It's year three of an 8-10 year buildout of the AI revolution. This is not going to be a straight line. They're making the bets just like OpenAI is making the bet, just like Nvidia has made, and that's the bet where we continue to believe there are two more years to this tech bull market given where we are in terms of the buildout.
All right, welcome to OK Computer. I am Dan Nathan. I am joined by my friend Dan Ives. He is the head of global technology research at Wedbush Securities. Dan, welcome back to the podcast.
It's always great being here with you.
First time you and I podcasted a few years ago under the Risk Reversal umbrella. You were on OK Computer, it went away last year. You did podcast with me a few times under Risk Reversal. So this is one of the first ones that we're doing again. So we brought back the whole brand. I want to say by demand. The folks liked it.
Look, I watched the episodes, and I think it's especially at this time so interesting and insightful to get the in-depth conversations that investors need to know.
You know, I don't get to sit in on the calls that you have with your clients and with the suppliers and the companies you cover. And I think that throughline is really important if you're an institutional investor. So what we're trying to do is a little bit of the same for some of the folks that are self-directed, retail, whatever you want to call them. I will say this, we've had a couple great conversations. The first OK Computer, Jim Lanzone, CEO of Yahoo, we spent a lot of time about the turnaround of that brand, of that company. You probably covered that company for a while. But we also talked about AI and how that's disrupting some of these legacy media and digital media.
And they're doing great things.
And they are. So stick around. And then we have Adam Singolda, another really good friend of mine. He's the CEO and founder of Taboola. The three of us are going to sit down and have a conversation about all sorts of things. I wanted to spend a few minutes with you because you and I have had a bunch of conversations over the last couple of years. The last couple of years have just been monsters as far as tech is concerned. You launched an ETF, and it is an AI revolution. It's IVES, and if you go and look at the holdings, it's a lot of names that you know, but the weightings that I find really interesting, and it's meant to capture the themes that you have about your research ratings. Tell me what it's been like. It's a billion dollars basically market cap. What's it been like running that fund in the midst of all the volatility in this space the last six months?
Yeah, it's been great. Performance has been really good. It's my view that you can't just own one name or seven names. The idiosyncratic risk is too high. My whole goal of the ETF is to have the 30 names, the Ives AI 30, that I believe are the best ways to play the AI theme. Four times a year we switch them out. We have a few that come in and a few that come out based on technologies that are succeeding and those that aren't. We took Adobe off and some others. It's been really rewarding because the whole goal of this is that a retail investor in Spain, an institutional investor in Australia or New York, anyone can own this. It's my way to present a vehicle based on three and a half million air miles over the last 25 years, so investors can understand what's happening and how to play the second, third, and fourth derivatives.
Yeah, and hopefully you'll come back in January because I'd love to talk about some themes for 2026 because I think they're going to be different than some of the themes of the last couple of years. You just mentioned energy and infrastructure, but enterprise software is one that's been left for dead. Adobe, as you just mentioned, has been massively disrupted over the last few years. That might end up being an interesting opportunity if they ever get their act together. We'll talk a bunch about that. Before we get to Adam, I want to hit some stuff that's going on. We're recording this Wednesday afternoon. By the time you're listening to this on Friday morning, we will already have had earnings from Micron. And when you think of high bandwidth memory, this is a name that's in your ETF. The stock has been an absolute monster this year. It's up 167% from its lows in April. It started out with the GPUs, then it got to Taiwan Semi, then the hyperscalers, then the server makers, and then the memory and storage joined the party. There are also contract manufacturers and the energy component. But right now it's kind of stalled. Broadcom, in the last week since they reported earnings, is down 23%. A company that guided up and beat, but it was low single digits. It wasn't good enough. Valuations became an issue, competition became an issue. This was trading at all-time highs, massively outperforming its peers because suddenly there's competition to Nvidia. Same thing with Google and TPUs. So give me a sense of the landscape right now because we see Microsoft is down about 15% from recent highs, Nvidia is down 15%, Amazon is down about 13-14%, Meta is down 15-16%, Broadcom is down 20%, Taiwan Semi is down at least 10%. You're starting to see these names underperform. Many of the Magnificent Seven have been underperforming the S&P and the NASDAQ. Apple and Google, which you are bullish on, have massively outperformed. Apple is benefiting from not having spent hundreds of billions on an AI buildup, but they also don't have a product. Give it to me.
That's a phenomenal summary. The reality is that anyone associated with OpenAI right now is viewed as a 'do not enter' zone. We'll get to Oracle, which is front and center. Then you look at Google. We started the year with bears in New York, but look at it now. Apple, we believe 2026, just like Tesla, is going to be a huge year. It's the year they actually come out with their AI strategy, and I think Google will be key to it. My view is it's year three of an 8-10 year buildout of the AI revolution. This is not going to be a straight line. We'll go through moments of worry about whether data centers are overbuilt or if OpenAI is too big to fail. Have companies gotten over their skis? The bears are coming out of hibernation mode. But my whole point is that when I'm in Asia and I see demand, the supply-to-demand ratio is 12 to 1 or 15 to 1. It's not a question of demand or capital or capex. It's a question of speed. Oracle has 450 to 500 billion in revenue they now have to fulfill. To me, they're making the bets just like OpenAI and Nvidia are. And that's the bet where we continue to believe in two more years of this tech bull market given where we are in the buildout.
Right. What if OpenAI doesn't have the money? This goes back to Stargate. We had Masayoshi Son from SoftBank, Larry Ellison from Oracle, and Sam Altman from OpenAI at the White House. The first thing that happened, which I thought was really important, is Elon Musk tweeted that they don't have the money. SoftBank is down 40% from its recent highs and was supposed to fund $30 billion of Stargate. To date, they've done $7.5 billion. Oracle got a $300 billion contract from OpenAI to build 5 or 6 gigawatts over five years. The market immediately rewarded them, but now the stock is down and the CDS has moved to 150 basis points. The chart is about to crisscross. Alarm bells should be going off. Brad Gerstner asked Sam Altman on his podcast, and Satya Nadella was on that podcast. Sam Altman looked up and to the right and said, 'Brad, if you want to sell your shares, I'll find you a buyer.' That was freaky.
Yeah, really since that podcast, if you look at the chart, the OpenAI-related names have gone down, while the non-OpenAI names have gone up. I went through it with Zuckerberg during the Cambridge Analytica era. This is the beginning of the $3 to $4 trillion that's going to be spent. The question is, will they have the money? Will they have to raise more capital? How will they show they can monetize what they're trying to do? That's why 2026 is a fork in the road moment. It's the monetization of AI. It's autonomous, what I believe will be the true start of autonomous in the US. It's Optimus, what we talk about with Tesla in terms of robotics. This is going to be the 'prove it' moment for the tech sector.
Yeah, we just haven't had that valley of despair yet. It's been fairly euphoric except for April and May. Now we're starting to see a lot of dispersion in the space. Some subsectors are having a tough time. If there's a debt bomb around all the debt being taken to fulfill OpenAI contracts, and OpenAI doesn't have a revenue generator other than people paying $20 a month and the API, what do you make of the fact that these could be commoditized? But 75 to 80% of the next $700 to $800 billion in buildout is being done from free cash flow. Companies are not taking on debt. It's free cash. Oracle is not a hyperscaler. They want to be, but they're funding it with debt. If someone said here's $100, you have to spend $15 to get it but take out $10 in debt, would you do it? Only if you believe the revenue is coming. Microsoft, Amazon, and Google are funding it out of cash flow, and they'll be fine. They have these cloud businesses. Going back 25 years, all that fiber that got laid was massive overcapacity for five to seven years, but it all got used. That could be the same case, but there will be companies that get absolutely drilled because they don't have the balance sheets or cash flow. Oracle has negative cash flow for the first time since 1992.
But I would rather Oracle do it. Their RPO and revenue have gone from $15-20 billion six quarters ago to $470 billion now. These are not numbers they're making up out of thin air.
Well, I think they are. I think OpenAI is making them up. He wants to build data centers in space and whatever. Maybe it happens, but can you think about the complexity? It's actually kind of easy to build them here. Just get trucks digging in the dirt and servers and routers.
There are more data centers under construction today than active data centers.
Right. I get all that. You and I believe the same stuff, but I think we have different time horizons. I'll take the over rather than folks looking to take the under.
But Nathan, a year from now, you'll be wearing a pink Peter Millar vest and you'll be saying, 'Look, I was wrong.'
I thought you were expanding your wardrobe. The sports jackets are coming out soon. But the point is, it needs to prove itself. The shorts will win until they prove it out.
You and I will do this in January because I want to think about your 2026 outlook. I really appreciate you coming here and taking the tough questions.
I enjoy doing it because for investors, lovers, haters, however I talk here is the same way I talk at a coffee shop or in an airport. That's truly how I believe.
Right. So you're doing a lot. You've done a podcast earlier in the year, the clothing line, the ETF. Some folks might think it's getting a little toppy, but you joined a treasury company. It's an operating company. You've explained it on the podcast as the chairman of EcoOrbs. I want you to describe what that business is again, and then talk about the DAT landscape. Companies that don't have a strong operating business are coming undone. So give the 411 on EcoOrbs again.
The 411 is that I'm not joining a treasury company just to buy coins. That's the public vehicle to start off. Our whole goal is that Sam Altman is a big backer of Worldcoin, which are the orbs. There are 18 million people who have done eye scanning or retina scanning. It's all about verification and authentication. It's about consumers doing it, but my belief is that we are essentially going to be the enterprise arm of Tools for Humanity and World. EcoOrbs is focused on Infinity by Orbs, our product. The buyers are enterprises. We have pilot partnerships with Kraken and Coinbase, and a few others we expect to announce. The goal is that we've bought World and Ethereum, and we have cash. We are much more of an enterprise authentication play and an AI infrastructure play. That layer is going to be key to the AI revolution.
Right. Explain why you need to own Worldcoin. What does that do for the orbs and the ecosystem? You just said you own Ethereum as well.
Worldcoin is the crypto. Everything is built on that infrastructure. When you sign up, you get paid in World, and you get a wallet.
So this is kind of like Helium. I had a Helium router five years ago. You join a decentralized wireless network and get Helium for mining capacity.
But the big thing is that there are real use cases being deployed with Tinder and advertising companies in Japan. In an AI era, regular verification is not going to be satisfactory. I get the glass-half-empty view of DATs, but as we've talked about, we are a real business and much more of an AI infrastructure play.
We all know this because we went through the ICOs, NFTs, and Web3. There will be plenty of Johnny-come-lately frauds that tell a story but have no operating business. One example is MicroStrategy and Michael Saylor. He's out there talking to anyone, getting companies to put crypto on their balance sheet and become a DAT. I've known Saylor personally. He's a nice guy and tries to be transparent, but the market is calling his bluff. He owns 670,000 Bitcoins. He sold stock, debt, and perpetual preferred to buy Bitcoin. His average price is $75,000 on 670,000 coins, and the stock is trading at a discount. They might have to sell Bitcoin. Doesn't this have the potential to blow up the whole strategy?
I strongly disagree. My view is that what Saylor is doing with Bitcoin through Strategy, and what Tom is doing with Ethereum, I am a big believer in Bitcoin and Ethereum.
But if you're a believer, why not just buy Ethereum? Why put it on top of a business?
It speaks to what they can do in terms of staking and the use cases for Ethereum. World and what we do at EcoOrbs is based on Ethereum.
Built on smart contracts and the like. I remember the promise of smart contracts, DeFi, and Web3, but here we are three years later and not many people are interacting with it. The biggest use case was NFTs, which made Ethereum so popular, and now they are all zeros. You can buy Bored Apes for a fraction of what they were worth.
There will be companies that go by the wayside, of course. But what we are doing and what Tom is doing with Saylor has created the industry. I am a believer, but I get the pushback.
