Limitless Podcast - Why It's Not Too Late To Invest in AI | Dan Ives

Episode Date: September 2, 2025

In this episode, Dan Ives from Wedbush discusses the early stages of the AI revolution, introducing his AI Revolution ETF and forecasting a $2-4 trillion surge in tech capital expenditures. We explore key players like NVIDIA, the importance of semiconductors, and the potential for small-cap investments to thrive. Ives also highlights the intersection of AI and cryptocurrencies, offering bold predictions on future opportunities.------🌌 LIMITLESS HQ: LISTEN & FOLLOW HERE ⬇️https://limitless.bankless.com/https://x.com/LimitlessFT------TIMESTAMPS0:00 The AI Revolution Begins0:51 Are We Too Late for AI?4:23 The Fourth Industrial Revolution6:53 Mapping the Capex Super Cycle9:11 Risks on the Horizon10:34 Understanding AI Spending12:09 The AI Stack Explained13:38 Building the AI House16:02 Identifying Software Opportunities19:10 Power Law Winners in AI20:24 America vs. China in AI23:08 The Software Layer's Golden Goose25:58 Separating Substance from Noise26:30 The Private Market Dilemma43:19 The Convergence of AI and Crypto45:10 How Rich Can We Get?------RESOURCESDan: https://x.com/DivesTechRyan: https://x.com/RyanSAdamsJosh: https://x.com/Josh_Kale------Not financial or tax advice. See our investment disclosures here:https://www.bankless.com/disclosures⁠

That's what we spend all over time doing, right? We're in the field, like, 3 million air miles lost 25 years, right? Like, not from a hell. I mean, the point is, like, that's been our secret sauce, trying to find who the winners are. Who are going to be the winners in, like, robotics, autonomy, physical AI. The amount that's going to be spent in those areas could potentially dwarf what we've already seen today. We have Dan Ives on the podcast today. He leads tech research at Wed Bush.

they launched an Ives AI Revolution ETF. That's an AI ETF, and he's been pretty early and pretty loud about the AI CapEx SuperCycle. So Dan, welcome to the show. Yeah, great to be here. Okay, so Josh and I are investors. I have experienced mainly in crypto. Josh has more experience in AI. And I want to start with this question. Is it too late for me to get rich on AI, Dan? I mean, look, I believe were in the second inning

of the AI revolution. Because, see, it's my view, and this goes back to 2022 when it started with, you know, godfather of AI Jensen and Vidia and my view of when it was all starting. It's a 10-year buildout. So I strongly disagree.

Like when investors, like, oh, the stocks have run so much, I missed it. Dude, this is just to begin because the second, third, fourth derivatives across enterprise, consumer, semis, energy, robotics hasn't even started in terms of full scale. Autonomous hasn't started as well. So to me, that's going to be the holy grail.

Okay, that makes sense. So you're saying we're in this second inning of a nine-inning game, of course, but coming from crypto, we know that these innings, of course, can span multiple cycles and you can have boom-bust cycles across them. And so this is the Nvidia chart that I'm showing right now. And even by crypto standards, okay, like, we don't often see charts that look like this, which is just like a parabolic line up from 2023 up to now. Like, this is an incredible chart. And you're saying, Dan, you think that, like, buying this chart is actually a good thing to do at this point in time? It's hard for me to believe. Yeah, and that's been a, but remember, the last, I might like my career, what, 25 years in tech?

I mean, really for the last 15, investors have just fought it. No, it can't go out more. No, it's, because it's my view, we're in a fourth industrial revolution. And that's really been our view, not just Nvidia, a Pallenteer, of Microsoft, of everything we've seen across the tech sphere, is that you have three to four trillion that's going to be spent in the next two or three years. I'm not saying you're not going to have pullbacks, but it's my view.

We're going to be talking about $5 trillion, $6 trillion mark caps. And a lot of it just comes down when you look at Nvidia specifically, like there's one chick fueling the AI revolution. And it's led by godfather of AI Jen's in Invigiant. There's no one else that's within miles of Nvidia. Okay. Do you think that this fourth industrial revolution will play out across cycles, though, is kind of my question. So 25 years in tech investing, you've seen a number of like, you know, bull and bust cycles,

including, of course, you know, the full dot-com cycle. You had in 1995 all the way to 1999. And then you had a, like a massive boom, of course, the end of the 90s, and then a bust, which a lot of people said it's over, internet's never coming back, all of this thing. There was some recovery time. And ultimately, I think the Internet Bulls turned out to be right on that. But it took some time. And so even a fourth industrial revolution, even, you know, second inning of the game here,

but do you think that this still could play out in cycles? And the cycles may be just because of human psychology, right? Like, we're in crypto. We've seen so many of these cycles. Things get ahead of their skis. Yeah. Do you think this plays out in cycles? Yeah, see, normally I'd say yes, like in terms of cycles, right?

Because my whole career, that's really what I've seen. Here's the reason I don't think this plays out in traditional cycles. It's because when you go back, when I cover tech stocks in the late 90s and the boom and the bust, those are basically were frothy companies, business models that were really untested. The use cases weren't really there. And these companies were essentially back by VCs and why companies that are losing tons of money. Now, because big tech is to send you back. 325 billion of capax.

Sovereigns haven't even gotten in it. Enterprise in the U.S. only 4% are actually starting to focus on AI. There's none in Europe, none in Asia X China. I believe the next few years, this trend actually just multiplies. Like it's our view. Like, look, we spend so much of our time in Asia.

I believe there's more growth over the next two years. when it comes to semis and when it comes to just overall demand globally, then the last 10 years combined. So you believe this chart just keeps kind of going up. Maybe there's some pullbacks, but it's nothing like kind of like a 90. I think Amazon had a 95% drawdown, for instance.

One of the dot-com darlings, of course, that's quote-unquote went bust. And then I think it took them, what, to 2008, 2010, something like that, to like recover just from the dot-com highs. You don't think anything like that will happen in AI. Big tech companies, they have a trillion dollars of cash in the balance sheet. And they generate $200 billion a year in free cash flow. So the different, that's why a lot of times, like, you know,

when I talk to investors, retail, institution, whatever it is, when it's about dot-com bubble, like, I was a tech analyst in dot-com bubble. So, like, I was there front and center. It's so dramatically different because of the use cases and because of computing power and because of what we're seeing on the enterprise. But on the consumers, I'll be like, oh, chat, GPT, I'm not using it that much.

It's not about that. It's about one in every 10 to 12 households is going to have a robot. It's about autonomous is going to be 25% of mobility. See, we're talking about the beginning. of a transformational period of time as a, as a, not just an enterprise, but as a consumer. And we talked about cycles. I want to talk about the super cycle, and particularly the KAPX super cycle. And I'd love for you to kind of walk us through, kind of map the dollars

for us, Dan. So what exactly turns into shareholder returns over the next few years? I mean, you flagged two trillion dollars of AI spend over the next three years. Who captures that margin first, who captures it last? And where is it all heading to? Look, it speaks are like our I, is AI 30, and obviously ETS based on that, is that, okay, it starts with semis. Of course, starts with NVIDIA,

but then TSMC, Broadcom, you know, what I view AAMD, and there's a number of other partners in there. Okay, so it starts with semis, and of course, semi-equipment. Then it goes to software, hypers, microscalors, Microsoft, Amazon, Google, look at everything,

what Oracle's doing, renaissance to growth. You've even seen with IBM and others. But then it goes back to who wins in the use cases. So I'm just going through like almost a pattern. That's where messy of AI Pallantier, right? They hate the stock at 12, despise their 50, yelling from the mountaintops at 100.

And the bears, when they're in hibernation mode, they can't say AI in spreadsheets from the stock's 200. Then you think about data dog, MongoDB, Elastic. Okay, so now how do you protect it? Cybersecurity, Crowdstrike, Z scale. thing about Palo Alto by cyber arc. You start to see more and more consolidation. The consumer side, meta, alphabet.

And then it's my view ultimately where does this go with like physical AI, robotics, autonomous, Tesla, nuclear, names like Aklo. So that's my view. My view is like you can think about the theme in a very like tunneled vision. You have to now think about in terms of the second third, fourth derivatives playing out. So when you look at these, is there anything that possibly derail spending? Because it looks like everything's going well, right? Everyone has a lot of money. They're pouring into these data centers.

Is there anything along like a power issue or supply chain or policy? Anything that you're looking for that we should would raise a red flag when you're evaluating these things? No. Like down the road, call it 6 to 8x more power given the amount of data centers out there. That's a restricting factor. But if you'd think about where we are, you're 3, 4 percent in terms of through. That doesn't really, till we're about 20% through where that's a restriction.

It's probably not until 2008. So I'm just trying to give you like some like maybe like goalposts or time lines. Look, the biggest risk to this whole thing was US China. Trump tariffs, cut tech off at the knees, chips selling into Nvidia. But obviously, look, we've spent a lot of time in D.C. meeting with, you know, lawmakers and people on the, like, cooler ads prevailed, right? Like, they recognize, like, okay, we cannot cut tech off of these. That gives China advance.

For the first time in 30 years, U.S. is ahead of China when it comes to tech. So that's why you're starting to see a S&J barter system, right? Invidium-D, you want to go in there? Pay us 15% fine. It's a tax. It's a toll on the G.W bridge. So to me, it's like, okay, there's geopolitical, there's supply issues. There's power issues.

But the biggest thing is if the use cases work, this story plays out. And if I think about where we were, like being early in AI, I can do that on 22 relative to where we were in 2025, like we are probably a year to a year and a half ahead where I thought we'd be. While we're talking about this CAPEX, so $2 trillion CAPEX, where's the CAPEX coming from? Is it primarily just strength of these like company, big tech company balance sheets so far? or some of it debt at this point? Because that starts to, I guess, trigger my meter

for how far we are into a CAPEX super cycle. It's just like when we start needing more and more debt, let's say, and going up the risk curve from a debt perspective. But so far, is it all retained earnings, basically? All retained earnings. Really? So it's these just large tech companies who've made billion selling ads in cloud and everything else,

and they have that on their balance sheet, and now they're just spending it. Spending it? I mean, see, that's the biggest difference. Like when it, you know,

if you ever like, you know, with a bunch of friends, like, oh, there's like 99. Yeah, look at Cisco.

They're like, oh, like, look at Cisco's chart. Compared to Nvidia and it like looks like a same way. Look what happens. Yeah,

but there's one thing you're missing. Like Silicon Valley essentially didn't exist then. And when you think about capbacks and everything that we're seeing, it's, it's new rules of the road, right, it's new rules of the game because big tech is essentially running it, but here's what's crazy. Sovereigns haven't even gotten involved.

Middle East, they're just starting. Enterprises just start. The consumer piece, that's just the robotics still in the lab. Autonomous just starting to get, whatever, Waymo in five cities, robotox, obviously now, Austin and, you know, some other city. So I'm just trying to show you, like, where we are. in this. Help me navigate the AI stack here a little bit further, Dan.

So say I'm kind of new to this. I'm newer to AI investing than Josh is, right? So you mentioned kind of semis at the bottom, right? And this is, to me, maybe this is the compute layer. As I look at what you're holding in your ETF, things like Nvidia, AMD, TSM, even Broadcom, which is somewhat interesting to me. That's one layer. And then there's maybe the data center layer.

Like we have to build out a whole bunch of data center. Yeah. Think about it like foundation of a house. Okay. The foundation of everything are chips. Okay. So just think about foundation house. Because now we built it, those are the chips.

As the concrete's getting filled in and you're starting to build the actual structure of a house, that's the hypers. Microsoft, AWS, Google, Oracle. Then when you're thinking about use cases, software, that's the actual building of a house, the rooms, the furniture. Protecting it, that's a ring, that's an alarm. That's what cyber security is. See, I'm trying to explain to you.

But here's, think about it like this. You're essentially building Vegas or Dubai, and it's desert, and you're just starting like 1914. 45-50 style to where we were when you think about build out Vegas. That's what AI is. Interesting, Vegas, you're using the casino analogy here, which is, you know, the fast money analogy. But let's continue with this kind of house analogy.

And for maybe, you know, two innings in that were like 15% of the way there, 20% of the way there, something like that. So 80% remaining. I think it's really more 5%, 10%. 5% to 10%. Okay, all right. So we've got semis and compute, NVIDIA chips. that's kind of the foundation of the house then, right?

And I guess the framing of the house. I'll have to get some construction people on here because I don't know my house part. But the framing, okay, of the house, that's like you're talking about the hyperscalers. All right. So we got the Google. The storage.

Okay. And storage, all of that. And then we have kind of the app layer, which I guess is like things like Palantier. I'm referring to all of these things publicly traded companies. These are great. That's what's in your ETF, okay? And then where do data centers fit?

Is that like the attic space or something? Or is that what they, you know? Yeah, I mean, but when you think about the hyperscalors, okay, and you think about remember they have different, they have like different functional areas. They're storage, they're computing, they're the application layer. Right. They build out the clap. So when you think about like part of how Microsoft and Nadella will be on the Mount Rushmore

of CEOs. the next 200 years. Because Nadella recognized, okay, we're Windows, we're Enterprise. Going back to like when he, 2014, you know, when Nadella came in 2015, he recognized all about cloud. So when he built out the cloud, you know, Amazon and obviously Bezos was ahead in terms of AWS. But here's the key.

They built out cloud, built out cloud, built out cloud, built that cloud. When AI came, everything has to be in the cloud. Yeah. So that was a further catalyst for these companies, where it's not just AI-driven workloads. It accelerates the cloud. It accelerates companies going from on-prem to cloud. So that's why these companies have almost had an accelerated growth,

like a Microsoft, like Google Cloud, like look at Oracle. Like, if you look at that stock, that's a good example. Okay, I get it. So most of the time the hyperscalers then, they have a ton in terms of data center and, you know, compute resources, I suppose. And there's not generally, like, independent companies here. So in each of these kind of layers of the house or the parts of the house, I would just say one thing.

Remember, a lot of the data centers, those are data centers all around the United States, all around the world. Right. Where they're basically physical data centers, reads, equine, you know, equinole. And there's different, like, players that play in the data center space. And those, look, there's more data centers under construction today. Sure. than active data centers. That's incredible.

That does remind me of the 1990s, actually, when we were building it all that. But the deal with the 1990s, but the deal with 1990s, this is a 1995, 1996 moment. Right. Not a 1999 moment.

And then you could be like, oh, in three years it's going to be in 1999 moment. Look, my view is, it's totally different because the system's not leverage. The business models are there, and it's being supported, not by debt, but by the, you know, essentially trillions

is going to be generated, not just big tech, but across the board. In each of these layers, do you see a kind of a breakout network effect winner, you know, sort of like Nvidia has sort of won the semis? And they haven't won, I guess, you know, there's TSMC. Like Palantir has clearly been that one on the software.

Okay. I think Microsoft's been the one on the hyperscalor. I think on the consumer side, it's been meta, because of the way they're going to monetize their, that user base, the $3.5 billion, you know, that, like, meta is going to be the one on the consumer side. But we're just starting, it's like, okay,

who's going to be winner on autonomy? Like, it's my view, like, autonomous and robotics, Tesla is going to be the queer winner. It's a whole part of our thesis, right? Like, with Tesla, like, it's very easy. When you go get Tesla, like, oh, it's an EV, it's a car. I've never viewed like that, and I view the holy growl from Musk and Tesla, which why Musk's now wartime CEO, to focus on autonomous and robotics.

Is it the case that there are these power law winners in each of these categories from some kind of return to economy to scale? Or actually, you know, we had Kathy Wood on the podcast not too long ago, and she talked about the notion of rights law, which is basically the more of the thing you make, the cheaper they get to make, and you kind of build a network effect this way. And if you're saying, look, Tesla for robotics, maybe it's because they make the most robots effectively. That's why they win. And video.

And scale and cost. Yeah. So is it the case that for each of these categories, we should expect some like really dominant players and returns to scale? But it's not a zero-sum game. Now, at one point in the next five, seven years, you'll have more and more competition. But that's why it's my view. Like, you're going to have winners in China, Baba, Baidu, and Biden, and. and others, like, this is not one of those,

it's a zero-s-s-one, it's one or the other. Because the market opportunity is so massive, so early to where we are. And that's why it's my view that, like, we're going to be telling my NASDAQ, $25,000, $30,000 coming years. But again, like, for something myself has done 25 years, like, you know how many investors I've known

that have missed every transformational growth stock because they're focused on evaluation Is they're too expensive? They're always too expensive? At the time it's like, well, the Greek crisis in 2000 No, but the geopolitical No, but the Fed But like there's always an excuse right

So the point is like That's you know I think we're most well known For just seeing like more disruptive themes Picking the winners And we don't focus We don't get so bogged down with the valuation of the next year, because I think with transformational growth trends,

you can't do that. So if I'm looking at AI and I was looking through kind of the stocks in your ETF, like the 30 ETS, right? One thing that struck me, too, is it's primarily American, right? These are primarily... Yeah, just a few Chinese players, yeah. Okay, so is that because you think America is going to kind of win or because it's just too hard to access China's capital markets?

This fourth revolution, where do you think it happens? Is there a pretty, like the internet primarily happened in America. Do you think AI primarily happens in one geography? First time in 30 years. Like I could tell you, someone that's like always like, you know, in Asia four times a year. And all the time I'd have those trips. Then, you know, I land in Newark Airport. And I'm like, oh, we're so behind.

Like, okay. Guess what? The last few years we're ahead. See, it's the first time. in my view, since mid-90s that U.S. is ahead of China when it comes to attack because of AI, NVIDIA. China is going to have queer winners,

Baidu, Bab, you'll have other one 10-cent. I mean, you're going to have winners there, but it's because it's the one-chip fuel in the AI revolution is Invidian. And you're not selling, when you think about Blackwell and next-gen chips, you'll have scaled-down restricted chips you sell it in China,

but you're not going to sell the best chips to Chinese companies. You sell it to U.S. companies. And that's why it also goes back to when deep seek happened, you know, right away, like, over that weekend, we're like, dude, there's a better chance of me playing an NBA than them spend $6 million. Like add two commas.

The chance that they didn't use Nvidia chips, again, same chance like I'm playing Ryder Cup, September and Beth Page. So it comes down, like there's a lot of like, there was moments like, oh my God, this liquid ended up happening, right? Like they essentially had Nvidia chips and you had two commas to what they really were spending. But that's why Trump administration understands the biggest asset they have when it comes to U.S. China. It's got by Argentina by Argentina. Okay, so we have, we're investing in the United States.

We're investing in these categories. We have the semis. We have data centers, hypers, hypers, hypers. curious, I mean, for the people who are listening, is there a section of this house that we've just built that you see the most potential in that is the most exciting, that has the most upside that maybe most people aren't paying attention to? Is it in the foundational, like, data center layer? Is it the hyperscalers like Palantir who's had tremendous returns in the past couple of quarters? Is there a place that kind of is asymmetric in expected returns versus the

others? I think the area, like when you said where people are missing, it's not really chips. And it's not even cyber secure. To me, it's software. Like what ultimately this is going to do to Salesforce, Adobe, Microsoft, Alphabet, in my view of Amazon and some of the parts, you know, specifically on AWS, the MongoDBs, the elastics. Like, you know, it was like, this is, this is going to be, like, look, for every dollar spent on Nvidia chip, we estimate there's a $10 multiplier across the rest stack.

But it all happens in the software layer. That's the golden goose. In the software layer, how do you identify, like, who's executing against this? Because it strikes me that some companies will. Some companies will probably miss it. Have you identified that yet? And what do you look for if you haven't yet? Yeah, I mean, being like, you know, exposing, like, identifying Palantir, like, so early, right?

That was a good example. Where it was, like, we identified it and we saw, it just took time. play out. Yeah. That's what we spend all over time doing, right? We're in the field, like, three million air miles lost 25 years, right? Like, not from my house.

I mean, the point is, like, it's, that's been our secret sauce, trying to find who the winners are. But no different, like, who are going to be the winners in, like, robotics, atomic, physical AI. I mean, I believe, like, the amount that's going to be spent in those areas

could potentially dwarf what we've already seen today. I think it's interesting because it's also going to be difficult to sort the signal from the noise because, I mean, I'm sure this has already happened. I don't listen to many quarterly earnings calls, but I'm sure every CEO in the country is talking about their AI strategy and like what they're doing and using more and more buzzwords. And so the software layer, you have to really separate the substance from the noise. Yeah, and that's actually, that's probably the most, like, difficult. thing because there's like so much noise just because you say AI 40 times in a conference call

doesn't mean your AI. You know, so like, and that's, that's, you're right. Like, it's a challenge because there's so many companies like I'll meet with. And management seems like we, we are reconverting into AI. And like, I'll sit down with them. And after an hour, I'm like, you speak in Mandarin? Like, I don't understand. Like, I don't understand what you're doing.

No, but the AI is doing, and I'm like, oh, it's really your core technology that you basically have reconverted with some change in code to AI, but AI is really 5% of the code. 95% of it's a way you see. It's not an AI play. So I'm just trying to explain. That's why we've spent so much of our time trying to identify the next steps. Back to these areas. So America versus China, you think America has the lead. Of course, we haven't quite

seen robotics and maybe what China will do there. That could be another area. But like when it comes to, I guess, Chinese AI exposure, is it even possible to get this sort of exposure that you'd ideally want in in U.S. capital markets? Or is that just like not available to U.S. investors? No, you can't because like, look, it's like Huawei has like a massive, you know, scale product off, right? I mean, you're going to be able to like put, see, I almost view it as like, no, like there's a decoupled, but I actually do think it's important

to play China because I don't view it is like, it's usher. Like, it's my view. Like, you also have to play that because that's exposing, not just China, but you're exposing to the rest of Asia and a lot of the other areas. So, but it's in arms,

and the good thing in arms race is, a lot of money spent, a lot of winners. Now, to your point, Like, okay, are we going to go through a point or there's a capax digestion? We could go to do points into the next year that you have capax digestion. But again, it goes back there. I don't view that as like a multi-quarter digest. It's like you're just going to go through some of these pockets where investors will be

hypersensitive. Are they lowering growth? Is it over? But it's my viewing. We're just these companies have to continue. to put pedal to the metal. You go back, like, and like going in, the question is like, oh, my God, is Nadella going to recommit to $80 billion that they said that they're going to do?

And everyone's like, and then Nadella's like, I'm good for my $80 billion. Mm-hmm. Okay, it's seven months later they're doing $120 billion run rate. So just go back. I mean, because they understand you take foot off pedal, like, what can intel? that didn't exact work out great. Another way that I'm trying to understand this market is like where the primary value accrues, is it all going to be weighted towards the large cap side of things or is there some room

for small cap plays? Because what I've been astounded in this most recent kind of like run, the AI stock market run is how lopsided it's been in favor of like large tech company incumbents. I mean, the returns on the S&P, the NASDAQ for that matter, are heavily weighted. towards like, I don't know, the top five companies. Sure. And that's like, that feels different from a market structure perspective than other things I've seen in the past.

But do you think there's a small cap play? Maybe they make a comeback? I'm sure. Because big tech, their first ones to benefit. Then the ripple effect is going to be small midcap. But there's so many names that we've identified. I mean, there's so many small cap names. Sound town, like stocks like $2.

Like, now it's 15. Like, that's the example of like a small cap name on like the AI speech side's benefit. We just got a small benefit. Like names like Pegasus. I'm just giving an example. Names like Inodato. Like there's a lot of small cap plays.

There's a lot of small cap plays that are going to be midcaps. A lot of mid caps are going to be large caps. So if you're talking to someone like me, let's say, and I'm just kind of like I started the question of like, hey, is it still possible to get rich on AI, right? So what are the plays now that you think are undervalued, right? Because it does give me some pause to sort of, if I have 100% of an allocation, I don't necessarily want to pour it all into that Nvidia stock that's already kind of, you know, gone

to the moon. I will put a portion there. Okay. I'll put a portion there because that's the, not the, that's more than the foundation of the house. That's kind of like everything. That's the basement. That's like the core piece.

So I get that. But I'm also looking for the undervalued play. Do you think I should look more at sort of midcap? small cap type equities for that. And it's like, is there an easy button there or do I just have to do the analyst research? I mean, look, that's why we created the Ives A-I-30. Like, those are the names at all.

I don't know there could be names. Like, right now it's rotating, like where I think software and cybersecurity and the autonomous piece are the ones where maybe there's more pedal to the metal relative to, you know, evaluation, upside. But it's what, see, I view,

it less about small cap, midcap, and more about sectional software, cyber security, the buckets the way that we break it out. Tell me more about the Ives AI-30 then, because I was looking at the stocks in there. There's most, many I recognize, some I didn't, but how do you guys kind of like adjust what's in that basket? Yeah, and every quarter we adjust, you know, we just launched a few months ago in terms of based and the I've said I have third of research. And it's really, look, it's giving investors the ability,

okay, how do I play AI? How do I play AI? It's not just these names. Like, okay, you got to focus on a second, third, fourth derivative. It's us putting it into the buckets the way that we view, whether it's enterprise software, cybersecurity, consumer, autonomous, semis. And that is something where every quarter, if we feel like there's names that are going to be more relevant

based on our research and others that are less, then on a quarterly basis, like some come in, some come out. And that's the way the ETF is based off of our research. So that's a way

to easy button play it, I suppose, for a retail investor. Can I also just buy the S&P? Can I just buy NASDAQ as an indecy as well? Of course, you can always buy it.

I just, it's my own view. The whole reason I create the I've Zay out there is that this is giving true AI exposure. You know, the problem is like in so many of these others,

it's like, they're not, there's pieces of it, but this is more contrary to what I'd be used, the AI theme, true AI, A to Z. And, you know, and look, and I think it's the biggest tech theme we've seen in the last 40, 50 years. It's a fourth industrial revolution,

but it's not a one or two year theme. Is there a point, like, at what point would you look at kind of the market and say, okay, okay, at this point, I think AI is overvalued for the point in time. Like, of course, long term, it could be the fourth industrial revolution.

But are there some metrics or numbers in the back of your mind where you're like, hey, if this happens, maybe things are getting a little frothy at this point in time. Look, a lot of it's based on, like, when we do, if I see ideal activity is just continuing to massively increase.

Conversion, ROI, use cases went from 10 to 80 to what could be one, exponential, then to me, we'll continue to look out next two, three, four years. And if we could justify the valuation, and I believe the street is always underestimating. Like, that's always been my view as tech analyst for decades. It's like street underest, maybe in the near term it's fine, but it underestimates transformational growth themes.

I agree. Because it gets caught up and she had political fed, this, valuations. you know, like typical themes. And I think that has created the opportunities, right? The other monkey wrench, it seems, could be like the use cases. For whatever reason, kind of the chatbot intelligence, LLMs, that doesn't matriculate, that doesn't kind of get into the app layer and create economic value. I think I just saw, you know, OpenAI hit, what, 10 billion in annualized revenue,

which is like a pretty good side. I mean, that would, that basically means they've got 10 billion. and people paying them $10 billion just for access to their model. That's a good sign. But there could come a point where we just can't bake the intelligence in our systems and automate, reduce costs, you know, like generate new revenue sources. I mean, is a fundamental of driver here like how smart these models actually are and how far they distribute in our economy?

And how are you watching for that? Yeah, and the enterprise piece, that's been, that box, I think, essentially been checked, right? Like you're seeing that with hypers, the Oracle, and the enterprise is booming. And remember, the AI revolution is really enterprise that's driving it. The consumer piece hasn't started. So when we go to the consumer piece, 26, 27, 28, that's going to be the key in terms of on the consumer side.

But that's why use cases are key. It's key to understand where this is all heading. and it goes back to like when we talked about in 2022, like we viewed as like this is a theme we're going to see till 2030 and beyond. But we're only early on because it's just the enterprise. So I've went through the allocation of these 30 and I kind of looked at a lot of the stocks that you had and some of them were really exciting. I want to talk about the software because you mentioned a bit earlier a lot of the value accrued that software layer, particularly around Palantir, who I guess would be the poster child of this.

I was looking at the chart. It's unbelievable. It's up 500%, over 500% just in the last year. The last earnings report was incredible. I'd say half of the people invested in it probably still don't even know what it does. But Webbush has said, I mean, Palantir is going to hit a $1 trillion market cap. Wait, what is it now, Josh? Dan, do you have a, do you know?

It's like 350. It's a long way to go. Basically, to deferred to hit a trillion dollars. It needs to continue this type of growth. Look, it's my view. What are you looking for? First of all, I'm the hugest believer in fan

of carp. Like, I believe carp. I put them up there with Musk, Jensen, you know, Nadella, just, you know, hook to somebody, in terms of like seeing the vision.

Well, Palantir has a mousetraub than no one else has. I mean, remember, it started off as government. Basically, you know, they really foundationally or essentially the big data platform

from most Western governments, military, a lot of three-letter agencies, and it was taking that technology from a data perspective on the enterprise, and basically like doing what they did in government to the enterprise. But carp from the beginning, and Palantirians and Palantir,

they understood AI wasn't going to me about the LMs, about data. See, they made very strategic decisions early on. So they were so ahead of the trend that when companies basically started to buy Nvidia chips and wanted to build out AI strategies, there was essentially like one company to call. It's Palantir.

Now, I get the whole, like, so expensive. But again, it goes back to, like, I could have said the same thing when Facebook bought Instagram, and everyone's like, how is you, that acquisition makes no sense. That's been a hundred bagger.

It's my view. You have to be able to look out the next five, ten years where these companies are going. I'm looking at the numbers here. So the market cap, just over $400 billion, trading at 600 times price to earnings ratio, which is outrageous. But again, that's, but see, that's my view. It's like, you know, so many times, like, I've had companies, I'm like, well, it's true. Then all of a sudden, like, it's a year later, and you're like, huh, free cash flow is $2 billion. It's now $8 billion.

What's free casher going to be next year? $12 billion. Then all of a sudden, the company's not trading a rev, train of free cash flow. Like that switch flips. So what milestones are you looking for along the way that signal we're on our way? So we're at about $400 billion. We're going to a trillion.

Is there anything in particular you're looking for along the way to kind of check marks? The biggest thing that I look for, Pallantir, it's customers we talk to. What do deployments look like? like, what are the use cases, as deal sizes continue to increase, what the commercial U.S. business could ultimately look like over the coming years.

The government, I mean, like, the U.S. government, like, when they go toward AI, the red phone is essentially Pallentere Nvidia and Microsoft Oracle. Okay, there's another company on this list here that feels a little out of place, although it shouldn't. It's Apple.

Apple is, I mean, very clearly missing on the AI front. I saw it recently. They've spent about a trillion dollars in buying back shares. And it doesn't seem like they're putting nearly enough money in R&D and AI development. And yet it's still on your list. So I see you shaking your head. Why is Apple there? Because they're on the list because, in my view, the consumer AI revolution in 1.5 billion iPhones,

2.4 billion iOS devices comes through Cupertino. They're a toll collector on the highway. But probably the most disappointed I've been at any point in covering Apple over the 15, 20 years, probably now because they're missing the AI revolution. Innovation's not going to happen internally. You have wartime CEOs across tech, Zuckerberg,

must, Nadella, Jensen with a black leather jacket. It's an F-1 race in Monson, cooks watching it on the park bench, drinking a cappuccina. So it's my view, whether it's perplexity, whether it's doubling down the Google Gemini partnership, bringing new leadership in, that's what they need to do. But look, I think they're going to do something significant. One other area to kind of look at is private markets. I think there have been a lot of people, retail investors, myself included, who have been worried that some of the biggest gains that we've seen, at least from tech, you know, maybe since our business,

Reins-Oxley, I don't know, people playing various things. It's a blast in the past. It's the private side, though, right? The returns going to the private side and then companies waiting to go IPO and go public later and later. That's a big worry I have with respect to AI. It's just like, well, I mean, if the value doesn't accrue to these public companies, it has so far, what if it goes into private?

And most retail investors don't have access to that. Is that something you're monitoring? Is that something you? Yeah. It's been an issue, right? Backlog, you know, of IPUs, a lot of the most successful companies, open AI, anthropic, perplexity, you know, they're private companies don't need to tap public market. But look, ultimately, you know, with circle and you've seen other, I mean, public is the path. Like, I think you're going to see more companies forced into the public market just because I think brand, brand, currency, value.

And then there's a lot of companies where, like, they had down rounds after down rounds, after down rounds, like, okay, rip the band it off, get a public valuation, here is the private, here is the public, do it, prove it, and then you'll get. So then there's some of that that definitely needs to go on. Dan, I want to give you kind of a crossover section. So the Liminalist podcast where we do a lot of our frontier tech and AI investing, and we've got another podcast called Bankless where we cover crypto all the time. We recently, We had Tom Lee on the podcast. So you probably know he's hugely bullish in many of the same things you're bullish on.

He's also started this Ethereum Treasury company called BMNR. Basically part of his thesis there is that stable coins are going to be a very big deal. We have a new money system. It's a smart contract driven. It's programmable money, all of these things. And you dot, dot, dot into the future a little bit. And you have a world where you have LLMs and AI agents. The question is, what money,

system are they going to use? Do you think an LLM and AI agent can walk into a bank and get an account? No, they're going to use crypto systems. I guess my question is, do you see some signs of convergence in there? I noticed there's no crypto stocks in your portfolio, but there could be. And what do you think about that? Look, first of all, like, Salar, Tom, you know, Scarmucci, I mean, some of the people have known for, you know,

basically my whole career, like, I agree 100% with their views, right? In other words, like, crypto and disruptive tech and where everything's heading, it actually aligns with the broader thesis in terms of AI revolution. AI revolution, when you think about where it's going to go, it's not getting there without crypto, because ultimately, there's a convergence that's going to happen here. So I'm very in line with that view, and I think more investors are starting to piece it all together, you know, in terms of crypto, there's Bitcoin Ethereum and some others with what's happening in AI.

Very cool. Well, we're definitely very excited about that convergence. Dan, as we end, are you ready for a quick lightning round? Yeah, let's do it. All right. You can only hold one AI stock, which is it? And why? Nvidia, godfather of AI.

Only one chip in the world fueling AI. What's the best AI model right now? Oh, I believe perplexively. Perplexity, interesting. Perplexity is probably the one that's probably the, like, I think the underdog, chat GBT clearly being the leader. But I want to put perplexity in there

because chatGBT is the leader, but perplexity, I think, is the one under the radar. Best AI leader CEO specifically for war mode? Cart, Poundeer. Carp, okay. Over Elon Musk? Look, Musk. clearly, obviously, he's a wartime CEO,

but I'm just saying if you look at what Carp's done with AI, with Palantir, you've got to give him a little nod over Musk for now. Most undervalued, underrated AI asset. I believe it's hyper-scalers. I think that's the one from Microsoft to Google to Amazon. I think those are the ones that's very undervalued in terms of everything they were doing. I could have said Oracle put them in the same category. Okay, so maybe I want to end this episode with where we started,

which I asked you the question, is it still possible to get rich on AI? You said yes. Now my question, I guess, is, so how rich are we talking? All right? Like, how much do you think

this will continue to run? I don't know what the size is now for all of AI and the kind of the stocks that you put in that category. But tell me about the size now and where you think this could get in the next, you know,

three to five years. Look, it's the biggest tech trend in the last 40, 50 years. And to our view, and this AI party, it was 9 p.m. It's now 10 p.m. That party goes to 4 a.m.

Everyone's waiting behind the velvet ropes. Who's getting onto a dance for a jensen car that, you know, obviously, you know, others are waiting. I would much rather be a bull in that party than a bear outside the party looking through the windows. When you meet up at 6 a.m. at the diner, bulls had a lot better night than the bears. It's no fun being a bear. Dan Ives. Thank you so much for joining us today. It's been a pleasure. Thanks for joining.