Herzlich willkommen zu einem Sonder-Podcast, featuret bei Handelsblatt. Die KI-Rally steht vor einem Realitätstest. Ich spreche mit Dan Niles, einem bekannten Wall-Street-Tech-Strategen und Gründer von Niles Investment Management. Er warnte am 20. Juni vor einer Korrektur im Tech-Sektor und lag damit richtig. Warum er dennoch an der langfristigen KI-Story festhält, welche Risiken durch sinkende Tokenkosten chinesische Open-Source-Modelle und teurere Halbleiter entstehen und ob KI-Modelle zur Commodity werden, während die Infrastruktur knapp bleibt. Darüber sprechen wir in dieser Folge. Wer Untertitel auf Deutsch haben möchte, die findet ihr bei diesem Beitrag auf YouTube.

Tja, wie so oft freut es mich, einen Dauergast bei uns zu begrüßen, Dan Niles von Niles Investment Management. Am 20. Juni hat er auf Twitter gewarnt vor einem AI-Speedbump, einem Schlagloch bei den AI-Werten. Das sehen wir seitdem auch mit einem recht scharfen Rücklauf bei vielen dieser Überflieger des Jahres 2026. Ich möchte heute mit ihm über seine drei Kernthesen sprechen. Einmal das Thema Token Minimization, dann den wachsenden Wettbewerb durch Low-Cost-LLM-Modelle, Stichwort China und zu guter Letzt auch die steigenden Halbleiterkosten, die wir ja in dieser Woche auch infolge einer Ertragswarnung gesehen haben bei einem kleinen Tech-Unternehmen bei Calix. Dan, we've known each other for quite a while. I know your German is impeccable at this point.

Great to have you here. Before we dive into your three main concerns, let's have a quick look at the market. You recently this week posted on Twitter and you also looked at the Morgan Stanley Momentum Index oder in diesem Fall, der Goldman Sachs High Beta Momentum Index. Sie sind beide ziemlich die gleiche Geschichte. Sie sehen einen dramatischen Dramatischen Dramatischen in Momentum stocks. Generally, der Goldman Sachs Momentum Index ist um 33%. Looking at die Geschichte, da sind die häufig levels, wo man starten zu erwarten, eine Bounce in der Sektor zu erwarten. Wie ist das? Yeah, so I put out a Twitter post this last Sunday, basically talking about that, where if you look at the Morgan Stanley Momentum Index, and so that's you take the longs, momentum longs, you subtract the momentum shorts. Since I put out that post on June 20th, that momentum index from June 22nd through last Thursday was down 28%, and it happened over 24 days. If you look at history, whenever that momentum index has been down over 10%, you've had a drawdown.

The average is 20% only, this time it's down 28, and it's taken 48 days to do that, this time it's 24 days. So the velocity of this move is greater and it's happened in a shorter time than the 60 plus occurrences since 1996 when I went back and examined this. So you would think unless there's something structurally wrong, so you're going into a recession, the Fed's hiking rates, you have a global pandemic, you know, something that's really problematic, you should be pretty close to where you want to start adding back in exposure, in momentum names and specifically AI infrastructure related names that have gotten just absolutely clobbered, which they rightly should have in that they got overheated. But I think now is a period of time when you can start to do that. Obviously, the biggest risk is you're going into earnings season, coming up with Google on Wednesday night, and that's going to be a huge deal for everything AI-related in terms of what they have to say. We're going to dive into that in a minute Of course you also had a dramatic rise almost a parabolic rise in the SOX Philadelphia Semiconductor Index, Isn't that a bit of a risk regarding a potential bounce and how sustainable is such a bounce going to be Sure, I mean.

So, just if you look back at history and you say, okay, what happened during the internet infrastructure buildup? So, from the end of 94, which is when Netscape Navigator was introduced to its peak in March of 2000, the semiconductor index went up about 850%. Right now, it's up less than 400%.

Und wenn du siehst, okay, die Posterchild für AI ist natürlich NVIDIA, ist es um 24. Die S&P ist um 22. Und Cisco, in der heydayen, in 2000, wenn es die größte Unternehmen in der Welt war, war es um 100. So es ist hart zu sagen, das ist übervaluert. And the rise is not even nearly as great as it was back then. The second part of that is that back, if you look back at the internet infrastructure build-out, there really wasn't new quote-unquote technology per se that happened during kind of that five-year plus expansion. If you look at AI though, you've gone through three distinct phases. You went through the training phase in 2023. You went through inference during kind of 24 into 25. Now on January 30th you started the Gentic AI phase and that requires 10 to 100 times more tokens. The only way you get those tokens is with more semiconductor chips. So, you know, unless you believe a Gentic is not real, if you need 10 to 100 times more token production.

You're going to have a continued expansion, I believe, over the course of the next year, at least. We're going to dive into that in a second, but are you indicating that even if token prices end up getting commoditized, the revenue will still rise faster than the current estimates on Wall Street? Well, this is where it gets complicated. So it depends on what layer you're talking about. I believe that the large language model layer is going to get commoditized. So the Anthropic, the OpenAI, because you've had, as you saw last week at the World Artificial Intelligence Conference in Shanghai, that was the ninth one. You saw, for example, new models coming out, and specifically Moonshots Kimi K3. In the past, people said, oh, the models out of China, they may be cheap like DeepSeek, but they can't compete at the leading edge. Now, you're seeing these models compete against the best models coming from closed source, OpenAI, or Anthropic. So I think at the model layer You're going to have commoditization And that's going to slow down The revenue growth to your point Which is one of my concerns Because you are getting that commoditization Having said that.

That frees up more money, if you will, for companies to say, okay, I don't have to spend a bazillion dollars on my large language model because I can use an open source version that's 90% cheaper, but I still need to produce the tokens. And so that leaves money to actually spend at the infrastructure layer, which I think is going to continue to grow. And you can't open source that, right? You need the chips to actually produce the tokens. And so I think, now is a phase where you have to get a lot more selective around which portions of the AI trade you want to be involved with. So if the LLM models are getting commoditized, is that already reflected in investors' expectations? Und wenn nicht, welche Firmen wir sehen die größte dramatische Impact? Also, looking at die IPOs von Anthropik und OpenAI in den nächsten Monaten oder vielleicht in den nächsten Jahren. Ja, ich glaube, das ist in die Leute, die denken, weil viele Leute sind still sort of fighten.

That thesis, that large language models will get commoditized. And obviously, OpenAI, sorry, Anthropik going public soon, that'll be very interesting to see what they have to say for their revenue growth, because they've seen the fastest revenue growth this year. I mean, we've never seen something at this scale. And to give you an idea, at the beginning of 2025, the two companies, OpenAI and Anthropik, together, They had an annualized run rate revenue of about $7 billion.

At the end of, sorry, what did I say 1995? I meant 2025, geez. At the beginning of 2025, they had a combined run rate revenues of $7 billion. At the end of 2025, they had a combined run rate revenues of $29 billion. Right now, the combined run rate revenues of those two companies combined is over $70 billion. So we've never seen companies at this size grow this rapidly. Now, the question becomes, as you pointed out, if you've got some pressure on revenues at the model layer, those two are at the epicenter of that. Und der beste Beispiel ist Coinbase der CEO hat eine Post auf June 26, wo er sagt, ja, ich habe diese Open-Source-Models und ein paar andere Sachen. Und sogar mein Token Produktion ist auf, mein AI Bill oder was ich spenden auf AI hat sich um 50%. Und so, das ist die Part, that gets very tricky because if you have an issue with Anthropik.

Because we're dealing with stock prices, right? And stock prices are two things. It's earnings times the multiple. What multiple are you willing to put on it? And that's where this is going to get tricky. Because if you do have an issue with the revenue growth at a company like Anthropik or OpenAI, and OpenAI, I think, is the one that's going to really have the problem. Because they're stuck between Google and consumer AI. And Anthropik, which from the very beginning focused on enterprise AI because corporations will pay for it. Only 5% of OpenAI's consumer customers actually pay for the advanced versions, right? They use the free stuff. So I think OpenAI's got the huge problem, which is why they've postponed their IPO into next year.

And we don't know what the numbers are, etc. We haven't gotten an update very recently, But I think part of that is tied to the fact that their growth compared to Anthropik has really slowed down. I mean, you've been skeptic about OpenAI and the whole OpenAI universe for quite a while. I think since I've pretty much known you and a big fan of the Google universe, with Gemini capturing more market share as well. But let's go back to Coinbase for a second. So if Coinbase manages to reduce the token budget by 50%, is this a one-time phenomenon or do you see the top 1% of the dominant token buyers in the United States doing the same thing? What would the consequence be for the sector?

That's a great point. And this is why on June 20th I said, you know, I'm expecting an AI speed bump. Because if you think of the timeline of this, what happened was the following. January 30th, you get the formalization of Open Claw, which you could say kicked off the whole agentic AI move, which demands 10 to 100 times more tokens.

By March, you had companies like Meta talking about token maxing. We have token leaderboards and employees could see how many tokens they were actually using up. Then Uber came out in April and said, hey, we blew through our entire AI budget for the year in four months. That's a problem, obviously. And then you've then switched in June to companies going from token maxing in the month of March. In June, they were talking about token minimization, which is why I put out, hey, I think we have a speed bump problem here. And then you had Coinbase CEO coming out on June 26, talking about, hey, we managed to cut our AI bill by almost 50%. And I think to your point, the top 1% of corporations that spend, they account for probably over 50% of the total AI spend. They're going to try to do a Coinbase system. Now the question is, can the other 99%.

Grow enough to offset that 1% slowing down. And even though token, and you can see it in the data, the token production has skyrocketed since January 30th when OpenClaw got formalized to now. You've also seen the price per token dropping since about June, which coincides with when you started to hear about all this token minimization stuff. So this is probably one of the most complicated environments to deal with. And what I'm trying to think through is, can, the stocks of semiconductors perform really well if companies like Anthropic and OpenAI are seeing slowing rate of revenue growth? And we're about to find out very soon.

Und, you know, we have the earnings of Google Wednesday night. We have the earnings of Microsoft and Amazon next week. What are you looking for? What do you think the market will care about the most? Yeah, I mean, the market's going to care about a lot of things. I think with Google, Google's really tricky for the following reasons. Google is picking up the fastest-growing customers in AI. They're the smallest of the big three. Amazon's the biggest. dann kommt Microsoft, dann kommt Google. Google's Cloud Platform Revenue growth, that was growing 34% in the September quarter. That accelerated to 48% in the December quarter and then accelerated further to 63% year-over-year growth in the March quarter. I think that's going to accelerate again to over 70% because remember, in the June quarter, you had token maxing going on. Google, though, doesn't guide. So you're not going to get any guidance. Now, you may get some qualitative commentary around what they're seeing in Google Cloud platforms. But unlike Microsoft and Amazon, which will guide to the September quarter, Google won't. I think Google's revenues and EPS forecasts after they report the quarter are going to go up for the future quarters because their core business, which is search, obviously. And YouTube and all the other stuff, that's doing quite well.

So it's really, to some degree, going to come down to the commentary and then what do they have to say about things like CapEx? Because Google's obviously trying to figure out how to control their CapEx budget, which is why you've heard yesterday there was some talk that they're trying to design a chip, where they're etching into the actual silicon parts of the Gemini LLM model to cut, the energy and the cost to produce these tokens. And so everybody's trying to figure out ways to control these bills because, the free cash flow for all of these companies has dropped dramatically and in some cases going negative for some of the most cash-generated companies in the world, which is why you saw Google do that $85 billion, equity offering, which I never thought we'd ever see them having to raise money. But it tells you how quickly.

Cash-consumptive this AI infrastructure spend is, and how much they're trying to get this under control. And there are rumors that Meta might follow in the footsteps of Google, also with a capital raise. Of course, you had also a lot of IPOs, SpaceX IPO, which is now below the initial IPO price. I know it's a different ballgame with SpaceX. You can't really compare those companies in many ways. But, you know, I don't know if you saw the most recent Bank of America fund Manager Survey, they had an interesting question in the survey and asked, do you expect any hyperscaler to announce a cut in CapEx in 2026?

And 61% said, no, such a cut will not happen. Yet, when you argue that you're going to see LLM models being commoditized, wouldn't that.

Wouldn't that be an indicator that capex expansion at its current rate might run out of steam, oh 100 and and that's are you that's arguably the wrong question the right question is do you what do you think the capex expansion is going to be for 2027, because right now for 2026 you're running capex growth of over 70% right and last year was about the same you know levels, and to your point Marcus which I think is a great one the survey should have been well do you think growth in 2027 is, 50% over 30% over 20% and what I would argue is if that growth rate is much lower than what people are thinking Bis zum nächsten Mal. Then you've got a potential crop. And so that's more what I'm focused on because you're not seeing all of these companies trying to design custom ASIC chips and, like Google's results yesterday or the news yesterday about them wanting a specialized chip with.

The LLM infrastructure kind of etched into part of the model. And they're already drawing ASICs with a specialized chip for inference versus a specialized chip for training, they're all trying to figure out how do we get these costs down because they're exploding higher. Because I'm sure they're seeing this too because if people want lower cost tokens because they can't afford the higher priced ones and quite honestly, they don't need it. And so for your viewers, right? It'd be like saying, okay, what's three plus three? And you're asking a PhD in physics for that answer. Well, that's going to cost you way too much, right? They can ask you or me that question and pay a whole lot less. And so there are certain things going on underneath the covers that we've just started to see this play out. And so that's why you've got these three different vectors happening at the same time. One percent of corporations trying to cut their AI bill. The other 99% growing their AI bill. You've got token production going up, but you've got the price per token dropping. And you have those three things happening at the same time. And then, of course, when it comes to stocks, you have to add in, well, what multiple do you want? And that's where it gets really problem. Ich bin wirklich interessiert.

Hypothetisch, ob ein Hyperscaler-Einrichter sollte das Unthinkable und tatsächlich signalen, dass CapEx-Expansion ist zu erreichen, dass es max ist.

It'll be Armageddon the next day. But you see, that's exactly my question, because the stocks of the hyperscalers did not really perform very well this year. And there was a time where people wanted to see ever higher CapEx spend to make sure that they're going to basically win the race. But now, if commoditization is becoming an increasing topic, Alex Karp of Palantir criticizing the LLM-dominant players for the pricing models they have, wouldn't a potential reduction of capex or even the signal of just keeping them steady not lift the stocks of the hyperscalers?

Ja, ich meine, so ein paar Dinge. During this Momentum Drawdown, Magnificent Seven didn't actually perform that poorly. Und part of that is people, as they thought about, well, if I do have, you know, less money being spent on this, well, that'll be great for the hyperscalers, because what people really hate is the fact that they're having to raise money and their cash flows are dropping or going negative. So for the hyperscalers themselves, depending on who they are, the stocks may actually work well for that, right? Remember the day, I'll give you a good example of it, you know, the day that Meta came out or there was this news story that Meta would go ahead and, you know, potentially be launching their own, public cloud. And that they might end up, you know, being able to monetize that. The stock was up, I think, 9% that day. And so what people want to see is better cash generation from all of these companies. And so what you may see that day is the hyperscaler companies, their stocks go up, but the semiconductor companies get hit at least for a day as people worry about, well.

Where exactly is the spending getting cut. I mean, I lived through the late 90s. This is probably the most complicated situation you've been in for a very long time because if you do, get the poster childs for this trade in trouble like OpenAI or Anthropic with what rate of revenue growth there is, then people are going to start to think all the way down the stack, down to, you know, through the hyperscalers, to the semiconductors, etc..

What are the implications all the way through? And that's, you know, that's what I'm trying to think through right now. I think it's way too early for these guys to say, hey, we're, you know, slowing down CapEx right now because there's no reason they'd all be raising money, right? Amazon went ahead and did a big bond offering, multiple bond offerings already this year. Google did that equity offering. Meta's raised a bunch of money from bonds. That doesn't seem to indicate, based on what they're doing, that there's any slowdown in the near term. Because the riches for the company that survives is so great that none of them want to lose that race right now. Like Larry Page, there was that famous quote of his, I'd rather go bankrupt than lose this race. So, you know, I don't see that in the near term. Do I think in 2027 there's going to be a plateauing of this? I absolutely do. I want to touch base on that in a second and look at your third concern, the cost of semiconductors. But first, you know, a quick touch on the open source models in China, Moonshot, Kimi, K3. First, people were saying it might be another deep-seek moment. Now it's sort of a deep-sea moment. But I would like you to help us.

Explain the difference you see between the Chinese approach to AI and LLM models compared to the American approach. And who's going to be the winner at the end of the day? Or can they both? You know, America will have the leading frontier models and China will have the more efficient, open source models that will get established worldwide on a faster clip because they're cheaper. Yeah. So if you think of the evolution, when DeepSea came out in early 2025, the main advantage was it was a lot cheaper because as they say necessity is the mother of invention because the Chinese were cut off from the leading chips from the US, they had to say okay we can't get better performance just by buying more hardware, we've got to get creative with how we're thinking about the actual large language model, and so they introduced DeepSeq introduced for the first time this mixture of experts. And so it's what I said before, right? Where, let's say you have a math problem, you don't need to give it to the entire model. Like the healthcare part of that large language model, it doesn't need to look at that, right?

You only need to give it to the math portion of that model. And it's called mixture of experts. And so they had this mixture of experts thing set up, where for the first time, you'd have sort of artificial intelligence sitting on top of the questions, looking at the question and going, well, do I need to give that question to the entire model? Or can I give it to a small portion of the model so my costs go down? Now, the performance though for the entire model was nowhere near where the leading models from Google Anthropik OpenAI were. What's different about what happened this last week with Kimi Kay is you say, okay, even if that model was a distilled version of the Anthropik model, which I think it was, the performance was actually better in some cases than the Anthropik model. And so people looked at that and said, wow, okay, if it was just a clone of the Anthropik model, the performance should have been worse, but in some cases it's actually better. And that's a big difference from DeepSeq where the performance was nowhere close, it was just a lot cheaper. So that's the big change. To your question about can the two ecosystems coexist, the answer is absolutely. But the question is...

How much of the tokens end up on open-sourced models? And how much of the tokens go to the leading-edge models? And that's how Coinbase was able to cut their bill by so much.

So the leading companies, from what I've heard, they're putting 90% of their tokens now through the open-sourced models. Wow. And they're only putting 10% through the leading-edge models. But they're not suffering a performance hit, necessarily, because they're just being smart about how they do it. They're asking Marcus and Dan what 3 plus 3 is, and then they're asking the nuclear physicist, how do I design a nuclear reactor, right? They're not going to ask you and me that question. That's not true, Dan. They'll ask me what 3 plus 3 is, und Sie werden die wirklich toughen Fragen stellen.

Da gibt es ein rumor, Sie haben sich über die White House über die Strecke oder blockieren die Purchase der Chinese Open Source Modell für eine American-Cooperschaft. Do Sie think das ist eine gute Idee? Well, so big picture, es ist eine horrible Idee wenn Sie sagen, US-Companyen kann nicht mehr access zu Open Source Modell. Denn think about it this way. It'd be like saying, well, the rest of the world, can buy gasoline at $4 a gallon, and the US has to pay $40. Like, that's not going to go well for the US, right? So now, do we have some good open source models in the US? Absolutely. So if you think about what Meta introduced, and I think Meta would be the big winner if we went down this path, which I think would be disastrous. But let's say that it occurs. Now, don't forget, there's also a push from the Chinese side where the Chinese are going, well, maybe we want to restrict the US from being able to get access to these open source models because they're so good.

So, there's a push from both sides here. And now, the good news is the U.S. Has some excellent open source models. So, as I said, Meta, one of the things that came out is that they're planning on offering, their own cloud offering with their Metamuse Spark model at about one-fourth of the price of OpenAI or Anthropic. Now, that's not one-tenth of the price, but it's still a pretty big reduction.

So, we do have some alternatives in the US. I think the one thing I'm hoping that the US administration has seen is we may have cut off China from the leading semiconductor chips, but has that slowed them down in AI? No, it's actually made them have to get much more creative and go, all right, well, how do we make the models better? Not how do we get even more powerful hardware? Because really for AI, it's both, right? You want an efficient model sitting on efficient hardware. And the US companies have been able to get away with, well, we got much better hardware than everybody else. And so we don't need to focus on if the models are good. Because think about it. Since DeepSeek, the gap between the open source models from China versus the US or the leading edge models in the US versus the open source models from China, that gap has actually, narrowed over the past year, when China hasn't been getting access to the leading semiconductor chips. So it's clearly not working on semiconductors and it's actually forcing China to be more creative. And so the last, for me, I think the last thing you want is to repeat that with the actual large language models themselves where you're trying to restrict that too. Yeah, couldn't agree more. Let's get to your last point in the last few minutes.

The semiconductor stocks and the pricing of memory. Morgan Stanley a few weeks ago pointed out that the semi-earnings revisions breadth is beginning to decelerate on the upside. And we're reaching extreme levels in earnings revision breadth, historically also a sign that we might see a normalization in the sector. Und based on their prediction, Looking at memory prices, which Sky rocketed dramatically, we will see a peak pricing in Q4 and with a reduction in pricing starting in 2027. And.

Looking at Calyx's earnings warning that we got this week, and Calyx is a tiny company, It's not necessarily a sign that it's a general issue. But one of the aspects was pricing. Memory prices are too high. They are raising prices to pass that on. Also, with looking at the earnings season, Dan, what are you looking for in the semiconductor stocks? And when do you see a top in the pricing environment with all the supply coming into the market also the next one or two years? I mean, everybody from SK Hynix to Samsung, everybody seems to be increasing capacity by incredible volumes. Yeah, well, there's a lot of questions in one. Yeah, sorry. I just realized that. All right. So if you look at what happened last week, you saw ASML and TSMC both put up great numbers. Both have forward estimates go up. And I think ASML was down 3% last week and TSMC went down 8%. Samsung went down, Micron went also down when they reported in June. Yeah, now, remember, this was during the momentum drawdown. So you have to separate out stock prices, valuation from necessarily fundamentals. Now, if you're asking about memory...

The one thing is you have to remember the new platforms ramping up, the new Vera Rubin platform from NVIDIA that's ramping up, that's replacing the Blackwell platforms, they take three and a half times more memory to function. So memory usage is going to still remain really strong and you can say it's actually going to accelerate in the near term. Now, if you're asking longer term, you have all of these people saying, oh, you know, memory is not cyclical anymore and all of this. And I go, that's completely idiotic in my mind because, you know, I was thinking the same things in the mid-1990s when I was young and just started in the industry. And then, you know, 1995 rolled around and you had the worst semiconductor correction you'd ever seen after a good solid five years of expansion. And I remember people saying back then that semiconductors weren't cyclical anymore. And obviously you saw what happened in 01, 02. And even you can argue coming out of COVID, where you had a lot of semiconductor companies signing long-term supply agreements. And then what did you see? And that was in 2022. And then in 2023, you had a horrific correction.

Because you had just, you know, demand slowed down, and you had all this supply and you got, you know, microchip is probably the best example of this, just got absolutely killed because they had all these long-term supply agreements that then all of a sudden had issues. So that's the part that makes me nervous on semiconductor memory, which is i think a lot of people have assumed it's different this time and it absolutely in my mind is absolutely not because as you pointed out right you have hynix going public they raised money in china which should be what really scares people you have cxmt, going public there and they're raising a bunch of Und Chinese memory ist ein großer Faktor in der Zukunft. Ich meine, Apple...

Is trying to get the okay to use Chinese memory in their China iPhones. And the question is, will the US government allow that? Because you go, well, if you're selling a phone in China, then, you know, maybe you should be able to do that. Because remember, semiconductor pricing has gone up so much that it's actually feeding into inflation metrics. I mean, you brought up Calix, but Apple went ahead and said, hey, we need to raise prices between 17% to 25% across a lot of our products. They didn't do it on iPhones, but then just earlier this week, they actually are raising prices in Japan because on top of semiconductor prices going up, you have the yen dropping. And so they've had no choice. And Microsoft obviously raised prices on Xbox because they had semiconductor pricing issues there. And I think going into the November midterms in the United States, Inflation is a big topic And so it's not going to surprise me If the US administration says Hey.

For products only sold to China will let this happen. China's only 5% of the global supply for DRAMs right now. Interesting. I think they're going to be 30 by the end of this decade. We're pushing for that, and they're pushing for self-reliance on semiconductors anyway. So you put all that together, and from a, you know, I see forecasts for people saying, oh, you know, memory's going to be tight all the way through 2028 and 29. I go, I just don't see that. So the multiples on some of these stocks, which have, when you're running over 80% operating margins, I'm not talking gross margins, I'm talking operating margins over 80%. You're way over-earning. Yeah, I mean, we basically, I mean, many argue we don't really have a tech bubble. We have an earnings bubble and a margin bubble mainly. Und das bringt mich zu der letzten Frage, Dan, looking at die earnings season. Du hast schon gesagt, Samsung, ASML, Taiwan, Semiconductor, Micron. All die stocks waren down after fantastic earnings. Ich meine, da war ein oder zwei kleine Probleme in die Ereignisse und in die Guidance, aber die stocks waren down. Looking forward zu der earnings season, was du denkst die Investoren würde sehen von Semiconductor stocks and the companies paying those prices.

What are going to be the biggest hurdles here and which companies are affected most? Yeah, I mean, I think you're going to get a good read on this after Google, honestly, on Wednesday night because what they have to say is going to impact all of this. And then we'll see, I always say it's never the data. It's how do investors react to the data. And so that's going to be the key, right? Because the good news is, as we talked about at the start of this podcast, the momentum index has gotten absolutely crushed relative to the historical metrics, right? It was down 28%. The historical average drawdown is 20%, right? Semiconductors got hit very hard. So the valuations were never big to begin with. And as you rightly pointed out, we have an earnings bubble. We have a margin bubble. We don't have a valuation bubble. The valuation bubble is what makes things super dangerous, right? NVIDIA at 24 times relative to the market at 22 times when NVIDIA is growing revenues at over 80%.

That's not a valuation bubble. You could argue this thing's a value stock. It's incredibly cheap if you believe the numbers. So that's why I feel like with semis as a group, after this latest drawdown, I am more bullish on that sector relative to the companies that are closer to Anthropik or OpenAI, because I think that's where the pressure is more going to be. You're going to have token production continue to skyrocket.

And for that, you're going to need semiconductors. It's either that or you don't believe agentic AI is real. And I think agentic AI is very real. And so I think you're going to need more chips in the future and that's semiconductors. So this is going to be, you know, I think look at what Google has to say, how the stocks react to what Google has to say. And that's going to give us a much better feel going forward. But to me, this is the most treacherous time to be in an investor that we've seen since this whole AI, you know, build out began because of all of these things, moving in different directions between the 1% cutting their spend 99% increasing their spend tokens go production going up, but pricing going down. Und, you know, who benefits and who gets hurt from open source. This is going to require some real selectivity to make money. I am so glad you're saying that because you, of all people, know best.

I'm just looking back at history. In the end 90s, I was at the New York Stock Exchange on the trading floor. And understanding the internet bubble, in many ways, was much less complicated than understanding what's happening right now. I think that you also need to bring such a depth in knowledge and know-how and technology that you did not need back in the 90s. I mean, you needed some knowledge, but by far not as much knowledge as now. And I think it's a good way to wrap this. When I look at the speculative action in also leveraged ETFs on Wall Street or in South Korea, I mean, it's a borderline insanity. I mean, it can go well for a very long time until it doesn't, and then you pay the price for it. Well, I think to that point, that's what should terrify people, because you not only have all these leveraged ETFs, but you have global debt to GDP for all of these governments around the globe. We're running deficits that you'd expect to see in a recession. You shouldn't be running 6% to 7% deficits when you're in a really strong economic climate. Und so, wenn die nächsten Däntrn hits, weil wir haben nicht ausgeliefert Däntrn, da ist ein Däntrn zu werden, dass es ein Däntrn gibt.

Und zu deinem Punkt, Markus, das ich denke, ist ein Däntrn, wenn du Leverage DTFs auf dem Däntrn.

Bubble in and of itself, your flexibility to deal with the next downturn is very minimal. And that's the scary book. I'm glad we're ending on a positive note. That's great for the German audience.

Man, you understand our culture so well. Thank you very much. It was a pleasure as always. Talk to you next time. Thank you. Sounds good, Markus. Thank you for having me on. Und jetzt der obligatorische und barfinsichere Risikohinweis von Mr. Markus Koch.

Meine Damen und Herren, liebe Aufsichtsbehörden, ich darf Sie darauf hinweisen, dass die Inhalte in diesem Video ausschließlich der allgemeinen Information dienen, und keine Empfehlungen zum Erwerb oder der Veräußerung bestimmter Finanzinstrumente und somit keine Anlageberatung darstellen. Insbesondere können Wall Street Correspondents und der Vortragende nicht einschätzen, inwiefern die im Video gemachten Empfehlungen ihren Anlagezielen, ihrer Risikobereitschaft und Verlusttragfähigkeit entsprechen. Wer also auf Basis von Informationen in diesem Video etwaige Anlageentscheidungen trifft, trifft diese ausschließlich auf eigene Verantwortung und eigene Gefahr. Und das wiederum bedeutet, dass weder Wall Street Correspondents noch der Vortragende, das ist immer noch dieser Mann hier, für Verluste haften, die sie dadurch erleiden, dass sie Anlageentscheidungen aufgrund von Informationen oder Kommentaren in diesem Video getroffen haben. Vielen Dank für die Aufmerksamkeit und weiterhin einen erfolgreichen Handelstag.