저자: DanIves · 외국인
요약 보기
Dan Ives는 AI를 4차 산업혁명으로 규정하고 현재를 '2회 말~3회 초' 수준의 초기 국면으로 봄. 칩 수요가 공급을 크게 웃돌며 균형은 2028~2029년 무렵 가능하다고 언급. 데이터센터 정치적 규제가 최대 리스크이며 중국은 물리적 AI·로봇 분야에서 경쟁 중이라 평가. OpenAI·Anthropic 집중 리스크는 과거 우려이며 기업 수요 확대로 해소된다는 입장. 미국 기술 우위 복원과 일자리 우려에 대한 낙관적 견해 제시.
다듬은 글
We've got with us for the next half hour or so. Dan Ives, he's partner and senior managing director at Yorkville Ives. It's the first time I've read that title for you, Dan, because, you know, for so long we knew you at Wedbush and we knew you as this, what many people called it a tech permeable. I want to get to, uh, your outlook on not just this earnings season and what to expect from Nvidia and the so-called circular financing.
But before we do that, because we haven't had a chance to sit down with you since the since this job changed. I want to just hear from you why you left Wedbush and what you're doing now at Yorkville Ives. Yeah, I mean, for me. And it was a great, you know, eight years at Wedbush, but it was to create what I view as like a modern merchant bank, you know, really something where when I think about this fourth industrial revolution and for my, you know, card going in Wall Street since the late 90s, it was seen around corners and seen the opportunity.
And I just felt like for me, it was an evolution in my career for something that to build a bank and especially have great research that's focused on tech, energy, you know, and and really across industries. And I've found great partners in Yorkville. And to me, it was the right place, right time to do something like this.
Given my view of the I Revolution. Well, I want to get to more more on that. And we're going to get to the I Revolution. But we spoke to you, uh, at Future Press, where you'll be there again this year. It's September of last year. And I remember the conversation because it was it was not just about what you were doing at Wedbush, but it was about the ETF that you had launched.
You were also chairman at the time of World Coin, this this crypto project or ECA. Yeah. Oh yeah. Co thank you. Which is part of which was yeah which was brought to world okay. Related to world Coin. So what is like what is the status of those projects now. Yeah. So that so in terms of like I, I left ICO.
Yeah. But what about you. Yeah. Yeah yeah. And that like you know no there's been no change okay. Of that. Um, and that's something where for me, look, it was sort of like simplifying, you know, I think some of my, you know, other activities and really to me was really like doubling down to some extent on what people know me as in terms of when it comes to tech, when it comes to, you know, great research.
And I think, you know, I think around the world, so many more investors have such a focus in tech. And it's something where to try to be some guiding light, you know, relative to why the confusion that you see in the market, what can you do now that you want that you couldn't do before? I mean, there's, you know, we're we're still awaiting our research license.
Yeah. But for now, this is like when it's all sort of built, it will be something that it will be exactly the same. Dan Ives, uh, research. And you know what we've done for investors over the decades. And that's really why we build the but also it's for me, it's finding the right people that want to do things.
And when it comes to innovation, when it comes to building, whether it's on the research side or, you know, across the industry. And I just think right now, like we're going through an innovation boom. And in my career, it's really been an evolution for me. That was a natural next step. Yeah. And I've been super excited about the feedback that we've gotten.
And just, you know, from, uh, you know, from, you know, many around the world, we have seen a lot of innovation cycles and you have been along for that ride. And I'm just curious how you continue to put this particular cycle in its place and in its spot, and how much do we really know about the ultimate impact?
I mean, some like myself, that's in so much time in Asia. Yeah, so much time talking to CEOs, CIOs that's done this since the late 90s. This is the true fourth industrial revolution. And, you know, we've talked about it with you guys may, you know, for many years. And that's why I think a lot of times like the skeptics or the bears.
You know they they're bearish from the 15 for their New York City office building or wherever it may be. But they're not in fabs in Taiwan. They're not seeing the demand. And the reality is like for the first time in 30 years, the U.S. is ahead of China when it comes to tech. And I think that's extremely important relative to where we are.
And I think what you've seen with earnings, whether it's on the NIO clouds, the hyperscalers, that it's just it is crystal clear that demand position feeds is now starting to take over. But do we have to be worried about China because they certainly have the government now saying, okay, there they are prioritizing this and with money, with efforts with, you know, how it works.
We've seen it with cars, we've seen it with EV, we've seen it in so many different industries. And they want to make sure they have a place at the table with this. So how do you how far are they along? What do we really know about their expertise. Yeah. So I think as someone they measured that spend so much time in the region.
It's one like counting them out is the wrong bet. Because the reality is, is that not just from a government perspective, when it comes to energy, you know just how ahead they've been when it comes to nuclear energy, when it comes to robotics. And it's one where I think part of why Jensen is so focused on selling into China, and I think why the chip companies?
Because if you do not sell into China, it just makes him that much more powerful. Yeah. And narrows the gap. And also when it comes to like the political, whether it's grandstanding, whatever you want to call in terms of like what's happened here in the data centers and data centers are the hearts and lungs of the AI revolution.
So if you have like moratoriums and you shut down data centers, the winner in that becomes China relative to what could happen here in the US. So I actually think like the biggest sort of risk, when you think about what could happen to err is that CapEx. It's not use cases that the biggest risk is just the political piece gets in the way in terms of the data center build out, because I think that's something where any every data center that gets voted down.
Yeah. China smiles. I think we have time for this question. It's a really good one. It's from Brendan in Maryland. Yeah, yeah. He has a question for you. Says you've likened the I Revolution to building out the Vegas Strip in the 50s. Can you elaborate on this analogy? What role is Microsoft playing here?
Are the hyperscalers like the big casinos? And famously, does the house always win? Great great great great question. Look, and the reason I talk about like Vegas 1955, because if someone told you in 1955 would eventually the sphere was going to be in where Vegas. But then obviously in 1956, there was an issue with the building.
You got the strip. Things not going to happen. It's not going to work. It just that's big tech companies like that's their view. Right. Like go back to Mike. If you like. My stuff is good example like two months ago. Write a New York City cab driver is bearish on Microsoft and net. Now look today right there was ticker tape parades for it.
It just speaks like because you have like Nadella is a North Star and he's able to see around corners and understand that on the Vegas Strip there's only one strip. If you don't have a place in that strip, are you staying at that hotel two miles from Vegas? Eventually you're in Santa Fe or whatever I'm in.
The point is, I'm just trying to explain, like, these companies know there is one strip, and that's why we're saying, well, we're about cap X. If we're tech metal like stocks down, does that mean that they start that, you know, they like slow down their bill because they know that if they slow it down, they won't have a place in the strip.
And then what happens when you look out a few years from now and you're three miles off the strip? But do we need all of this on the strip is, I guess, what I'm wondering. There's a there's a second part of Brendan's question. When we come back, I want to ask, um, Dan, Danny's partner and senior managing director at Yorkville.
Ivy's joining us right here in our Bloomberg Interactive broker studio. Stick around, folks, more to come with dad in just a moment. We've got back with us and we are back with Dan Ives, partner and senior managing director at Yorkville Ives. He joins us here in the Bloomberg Interactive broker studio.
I want to get to the second part of Brendan's question. What a great question on two parts Bloomberg.com ask radio. That's how Brendan was able to ask the question. Bloomberg.com ask radio. This is for, uh, terminal subscribers, Bloomberg Terminal subscribers, and those who also are subscribers to Bloomberg.com.
So he went on, Dan, to say even more, some question whether this AI boom is due to chip sellers logging revenues faster than reporting their expenses. Can you speak to that? Yeah. So I view it as right now demand supply for chips. Yeah 1314 to 1 demand supply. Yeah. The one thing is that look I view the whole thing is kind of like a Jenga puzzle having when one think about it, you put it together, right?
Whether it's Cisco, Neo, Cloud hyperscalers, WhatsApp and pound tier on the. The reality is, is that they hold the cards, the chip players because you don't have many of them and you're not going to create new ones. And that's hook that's similar what you're seeing carrying the cost with the memory players.
And I just think it's one where when some try to kind of poke holes in some of these too. The reality is, and it's always been our view that like investors are way underestimating the scale and scope of the AI revolution. What would you say? And certainly we've heard this around this table of the semiconductor makers who are very sensitive to the cycles of the booms and busts, that they are going to be very careful also in controlling supply in the market.
So, yes, they're going to invest and they're going to ramp up, but they're not going to do it so much that all of a sudden the price drops. So So there's got to be a I don't know what the the factor is that in that excess demand 1314 times whatever you said for chips. How much of that though is the semiconductor companies being very smart and kind of managing the situation and keeping those margins healthy?
I think at one point you'll hit that, but for now, know the equilibrium you won't hit till probably 2028, maybe early 2029. Everybody keeps talking about 2030. I'm like, I'm going to have it tattooed on my forehead because it just feels like I want it. And that's why the reality is we've always thought I revolution like what we thought it was third inning.
And now, you could argue, is a bottom of the second. When you go to China, when you travel like I want to ask you about China, what do you actually get to see in terms of what's going on there? Because I just feel like there's so I don't think it's so easy, but I'm curious because when somebody does get a peek into some of what's going at China, I think it really surprises everybody.
It's the innovation. It's the it's really what they're doing on robotics, on what they're doing in chips, on what they're doing on when you do have physical eye. To me, that is the thing that stands out the most because I think the name in the game and I today, you could argue like some of this stuff, some almost table stakes relative to when you think good of physical AI and when you go to a tour, the next phase is here.
And that's why right now, like robotics, autonomous, true physical, I, I, I've always viewed physical AIS who are the golden goose of so agent ai and agents that that's not the gold I view. That is a highway to get on to ultimately what's going to be the part of the biggest modernization piece. And I think that's why, like a lot of times, I think investors, whether it's a China in terms of open source and different models that come and we go through these sort of like mini scale, deep cheek type movements.
The value is in the data. That's our sovereign data. You know, when you look at carbon pound here, they talk about so much Jenssen story and talk about why. But it's an arms race that's playing out across the board which speaks to used to like circular financing and just you know, we go through these sort of ebbs and flows and the worries there.
Well speaking of worries, some folks out there are worried about concentration risk. Uh, we spoke to Ed Citron a few weeks ago from Easy Primary Research. He was on our program. Here's what he said about concentration risk. So in calendar year 2025, according to my own reporting about OpenAI's numbers, 69% of the year over year growth of Microsoft Intelligent Cloud segment was actually from OpenAI.
Without that, it would have only grown 8% year over year, which is barely beating inflation. And so everyone is being sold. What I consider kind of a lie. It's honestly kind of a scandal. Those ads that you're on of easy primary research. After he was on, just a few days later, our Bloomberg News team reported that Microsoft generates most of its AI revenue from OpenAI.
It said that, uh, what about 70% of sales, uh, in a year came from, um, just a couple of different companies. Um, excuse me? Microsoft's business, 70% of its actual I sell came from OpenAI during its most recent fiscal year. Is there a concentration risk? But that would be like the Chiefs the if they lost Mahomes they're done.
The point is like the reality is like OpenAI and anthropic their core essential foundations, they are. No one will doubt that. But the reality is, is that when you continue to see the funding there, and what's happening is that the tech companies, whether it's the hyperscalers and others, yet they have so much invested in them.
But that's just the first phase, that number two years now, three years from now. The concentration risk is now meaning there will be many companies that are paying, of course, because now for these services today, it's a small group of companies that are paying. But when you think about what's happening, sovereign AI and today really being much more just us based in China, because what's happened, the Middle East and India and just across the world, we're still in the early parts of it.
But that's why OpenAI and anthropic, when you think it as they go public and those are there, that just further solidifies the foundation for the AI revolution. I'm going to play devil's advocate to you, because I think what someone had said too is that, okay, so most of the revenue comes from the AI revenue for Microsoft comes from OpenAI.
But help me out here. Um, I'm giving you a gift because I'm usually tougher. I like to battle with you, but the idea is OpenAI has a ton of customers, and so that's where the diversification is. but it comes through that conduit. So I guess essentially if Microsoft and OpenAI go to battle, that could be a problem.
But is that kind of is that the right reasoning? Sorry, Tim. Don't be. That's fine, because but to that point, is that the concentration risk, it's not just viewed as it gets OpenAI OpenAI too much, thousands of customer anthropic. Same thing. What they're doing the enterprise. It used to be a good idea.
I guess they would from makers to be a concentration risk. But Microsoft is the core part of the enterprises around the world. So Microsoft represents so many enterprises that are ultimately going. So I just I view I understand like and that's been yeah that that's been a concern. But 18 months ago it was a concern.
Look where we are today. And that's why I think those that bet against this, it's it's betting against a cap cycle that actually continues to accelerate. You know, it feels like there was this there was this moment in New York in like the mid 2000 teens where all these venture capital backed companies were flush with cash and Uber and Lyft were duking it out, giving you incentives to like, take this car, take this car and you as a consumer.
And there was a service that came and picked up a package at my home and shipped it off, and it was free. They went out of business. But the point is, is that venture capitalists were subsidizing this sort of on demand lifestyle. It kind of feels like sometimes we're there with these llms. Where's the like?
It's free to use clod. You know, then 20 minutes later, you're like, okay, well, I can't use cloud anymore. I'll just go use ChatGPT. Okay. You use up everything with ChatGPT. Oh, I'll go just, you know, go to Google and use the AI that Google offers. Like, it kind of feels like we're in that moment again where these are interchangeable, at least the llms.
But I've never view that. The models. Yeah. Uh, the models are not going to be where the that that's why anthropic you know OpenAI. They're already see they're seeing around the corner of their focus on enterprise building, enterprise sales force ultimately really butting heads with your core enterprise players.
It's going to be in the data. I mean, if you look where it's all heading, it is just going to be a conduit deal because eventually models, you're going to have hundreds of miles, your models regional vertical. It will get more and more into industry specific construction models, retail consumer like.
The point is and that's also where like if you think like where Apple is trying to go in the consumer side and they'll finally be, you know, a shorter player. But I just think it's one where where AI is today and where it's going to be 3 or 4 years from now. I think for for many investors, you get 1 or 2 investors.
You have to be if you're a technologist and maybe an incoming that tries your own corners up, you're going to view it as like, this is true forward in this revolution. If you're one of the bears that have called ten of the last two downturns, then your you'll say, this is another one and they'll be in hibernation mode.
And they can't see eye on spreadsheets. But you're smart, man. What's the risk? We have about a minute or so left. There's a risk to everything. The biggest risk. Don't tell me it's is government oversight. When you look like. When you look at time in New York state, local moratorium data centers data.
So that's the biggest midterm elections. The. It's not just New York. There are a lot of it around the country. And I spent so much time in D.C.. And the point is, like a lot of politicians that are still using blackberries, do you want them ultimately determining I and I think that's the scary thing, right.
Like it's one where it would stifles innovation that you need regulatory know it out. Yeah. And PR problem like that has been caused by the tech industry in terms of wiping out jobs, 18 months scare tactics, things like that. So they've definitely contribute, but I think that is the biggest risk. When you think about AI revolution relative to the political piece.
Do you think I though I mean, it's going to wipe out tons of people talk about universal universal basic income. I've talked to developers. We've talked to developers about thinking about building houses for people who are on like because of I forgive me only about 40s, but I'll just take the other side.
And so are you somebody myself aware of a 3.5 million air miles and been around this country so many times the amount of towns and cities, they used to have 75,000 people. Now it's 10,000 factories that went away to China. Yeah. And I view it is for the first time in 30 years, the U.S. has added China when it comes technologies to the innovation boom and to what I view is really a renaissance.
I view that is to add jobs in the United States, despite maybe some of the more negative news today. All right. So go get that research license and come back. I will, I can't wait.
Comments 0