very few, if any, people have made money on Snowflake in the public market because, you know, it opened at a very high price and then it went up from there. You know, and it's subsequently down below its initial, it's below what its initial offering. price was when this market first opened, I think it was 240 or something like that. So that to me is disappointing, but I wasn't shocked by what happened. The board very much wanted a big bang IPO. They got it. They got it. Problem is it's hard to maintain that afterwards. So transitioning to some of the ideas in your book, one of the great charts you included was what you called the arc of data innovation. So can you talk a little bit about this chart? It sort of shows

Starting point is 00:31:56 the how you see technology, how it's progressed over time in the past and then in the future. So I'd love for you to paint some color around this. Well, I'm, you know, I've been in the industry for so long and I've been working with a number of the players for so long. I think I've had a really, you know, unique viewpoint on how the industry is involved. And what the arc really does is it talks about the key data innovations that have happened over really the last 50 years. And it describes key things that have happened, you know, 1960, it's more than 50 years. In 1960s, you know, 1960s, the advent of static data or structured data and, you know,

Starting point is 00:32:31 some of the early database products, relational coming out in the 1970s and then really taking off in the 1980s. We have text and internet and search, you know, appearing in the 1990s with semi-structured data coming from that, all the log files being thrown off by these business systems, these applications, web servers, and now you have the ability to analyze behavior. And then in the 2020s, the modern data stack making it possible for people to analyze that data at real scale. You know, I'd always seen an arc of data innovation and the book always had in it an arc of data innovation. But where my head was was that it was all about driving better decision making, the digital data economy,

Starting point is 00:33:14 where the world is today being driven by data. And that was the world I sort of saw when I started writing the book. And then in the 20 months or so that I spent. writing the book, I watched how the industry was advancing in the areas of AI, these foundation and large language models. And I was sort of caught, like many of us, caught breathless by how fast things were going. And I realized that the arc had changed in that this idea of artificial general intelligence, you know, a machine that is as smart as a human being. I have for my entire career believed that's where we were headed, that, you know, people were building and were going to

Starting point is 00:33:55 build such devices. I've always believed that. But I thought it would be more like 2100 or 2050 when that happened. And I figured I wouldn't be around to see it. And now I think it's going to happen by like 2030. And I sure hope to be around in that time frame. And so I recognize the horizon for progress had moved in considerably. And so the arc now changes and talks about things like artificial and general intelligence and even super intelligence and beyond. Tapping into artificial general intelligence there, you mentioned that's technology that's as smart as like the median human, I believe you say in the book. That's that was Sam Altman's, that's one of Sam Altman's definitions and seemed like a reasonable one to me. Yeah. So, you know, you mentioned originally

Starting point is 00:34:40 you thought it was going to be, you know, far out into the future. What were some of the key things that led you to believe it's going to be coming much sooner, say, 2030? Well, my first, I mean, I work with Microsoft, and sometimes I get a preview on things that they're coming at Microsoft. And I had seen in early 2022 the co-pilot work that the GitHub team was doing. And I found that breathtaking. I mean, I found that sort of breathtaking that this technology was writing a material part of the code for a programmer.

Starting point is 00:35:12 And Microsoft believed that that number would be around 40%. And that seems to be holding true. That when developers use copilot, that 40% of the code that they check in is actually written by copilot, which is a stunning number. I mean, you're talking about a massive increase in productivity of developers, which is one of the biggest gating factors to technology advancement, is how fast can developers write code. All of a sudden, we have a very significant 30, 40% increase in capacity. from this technology, and that seemed pretty breathtaking. And then I watched, you know, what the stuff was happening with stable diffusion and the dolly and the drawing apps. And then like the rest of the industry, I was just caught sort of

Starting point is 00:35:55 breathless by how fast and how far chat GPT has come. And so that made me realize the world is totally changing and changing it as a pace that's much faster than I anticipated. And the really, the remarkable thing about this is that for the first time we have what you can think of as intelligence, the ability for machine to make independent decisions that are not driven by logic that was created by a person. You know, programs are written by people and they're very logical in the way they do things, and that's the way a lot of artificial intelligence used to work, rules-based things. Now with these neural networks that have grown in very large scale, all of a sudden these networks have a type of intelligence that is very similar to human intelligence in the ability

Starting point is 00:36:39 to think through processes. It's not as advanced yet as human intelligence. there's definitely missing elements, but the technology is able to solve problems that literally were unsolvable two years ago. And all of a sudden, there's this vast set of problems that I've wanted to solve and that I have companies that want to solve that. Now, all of a sudden, you can solve them. And essentially this idea that you can take any task, any process that people do, and knowledge that is in people's heads on how to do that. And you can effectively bottle it and put it inside a piece of software and take that knowledge, that domain knowledge that you have and actually make it so that a computer can replicate that.

Starting point is 00:37:18 It's a remarkable advance, and it affects effectively everything. So it's an incredibly exciting time. One of the amazing things that sort of stands out to me is, you know, I look at the founding story of Snowflake and how they're in stealth mode and no one knew what they're really up to. And, you know, today everyone's familiar with Chad GPT, but what we're not familiar with is all the things. the Googles, the Microsofts, the open AIs of the world, what they're developing that they haven't released yet. Right.

Starting point is 00:37:47 And we're in that period. We're in a waiting period right now. I actually feel like, you know, there's, Gartner has this thing called a hype curve. I don't know if your viewers, if your readers or your listeners are familiar with this, but it is, is when a new technology is introduced, it goes on a hype cycle. And it's sort of the curve goes up at a very fast, at a fast pace of hype, high, pipe, and then the hype hits the peak of hype. And then it goes into what they call the trough of disillusionment where reality strikes.

Starting point is 00:38:13 And in that period, you know, people become disillusioned about things. And then eventually it goes into a cycle of usefulness. And it has a lifetime afterwards where people understand what really is possible. Well, the first six months of this year were the hype. You know, I've never seen hype like AI hype. It was the most fast hype, biggest hype I've ever seen in my career. And I think we actually hit the peak of that hype cycle sometime in early July. And I think we're now beginning to enter this trough of disillusionment as people are saying,

Starting point is 00:38:45 okay, fine, how does this affect me? How does this affect my job and what I do? And we're waiting for products right now. And there's a chance that those first products might be a bit disappointing, which is what often happens, when first generation products come out. And so that's all part of that trough. And now the hype cycle went as fast in AI as anything's ever gone. What's really going to be interesting now is how quickly do we go through this trough of disillusionment?

Starting point is 00:39:09 I think the fall will have some disillusionment? But by the time in winter, will we begin to exit that and start to see real products that are solving problems? I don't know. We'll find out, won't we? I'm hopeful, but only time we'll tell. In real time, it sort of feels like it's going to develop slow. But when you look at the bigger picture, it's happening at a very rapid pace as you've seen throughout your career and how the timeline is sort of shrunk with these technologies. I'd like to get your point of view on what industries you think will be most impacted by the advancement of AI.

Starting point is 00:39:41 Well, I think that, you know, you hit on the real big thing, actually, which is the shrinking of timelines. And that's what's happening. And if you, you know, you talk about the trend in the arc of data innovation, it's actually been a, it's been a constant speeding up of how things move over time. If you look back in the 1970s and 1980s, you know, this was before email even. And information moved between people at a much, much slower pace than it moves today. So technology has sped up and the ability for people to work with and exchange data has constantly been increasing the pace of innovation. AI is exactly the same way.

Starting point is 00:40:17 It will be a significant increase in the pace that innovation happens. And ultimately, it may continue to go faster than really we can even really understand. We'll see over time. You know, you ask about what industries are impacted. I think a bigger question is when industries are not impacted, and I can't think of any. I mean, every industry is impacted with this, because if you look at what is behind every industry, intelligence is a big part of every industry. I mean, if you sort of say, what's behind an industry, well, you've got labor, for sure. So there's that. There's intelligence

Starting point is 00:40:49 and there's knowledge. Those are all elements of every industry. Well, we have, computers have done an amazing job of storing knowledge. And that is really, you know, what is not knowledge. Knowledge is data that has been analyzed and a conclusion has been reached. That's thought of his knowledge. And, you know, sometimes those conclusions are correct. Sometimes they're wrong. Society over time reaches what are generally believed to be correct knowledge conclusions. Often that gets encoded in things like Wikipedia. Sometimes it's right. You know, mostly it's right. Occasionally it's wrong. But that's knowledge. So we've had knowledge. Now we have intelligence for the first time that we can combine with that knowledge.

Starting point is 00:41:29 to solve problems. And so every industry is, you know, is going to be impacted to some degree. What industries are the least impacted? Well, it looks like in the short run, the industries that are the least impacted are the ones that involve human labor in some ways. I mean, it's been said that the last, you know, we thought that the jobs that were going to be impacted in the short run would be, you know, drivers and things like that. Well, that may happen. But in the short run, the impact is probably on information workers and how they work in their jobs. And the people that are cutting lawns and doing all of the tasks that are required in life are probably the least impacted in the short run. Now, that may change. I believe that will change in the 2030s because

Starting point is 00:42:10 the 2030s, I think, is the era of robotics. And that's where we will really begin to have robots live and work with us as a part of our daily lives, whether it's autonomous vehicles or robots that help us to clean the house or care for elderly people or cut the lawn. I mean, all of those things are going to happen. Some of that is a bit further away, though. Let's take a quick break and hear from today's sponsors. No, it's not your imagination. Risk and regulation are ramping up and customers now expect proof of security just to do business. That's why VANTA is a game changer. Vanta automates your compliance process and brings compliance, risk, and customer trust together on one AI-powered platform. So whether you're prepping for a SOC 2 or running an

Starting point is 00:42:57 enterprise GRC program, VANTA keeps you secure and keeps your deals moving. Instead of chasing spreadsheets and screenshots, VANTA gives you continuous automation across more than 35 security and privacy frameworks. Companies like Ramp and Ryder spend 82% less time on audits with VANTA. That's not just faster compliance, it's more time for growth. If I were running a startup or or scaling a team today, this is exactly the type of platform I'd want in place. Get started at vanta.com slash billionaires. That's vanta.com slash billionaires. Ever wanted to explore the world of online trading but haven't dared try?

Starting point is 00:43:38 The futures market is more active now than ever before, and plus 500 futures is the perfect place to start. Plus 500 gives you access to a wide range of instruments. The S&B 500, NASDAQ, Bitcoin, gas, and much more. Explore equity indices, energy, metals, 4X, crypto, and beyond. With a simple and intuitive platform, you can trade from anywhere, right from your phone. Deposit with a minimum of $100 and experience the fast, accessible futures trading you've been waiting for. See a trading opportunity.

Starting point is 00:44:13 You'll be able to trade it in just two clicks once your account is open. Not sure if you're ready, not a problem. Plus 500 gives you an unlimited risk-free demo account with charts and analytic tools for you to practice on. With over 20 years of experience, Plus 500 is your gateway to the markets. Visit Plus500.com to learn more. Trading in futures involves risk of loss and is not suitable for everyone. Not all applicants will qualify. Plus 500, it's trading with a plus.

Starting point is 00:44:44 Billion dollar investors don't typically park their cash in high-eal. savings accounts. Instead, they often use one of the premier passive income strategies for institutional investors, private credit. Now, the same passive income strategy is available to investors of all sizes thanks to the Fundrise income fund, which has more than $600 million invested in a 7.97% distribution rate. With traditional savings yields falling, it's no wonder private credit has grown to be a trillion dollar asset class in the last few years. Visit fundrise.com slash WSB to invest in the Fundrise income fund in just minutes. The fund's total return in 2025 was 8%, and the average annual total return since inception is 7.8%. Past performance does not

Starting point is 00:45:34 guarantee future results, current distribution rate as of 1231, 2025. Carefully consider the investment material before investing, including objectives, risks, charges, and expensive, This and other information can be found in the income funds prospectus at fundrise.com slash income. This is a paid advertisement. All right. Back to the show. On this arc of data innovation, the last four parts I'm reading here is intelligent robots,

Starting point is 00:46:02 humanoid robots, superintelligence, and then technological singularity. So I'm sure you've thought a little bit about sort of the endgame of this arc of data innovation. So I'd love to get your take on this as well. Well, you know, again, I sort of have always viewed, I mean, I started by saying that when I was a young man, I spent a lot of time reading Isaac Asmoff in the early days and went through, he wrote over 400 books, which is ridiculous. And I can't say I read all of them, but I read a lot of them. I've covered a lot of Isaac Asmoth. And so I'd always had in my head this idea that people would have developed intelligent robots that are machines that work and live amongst us, because that's what many of Asmos stories talked about. And I always believe, that people would develop these systems that go beyond us. And in some senses, that's our purpose, is to build something that can take the next step. And again, I never thought I would see this. I thought this was all post my lifetime. And now I see it coming closer. The idea of, first of all, you've got artificial intelligence, which is essentially a machine that's as smart as a person.

Starting point is 00:47:05 Superintelligence is when these machines continue to get smarter and smarter where they're really smarter than all of us. That's the idea of super intelligence. And, what's happening through this process and when we feel it and see it almost every day in our lives is an increase in pace in innovation. Things are happening faster and faster. And that, you know, potentially will continue to increase. And if, in fact, we do build these machines that are very intelligent, that will continue to increase the pace of innovation. What a technological singularity is, is effectively a situation where machines begin to advance science and technology at a pace that is beyond human capacity to really understand. And that's this idea that things go very, very quickly.

Starting point is 00:47:48 Ray Kurzweil was the one who first brought this up many years ago with his book about the singularity is near. I've had a chance to meet Ray once in my life and I think he's probably right about what he mostly wrote. He would tell you, I think it's a lot sooner than even he thought now. And so that direction seems to be happening. It may not happen. We don't know for sure. But to me, the real key is is let's make sure that as we build these machines that may be able to do things beyond us, that we instill within them the values that we think are important. I mentioned values earlier on. I can't stress the importance of this enough in building companies, in leading your life.

Starting point is 00:48:25 But for goodness sakes, when you're building technology, the values of the people are imbued inside the service and the technologies that are created. And I can see these things. I know the values of Microsoft. I can see them inside their products. You can see the values of meta inside Instagram and Facebook. You can see the values as much as they exist in Google inside the Google products. And so as we create these new things, the values that we imbue in them will be very, very important and will direct what they do.

Starting point is 00:48:58 For those listening, what was the name of the title you held up there? Oh, it's the title. It's the singularity is near. He wrote this book. He wrote this book about 20 years. ago. But it just released a newer version, a new book that updates this. You mentioned Asimov, and he's someone you touch on a lot at the end of your book. He sort of helps you shape your framework around the governance of these sort of types of intelligences. So can you talk about, you know, the role of Asimov,

Starting point is 00:49:25 you think the role, some of his ideas are going to play into this? One of the, you know, pieces that are outlined in your book is what you call Asimov's Law of Robotics. So it's kind of a way of coding these things in a way where, you know, we can kind of control where this is all going to be going. You have to put this in perspective of this guy. I mean, he was, Asimov was a brilliant man. I think he was actually a prophet in the sense that he saw things ahead of where the rest of humanity was. In the early 1940s, before digital computers were invented, Asmoth was toying with this idea

Starting point is 00:50:02 of intelligent robots living amongst people. And as he began, and he was a science fiction writer, that's what he was. He was a very good one, but he wrote stories. They're all stories, right? Fictional stories. And he had this idea that unlike previous generations of people that thought of intelligent beings as Frankensteinian monsters, you know, that were created, you know, by man that proved that man should not create these things. I mean, that's essentially what thousands of years of history associated with humans creating super intelligent things or robotic sorts of things, it came from. Asimov, unlike all the previous that came before him, the writers that came before him, he saw these devices as machines that were created by people to serve people. And, you know, he recognized if you're going to have machines interacting with people helping us with tasks, there needed to be some rules that they operated by. And so in the early 1940s, he came up with the three laws of robotics. The first law, a robot may not harm, may not injure a human being or through an action allow a human being to come to harm. That's the first law and it dominates everything. The second law, a robot must obey the orders given to it by human beings, except where such orders would conflict with the first law,

Starting point is 00:51:15 so it has to follow orders. And the third law is a robot must protect its own existence, as long as such protection does not conflict with the first or second law. So this idea that these creatures would live amongst us, but following very directly these laws. And of course, you know, humans do not, right? We follow, we do what we do it. And most of as, Osimov's stories about robots are actually all the stories are parables of how you live and work with robots where robots are following these laws and people are not. And, you know, so this idea like, like, what does it mean to not injure a human being or to have a harm come to a human being? It's very vague, right? It's not obvious. Well, Asimov spent many of the stories talking about that in different types of harm and how that happened and how the robot would react following the laws.

Starting point is 00:52:01 And so it's a chance to think through these issues because these are. are values essentially. And as we create these large language models and these intelligent machines, unlike Asimov's robots, which follow these laws, you know, because they were hard-coded in their positronic brain, these are models that are very malleable created by people. And what they do will be dependent on what we tell them to do. And so it's very much based on values in, values out. And so I think the industry has recognized how important it is that these models operate with a high set of ethics and standards, and they will be built based on the values of people creating it. Now, since I wrote the book and finished the book, the thing that

Starting point is 00:52:48 has changed in the industry that is an incredibly positive change is that in addition to having GPT4 from OpenAI and BARD from Google and whatever. The big, big companies, we now are seeing open source models be introduced, which have very powerful capabilities, but can be used by anyone to do effectively anything. Now, some of those things will be not so good. People will do bad things with them. These models are tools just like anything else humans have built. And every tool that people have built has been used for every possible.

Starting point is 00:53:25 purpose, good, bad, and evil. And that's going to be true for these large language models as well. But because they're open, we'll also see lots of good things come from it. And I think the fact that there is a lot of innovation will allow us to stomp down the bad uses and focus on the good uses. That's the thing that's amazing about this. It's such a multipurpose tool. I think we're really getting at touching on, I think, a key player in this, obviously, is governments and regulators. I'm curious if you think that. that today's governments are taking appropriate precautions to safeguards against some of the potential downsides. In other ways, it almost feels like an impossible job. So I'm curious to get your

Starting point is 00:54:06 thoughts on this as well. Well, when January was, and I was finishing the book, it was unclear how people were going to react to this. And I was worried that I didn't know how strongly people would react. I'm no longer worried about that. The volume has been as set to 11 since then. And every possible concern has been written. I mean, there's a new article every day. You know, it's all over the business press. It's in the New York Times. It's in, you know, Vogue magazine. It's everywhere.

Starting point is 00:54:33 And so the concerns are very present. And there are valid concerns. And there are needs for some government regulation. To me, a very good example of that is these deep fakes. You know, it was never possible to build a video of a person saying something that they don't believe. I mean, you could never really get away with that. You can now. I mean, these systems, it is possible to create a deep fake, which is pretty indistinguishable

Starting point is 00:55:00 from an original. And while it's possible today, it's probably going to be trivial within a year or two. And, you know, there'll be apps for the phone that kids will be able to do it with. This is a potentially very dangerous thing. And we do need to make sure that laws prevent people from impersonating others without their consent or acknowledgement. You know, I don't have any problems with people kidding other people, you know, comedians have been mocking people for decades, as long as you say it was created by AI and you

Starting point is 00:55:31 don't misrepresent it, whatever. But when you misrepresent it, I think it should be illegal. And that's an area where regulation is required. Some of these other concerns like of how this things get super intelligent and things like that, it's too far out to regulate. There's no way to regulate those things. It's just too soon to understand what the issues are. But as we begin to build real, as real products emerge from AI and capabilities appear, AI will be an incredible spamming device. You can use it to spam people. You can also almost certainly use it to block spam. So we'll see all of these things happening. And some of the laws like anti-spamming laws will apply directly because they're just a new tool people are using. But there are cases like perhaps with defakes where new

Starting point is 00:56:12 laws are required. On the other hand, I think we should be cautious and not overregulate because government can't possibly anticipate the way the industry is going to go. And you mentioned spamming. I think it's already an issue today, you know, just being in the- It was an issue before AI, though, wasn't it? I mean, I got plenty of it before AI, it seemed like. But it'll get worse. It'll get worse. There's going to be plenty of spam in the YouTube video for this conversation even. So since we mentioned the, you know, potential dangers of AI, can't help but think of Elon Musk, who's been very outspoken about, you know, where companies like alphabet are heading. And he was actually involved in the creation of open AI, which, To my understanding, started out as a non-profit organization, but is now for-profit. And they also received a $10 billion investment from Microsoft. So I'm curious if you believe that Elon Musk or his companies play a role in the development of the future of these technologies and AI. They surely do.

Starting point is 00:57:11 And he's investing, you know, he's investing actively in building artificial intelligence. You know, I think it's connected to his new X service that he's creating. Whether you, you know, like Elon Musk or hate him, he has a right to build his own solution around there, just like Meta has their right and Google has their right, Microsoft and Open AIA have their right. And we'll see all these companies build things. To me, the great thing is now that we have these open source models, and by the way, I have to give Mehta and Mark Zuckerberg incredible credit for releasing the open source Lama 2 model recently, which is really having a dramatic impact in the industry. It is, it does appear to be. be a really, really good model in open source and it's competitive with some of the frontier models from companies like Open AI even. But what this means is that we're going to have every possible solution. You're going to have counselors that help counsel people that are AI counselors. Hopefully, we'll have all kinds of tutor bots. I look forward to tutor bots that help to tutor young children. Lord knows our education system. I don't know that that's the answer to our education system.

Starting point is 00:58:14 I'm not claiming that's the answer to our education system, but Lord knows something needs to be done to our education system. And this might help. I can see ways where this could help. I'm not claiming it's a magic bullet, however. You know, we will see, like I say, psychologist bots. We're going to see answer bots. We already see some of those that answer questions. They do a pretty good job of that today, actually, a pretty amazing job in some senses. Soon we're going to see action bots that do things for us, you know, like schedule a reservation at a restaurant for us where we don't have to go through the process of doing it. Just tell it what to do and it'll do it. You know, these things are all going to appear and they're going to be from different companies. And Elon will have his bot. He'll have his

Starting point is 00:58:51 ex-bot. And, you know, Mark Zucker will have his Facebook bots. So, and of course, Google will have theirs. And now lots of little companies will be there too. Given your extensive experience with Microsoft building out what all happened at Snowflake from zero to 200 million in revenue, now you're on the board of a number of different companies. I think it's fair to say you know a thing or two about differentiating a successful technology business versus an unsuccessful one. So I'm curious if you could share some of the key things you look for for companies that you invest in in this space. Well, I'm not a traditional investor. Let me just start by saying that because my number one concern is actually learning and advancing technology in areas that I care about, not so much

Starting point is 00:59:35 return on investment, although I try and make investments that will have rational returns. I try and be rational about investments. So my focus is always on things that are new and cutting edge that are in the data space that are solving problems that couldn't be solved before. Like an example of a company I've been involved in is called Docugami. It's solving the problem of taking business contracts and turning those contracts into data that can be actioned by an organization. I mean, today contracts are programs that are interpreted by lawyers and executed by people. And over time, that program is going to be interpreted by AI and executed by computing systems. And Docugami is playing a pivotal role in connecting those dots together.

Starting point is 01:00:17 It's a class example of a problem that could not be solved three years ago. I mean, literally with the technology three years ago, it was not there. Now it's there and it's actually working. So to me, it's about innovative things that break new ground in data. And I'm looking at new ways that you can apply the relational model to both analytics as well as to operational applications. And my investments in small companies fall into these categories. One of the key points that sort of stood out to me in your book,

Starting point is 01:00:46 you talked about how everyone's aware that software is eating the world. And you had this quote, in the next 10 years, you predicted that models will eat software. Could you explain what you meant by this? So we've always built software directly to do things. Today, we write software very, very directly. And it's focused on, you learn what something. something does, you write a piece of code that solves that problem. Where we're moving towards

Starting point is 01:01:10 a world is where we create essentially a twin, a digital twin of an organization. And that is a model of the organization what it does. Today it's very difficult to create to do that. And software does it in a way, but it's very opaque and it's not structured in a logical sense. I think over time, the way we operate our business and run things will be to have these models that define what our business process is. And as we learn from that, that model reflect what that business process does. All of these machine learning and artificial intelligence things are models. They are a type of model that emulate and essentially are that they're emulating some physical thing and doing it in a virtual world. And, you know, those models are going to become more and more of the software that

Starting point is 01:01:58 we build. If you look, it's happening already. The entire software industry is moving now this year to perfect models and to take these language and artificial intelligence models and tune them, fine tune them for different applications. So instead of writing code specifically to do things, we're going to take these models that are general purpose models and apply them to solutions. And that's the new type of coding. That's the way the coding is going to work in the future. So models are going to eat the old way of doing software. Quite interesting. And I think a point that sort of ties into this is thinking about the investment and asset management industry.

Starting point is 01:02:34 many likely speculate that, you know, eventually software is going to take the jobs of many investment managers. I'm curious if you have any thoughts on this. You know, I think investment managers are people that work with people and talk to them. So I don't, it's just like everything else. There's the software will make the job of an investment manager different for sure and maybe easier in some ways. But I think the human interaction is still going to be important. And I think that we'll continue to see that in in most industries. That, in fact, this software won't replace people. It will augment what people are doing. That's the way technology has always worked. I think this is the way it's going to work this

Starting point is 01:03:11 time. That said, when we have technological disruptions, while it does create a lot of new jobs, it does impact people in their current roles. So some people will, you know, whatever, how the world changes will change in a way that is difficult for some people to actually make the transition. So it has a human impact. And I don't want to, I don't want to diminish the importance of that human impact. But in general, I think it will advance society and help things for people, including for investment managers, who I think very much play a real role going forward. I think another glaring questions in our audience's minds as investors with the rise of Chad GPT is its impact on other business models, especially one like Google Search. Google Search is a

Starting point is 01:03:54 business that earned over $162 billion in revenue in 2022. I'm curious, if you believe this business model will be totally disrupted by technologies such as chat GBT within the next, say, five, 10 years? I think search will be totally different 10 years from now than it is today. Let me start by saying that. I don't think search will be, you know, the way the Google is today with the 10 blue links and the zillions of ads, all the DM ads in front of it. So I think that it will be disruptive in the sense that these answer bots are already very

Starting point is 01:04:28 disruptive. I use an answer bot, which is one that I'm an investor in called perplexity that I think does a nice job of giving you good summarized answers to your questions with references to tell you how it got to those answers. And it avoids hallucination by working with current data. And I think they already do a better job than search for a lot of problems. And I'm already switched, I've switched away from Google to those problems. Interestingly enough, where Google is still most useful is where you need the ads, where you want the ads. That's where Google is really particularly good these days. But other than that, some of these other things can solve a problem. You know, search is the most profitable. It's the biggest and most profitable app on the planet.

Starting point is 01:05:07 Let's just start with that. So in terms of, you know, you give the numbers, in terms of any app, nothing is bigger than that. It's the biggest it gets. And it was impenetrable. Google was impenetrable until now. I, you know, was at Microsoft. You know, we competed with using, we did and still, we did, and they still do compete using with Bing. I watched us try and compete. It was one of the most heart-wrenching things to watch because, you know, what I learned from the leaders at the time from Microsoft was how just because of the way the industry is structured, it resulted in one big winner. You know, typically in an industry, you get a big winner. You get a winner that is the leader, but then you have three or four other companies that have material market shares as well.

Starting point is 01:05:48 And or at least two, like you've got, you know, Android and Apple. You have at least something like that. Search was really different because it just, it was a singularity. I mean, it turns out to be its own type of singularity where everything went to one vendor and the cost of running it was so high in different regions around the world that was almost impossible to replicate that. And the only company that managed to do a decent job of it is Microsoft. And that's because Satcha just hung in there and did an incredible job of turning being from a money losing thing to at least being a decent business for Microsoft, not a great. business, but a decent business. And then he hung in there for all this time until this new innovation

Starting point is 01:06:29 came out, which changes everything. It's a total change. And it's the first time where Google is vulnerable. Now, will they be unseated? No, I mean, I don't predict that. I don't predict that because I predict that Google will respond and build great products that take on this new paradigm. But I do predict they will lose share. I think they will almost certainly lose share. The other thing that's interesting is that with a behavior of people has been trained to go to Google, when we have these bots in front of us, I think we're going to be trained to go to them. So there may not just be one place you go. There may be many bots you talk to. And those bots will talk to search bots. So it may be that the portal is not the search browser in the future. It may be that whatever app you're working on. So the whole model has a chance of changing, but we haven't really seen that fully play out yet. I do predict Google will lose. share, but I don't believe they'll lose their position as the leader. Yeah, that's one of the things I'm really interested in seeing is what Google's plan is

Starting point is 01:07:27 going to be to reinvent itself, because I'm sure it's been, who knows how long, but it's been probably a number of years where they've sort of seen this coming and they're ready to try and make that pivot to its reinvention. They didn't act very ready, though. They didn't act that ready? I mean, they got kind of caught flat-footed, didn't they? I mean, they really felt to me like they got caught flat-footed, and they had all this technology. You know, the idea that Microsoft is leading Google in this is crazy. I mean, Google was so

Starting point is 01:07:54 far ahead in AI. Everybody knew Google was going to be the winner. And somehow, because of open AI and some really smart moves on Satch's part, you know, it really changed some things. And I think it's good because it's a shake-up that the industry needs. In general, I think it's useful for a large numbers of companies to have access to this technology to build their own solutions. And now we know that's going to happen. Well, Bob, it's such an honor having you on the show. Thank you, much for sharing your time with us. I highly encourage the listeners if you enjoy this chat, pick up Bob's new book called The Databpreneurs. Bob, before I let you go, please give a handoff to the audience on how they can maybe get connected with you or give the handoff to your book.

Starting point is 01:08:33 Well, they can learn more about the book by going to the datapreneurs.com. I mean, you learn a lot up there. There's a bunch of links to other podcasts and things up there as well, some information about the book. And I hope that people enjoy it. And if they do, please leave a review on Amazon. Got it. Thank you. you so much, Bob. Thanks a lot. Thank you for listening to TIP. Make sure to subscribe to Millennial Investing by the Investors Podcast Network and learn how to achieve financial independence. To access our show notes, transcripts or courses, go to theinvestorspodcast.com. This show is for entertainment purposes only. Before making any decision consult a professional, this show is copyrighted

Starting point is 01:09:13 by the Investors Podcast Network. Written permission must be granted before syndication. or rebroadcasting.

We Study Billionaires - The Investor’s Podcast Network - TIP575: The Future of AI w/ Bob Muglia

There aren't comments yet for this episode. Click on any sentence in the transcript to leave a comment.