SPEAKER_00 0:00

Eric, we are here with the AI Investor Podcast, making good on our promise to shift to a weekly show. It took us two years, but we're we're finally getting into a rhythm here. And I know our listeners are very grateful. Uh, there's a lot to talk about today. The markets continue to rip. Of course, that's been true, feels like for the last two years, but we're actually at uh an even more exceptional point in history when we look at what's happened over the last eight weeks. I know we're going to talk about that. And I know we are going to discuss the seeming dichotomy, you know, that that things that seem like they shouldn't be true at the same time are, which is that NVIDIA has gone on a multi thousand percent run, and yet it is cheaper than ever despite higher growth rates. And that's actually a dynamic, you know, NVIDIA is sort of the poster child here, but that dynamic is playing out across a lot of this space. That and a lot more coming up next. Eric, like usual, though, we always want to kick it off with what's going on in the broader climate here. So the markets have now been up for eight weeks in a row. Uh, this is stunning. And to people who have not been long-term investors, this is one of the most euphoric couple weeks that I can remember in recent history. And I'm sure maybe this is not quite at some of the other manias of the past where you and I were less active, but in the last 15 to 20 years, this last week goes down as historic in my mind. What about you?

SPEAKER_01 1:17

Yeah, Austin, number one, I'm glad to have you. I'm glad you survived. You sent me some pictures from an Airbnb. You were at this past weekend. It looked like something out of saw, um, or or maybe a Texas Chainsaw Massacre, but I'm glad you made it. Maybe we could actually put the picture in the show notes if we want to upload it. I don't know though. I will be willing to share that photo.

SPEAKER_00 1:39

Let's do a quick side quest here. Let's do it. I I Brad, I don't know how you want to edit this for the intro or the cold open. Brad, do whatever you want with this episode. You've got you've got full license, but I have a bone to pick with Airbnb for allowing this location. I'm I'm not I'm not gonna shame the person here who has the listing because they seemed like a nice person and they responded to our texts and they were very uh you know kind. But there this place felt like it was falling down around us. We were forced to book this because we had our dog with us, and this was the only place um around us that took dogs. Eric, as you know, I went down to attend the West Point graduation to support a family member there who was graduating. And last minute we had to change our Airbnb plans. We got this wonderful uh farm that looked bucolic and none of the doors locked. It felt like the house was gonna fall around around us. It was quite literally over 200 years old and it showed cobwebs everywhere. It was it was the worst Airbnb experience I've ever had. How was your Memorial Day weekend?

SPEAKER_01 2:40

That's what's great about you, Austin. You break the divide between Amish lifestyle and an AI investing podcast. So seamlessly. So I love to see it. I had a great Memorial Day, and well, it was fine. We'll say that. The weather wasn't so great. But Austin, what we're going to do for these Thursday shows is sometimes we'll do a longer focus. I'll talk about what I want to do next week. Last week we had John Rotanti on. If anyone out there missed that, we got a lot of really positive feedback. He talked about some stocks we normally don't cover. So if you just didn't have the time because I was an hour, hour and 15 minutes, I would really encourage everyone, go back, listen to that. The the kind of feedback I've gotten has been incredible. And also for an episode like this, we will probably do our best to keep it a little shorter, but we do want to talk about what's going on in the market. So, Austin, like you said, it's historic times in the sense the market's now been up for eight weeks in a row. Um, there's only seven instances in market history post-World War II where the market was up for eight straight weeks and it gained more than 12%. Now, here's the good news if you're out there, uh, limited sample size, but on average, what came next is a 17% gain across the next year. And Austin, the other thing is too, we've been talking about kind of the euphoria in the markets. So it might be kind of interesting. We're gonna take a step back here and talk about this broader situation. You know, doing this weekly paint allows us to kind of talk about these week by week conditions. But the market, while it's up 9% year to date, um, it's actually gone cheaper, which might shock people because since the year started, earnings estimates for this year are up 14%. So when you look at the PE of the market, it's actually gone down. But Austin, what I wanted to talk about, and you teed up a little bit, is the sale of two markets that's developing. I think it's really important for everyone out there listening to kind of understand what's going on.

SPEAKER_00 4:43

Oh, go ahead. Well, and for this next segment, I want to pause, be right here, and say what what you're about to do is actually one of the most important and perhaps one of the most difficult things to do in investing. And that is to actually don't pull up the stock chart, right? If you pull up the stock chart for any of the stocks in your portfolio or the AI trend or the tech stocks for at large, the the you bias yourself by saying there's no way this trend can continue. It's up 500%, it's up 800%. The most difficult thing to do is to actually forget that and to hold yourself accountable based on the investing conditions today. What is true today? What is the backlog of orders that are going to be coming to these companies? What is the valuation we are paying for them? One of the things I really liked about your episode with John Ratonti is he talked about the durability and magnitude of this trend and he put it into a decade-long framework. And that is the only way to think about this. You almost have to forget what the stock chart looks like up until this point because that cake is baked. What we are trying to do now is invest and going forward. So this is one of the hardest things to do when things are going on a run like this is to pause and say, what are the actual investing conditions that I'm putting my money to work in today, absent that chart. And as you just noted, shares have actually gotten cheaper because earning estimates are increasing by a greater rate than the market appreciation itself, which means the price you're paying for that few, those future earnings, the future cash flow is now less than it actually was a few weeks ago.

SPEAKER_01 6:06

Yeah, and the one thing about the market, though, is it's often unevenly distributed. So we look at NVIDIA, it's down about 5% across the past week. We go back three months, it's up 8%. Um, or I should say three months. But here's the thing, Austin, across that time, expectations for NVIDIA's profits this year, they're up 14%. So this is what I'm talking about. It's getting cheaper. Price has gone up 8%, but the expectations for its earnings are up 14%. But when you extend your time horizon, you see expectations for next year, the one beyond this, for NVIDIA are now up 25% across the past 90 days. 90 days ago, Wall Street thought NVIDIA would make 1010 per share next year. Now they expect it to make 1266. So Austin, here's the thing that this is gonna really um this is really gonna shine a light on. NVIDIA, it's 8.5% of the NASDAQ, the QQQ, which is one of the most popular indexes, but it's expected to contribute 37% of next year's profits for the entire index, which is crazy. NVIDIA Micron Broadcom combined, they're just a little under half of all of the profit contributions to the NASDAQ index itself. But often this group right here, NVIDIA is now trading for 16.5% next year's projected profits, all while it's expected to grow 42%. Broadcom's now trading at 17.5% next year's profits, also with sterling growth rates. And Micron, uh, according to a new UVS estimate, we'll we'll discuss later in this episode, is trading for 5.8 times next year's estimate. So what we're looking at, Austin, is a tail of two markets. Yeah, the market's cheap, but it's because 50% of the market's profits are coming from a series of very high quality companies growing earnings that are priced now at their historically low prices relative to what they've been trading at. You know, we could look at, and we'll again we'll spend more time talking about this, but NVIDIA is about to release its new Ruben system. It's gonna enter production in the second half of the year. We've talked about how Ruben is going to create all kinds of new data, you know, it's gonna create new architecture for data centers, how they're designed. And it's leading the enthusiasm for Ruben is part of what's leading to NVIDIA's estimates going higher and higher and higher. And what we're seeing is third-party estimates. It looks like NVIDIA is going to take market share due to Ruben. But this is this is the story in a nutshell, Austin. All of this news is happening about Ruben. No one wants to buy NVIDIA, right? You you go on Twitter right now, and Austin, it's like a fervor. Everyone is looking for bottleneck stocks, these tiny stocks that are as speculative and as small as possible. Again, you you go on Twitter and you see now accounts often with just hundreds of followers, a thousand followers, they'll bring up a stock and it'll jump 20 or 30 percent a day, and everyone's kind of chasing this mania in these kind of narrow bands of semiconductors. So this is this is almost often a disconnect that's emerged in the market, where many of the most speculative stocks that are often pre-revenue or close to it, they're trading for absolutely massive multiples. But the most high quality stocks that are actually driving the earnings that make the market look cheap, well, they're trading at again the lowest point they've traded at. So Austin, here's the point I wanted to begin this episode with. Both of these things, it's hard to square both of them being true. I saw Gavin Baker, he was on um some podcasts this week. I think he did two or three of them. And he brought this point up repeatedly that there is this disconnect in the market. And I think this is a little bit, we'll move on to memory stocks next. But this is what's happening. You brought up earlier, Austin, you really like that durability point. You can't have NVIDIA priced like it doesn't have durability, Broadcom priced like it doesn't have durability, micron priced like it has durability, and then all the most speculative stocks priced like they are going to rise for a decade to come. So something has to give in this. Um, and we're just in a very unique market environment. And we're gonna talk about, like I said, micron and memory next. And I believe the price action we've seen in Micron is kind of a reaction to this scenario because in a way it is illogical.

SPEAKER_00 10:49

I I have wondered, and and no company in the industry will illustrate this better than the memory space right now. I have wondered if a bit of this is just the muscle memory and the pattern recognition that used to be true for years that these sectors are cyclical, right? And memory stocks were the one of the canonical examples of it in tech, but every every industry, every industry, you know, industrials are notoriously cyclical and caterpillar and deer. And to the point you're making, cyclical stocks always look cheapest right at the top. So I have wondered if a lot of this mispricing is just Wall Street continuing to act as if the things that were true historically are now true, that we're in a cyclical trend and this is just part of the regular cyclicality patterns of memory stocks and that they are not going to rise to the occasion of uh, you know, increasing output or increasing pricing power. I'm not sure if you have a theory on that. And I know we're going to talk about microns specifically, but I've tried to square this point that you're making. And I can't, I can't figure out why somebody or you know, why an analyst would look at the reality you are talking about and continue to price these positions this way, unless except that, except historically this was a cyclical pattern there's trying to miss the top. Is that what do you think? Off-base, something else.

SPEAKER_01 12:04

That's interesting because you know what else is a cyclical industry? Semiconductors have right. Well, I mean, that's what I mean.

SPEAKER_00 12:10

It's it's GPUs and memory, right? I mean, it's it's it's the same question. But is that does that hold or off-base? Because it doesn't make sense to me.

SPEAKER_01 12:17

I I think we might be seeing a couple movements of money in different places. So you might have a little bit more of a lot of institutional money around something like Micron. It's a trillion dollar company. And they're gonna have those long memories. I I mentioned in the past, but memory right now is at 80% margins. The best margins that ever achieved at previous peaks was 60%. If you're a student of history, if you've lived through this before, you're going to be very nervous about this. Now, on the other end of that, I've been in optics, I've been following this field since 2009. Um, like any space in semiconductors, when there's an oversupply, that you reach the end of a cycle, the other half of it's absolutely brutal, right? And I think that you have a lot of investors piling these spaces, including many of the smallest, most speculative stocks that haven't been through these cycles before. So, you know, there's not the same where the money is flowing from doesn't have the same muscle memory, you know, to talk about what you were saying, Austin. The bigger you are, the more you're gonna be controlled by funds. You know, NVIDIA right now, it's it's a five plus trillion dollar company. You know, it is going to be large institutional flows for money, things that are more concerned about what could happen on the other side. These smaller stocks have a lot of retail flow. People not as concerned, willing to pay valuations that are above historical norms for semiconductors. And that is really what we're seeing, right? You're seeing NVIDIA in many ways that historically cheap prices for a company growing at the rates it is, and many of these smaller companies training at historically high rates for semiconductors. Um, but I do think, Austin, you know, this week, Micron, they received an upgrade to $1,625 per share. The previous street high was $1,100. Here's an interesting set. Their stock price getting from $500 billion to a trillion, it took 48 days. So let's look at the other stocks that hit uh trillion in the past. It took Apple 1,500 days, it took Meta a thousand days, and it took NVIDIA 500 days, right? So anywhere from 10 times to 30 times as long. And it just kind of puts in perspective how quickly this is happening. But SK Heinex also hit a trillion. Austin, we look at the recommendation sheets. I was kind of astounded the morning to look. SK Heinex was up nearly a thousand percent since we recommended in January of last year. Micron's up 756% since we recommended it. But you know, the real reason, again, that this Wall Street bait, this recommendation or this upgrade came from UBS. The reason they're increasing their price target on Micron is they are saying their earnings are more durable. It's this word we keep using. How long is this trend gonna last? Those stats Rotanti cited about, you know, this gigawatt build-out, how long we're gonna keep increasing the gigawatts of compute from year to year. And UBS says that Micron's EPS is gonna comfortably stay above $100. Um, you know, the funny thing, I think we recommend it for less than $100 per share. So it's its earnings every single year are gonna be higher than the price we recommended it for. Uh, they're predicting $155 per share in 2027, $167 in 2028, and $117 in 2029. They expect free cash flow across this period to be more than $400 billion. And Austin, a big reason for this stability is um a lot of memory is being locked up into long-term agreement. So it's gonna just provide a baseline for them across this time period. So, you know, it is just a few things. Number one, as far as memory, I'm sure we have some people watching this who want our opinions on memory. Why is memory staying as elevated for this long when it's had no historical precedence? Number one is just demand. Um, you know, demand for compute is insatiable. We talked about that next week.

SPEAKER_00 16:29

And memory and let's also anthropic has also really thrown a lot and many logs, entire forests onto this fire. Uh, because now inference, right? Inference is really where we need a lot of this memory. And now that we actually have models that are capable enough of being reliably independent and autonomous for minutes or 30 minutes or hours at a time, the amount of memory that you need goes logarithmic. This there, there is no prior precedence to this cycle. This is not uh a case of micron printing 400 billion in cash flow simply because there's a supply constraint today and they can charge premium prices, which is true. But the much, much bigger trend behind all of this is inference, right? You need memory for training, but you need a thousand X for inference. And now these models are capable enough of actually pushing on that demand as well.

SPEAKER_01 17:18

Yeah, and Austin, you look at the the pricing power that's coming up from this is incredible in the near term. I was looking at a bill of materials for the next uh their Rubin system VR 200. Uh so it's gonna have about 4 million in GPUs. That's up 57% from Blackwell's G V300. Um two two million dollars of its total cost of materials is memory. That's up 435%. So, Austin, you're talking about the importance of memory here, right? In one generation, the price of memory is going up 435% from Blackwell to Ruben, which is just incredible. A couple other factors I want to touch on, because this will be important later. We're gonna talk about some stocks. Um, one reason people thought the memory market would crash is because you've got these external factors. You've got these Chinese companies who would accept lower prices, who would expand production as rapidly as possible. Well, Austin, the problem is they can barely keep up with domestic supply. I saw the other day ByteDance is looking to do 100 billion in Capex next year. Well, that means they're just gonna sell all their memory to ByteDance. And finally, Austin, another point, agentic AI is here. We've talked about its impact on CPUs. Well, agentic AI has incredible DRAM needs, right? So it is, it's it's what you said, this enduring need for memory that's happening. It's these unique factors I just talked about. And so what UBS is pricing out is they're saying, well, at 1625 microns at 10 times 2028, what we expect. 14 times they're 2029. And that's a number that they could be sandbagging a little bit. So they think the stock's gonna re-rate to get out of this, you know, historical cyclicality. But one one more time, Austin is just we we have two very different events happening. One is many of the most established quality companies in AI are trading for some of the lowest P's we've ever seen. And some of the companies that are adjacent to these bottlenecks or have a pitch you can write in 30 words, are often selling for 30 times sales or historically high prices. And something's got to give, Austin, right? And what gave in the case of memory is we are seeing Micron re-rated. It's not necessarily like that these very speculative stocks had to decline for this to make sense. It just meant eventually there's enough force behind Micron itself that it rose. So something like NVIDIA right now, I know if someone's invested in it, it can be frustrating that you see all of these AI stocks rising with such vigor. Um there there is going to be a reckoning to this, I do believe. And that's why if you want to, if you are worried about valuations across the AI space, one thing that's kind of beautiful about this trade at the moment is you can create an outsized position in NVIDIA. You can get a company that is taking market share that is at the red hot core of the trend, and you can get a valuation that is, when you look at the year ahead, significantly less than the market itself. So I just wanted to deconstruct the situation. And I think it matters for all investors out there. Number one, there is something a little scary out there chasing some of this momentum. Number two, if you are afraid that you've missed the boat in AI, it could be as simple as having a larger position in some of these stocks that are very high quality and haven't quite ridden the trend yet.

SPEAKER_00 20:58

Well, let's talk about some of those. So a remarkable run for Micron and literally every company, you know, SK Heinex, every company in the memory space. Do you feel like that run is over? Or is there still outsized potential to come? And when I think about the last three years investing in artificial intelligence, it has been this constant loop of it can't possibly get bigger than this, weight, it's bigger than this, it can't possibly get bigger than this, weight it's bigger than this in every in every dimension, right? Whether it's pricing on memory, the size of the data centers that are built being built out, the gigawatt needs, uh, the capex that's coming out of the hyperscalers, everything has followed that cycle. So is memory in another one of those loops where we see something like Micron or SK Heinex up a thousand percent, say the run has to be over, or is it still warming up and there are just people who haven't got on the train yet? What do you think?

SPEAKER_01 21:49

Well, what I think is interesting, Austin. We looked at memory and you know, I had talked about it right after we had the market kind of reach system deer. And I said, I I would actually be a Buyer of memory at this point when we did that episode. But we've also issued recommendations saying if you don't want to buy Micron if you're worried about this, there are actually companies that would benefit as the price of memory went down. We talked about those margins. Memory, Micron, SK Heinick, Samsung, they're all over a trillion dollars right now because they're getting 80% margins. There are companies that that could go back down to 60%, and these companies still win in a big way. So, Austin, we recommend ACM research on the March 13th episode. I don't know how many people out there went and picked up that stock, but it's been absolutely outstanding, and the situation continues getting better. It's up 92% since we recommended it in two and a half months. And as we noted, it's it's a relative pure play into Chinese memory. And as I just talked about, Austin, China can't build enough memory for itself, let alone to flood the Western market. And the main players in China, which is YMTC and CXMT, they're expanding their production next year. And Austin, the other thing too, necessity is the mother of invention, right? Historically, how semiconductors progressed was going to basically smaller and smaller nodes. And once we get to things like five nanometer, three nanometer, they're only allowed by EUV systems that can't be exported into China. So you're gonna need a new vector for improvement. You're gonna need to look at how you package ships, how you stack them in 3D. You're gonna need to look at technologies like hybrid bonding and Huawei, which is one of the biggest companies in China, they're driving a lot of the kind of agenda. Uh they're they're basically infilling for NVIDIA now that NVIDIA is no longer really allowed to sell in the country. They just announced a new roadmap with tau scaling. It relies heavily on hybrid bonding. And Austin, the the point is here that that's gonna benefit this ACM research company. Um so ACM research, they they've got the benefits of this massive production in memory across China. China trying to keep up with the US in this Cold War. These new roadmaps with new technologies that move them away from these other areas in semiconductors and spend their money in areas where ACM research is. So Austin, this isn't this isn't like a sexy pick. You're not gonna go probably onto Twitter tomorrow and see ACM research, the number one bottle of sod, get in for a thousand percent returns. But it's it's exactly what I think is a great company that's really rock solid and is positioned in an ideal place and it's already reward investors. The second one, just quickly, is uh, you know, onto innovation. It's weird to say something that we recommend in late February is um up 19%, it's a laggard, but it kind of is. And and it it just once again it ties into this thing that I'm talking about, Austin, that when we look at where the future of semiconductor innovation comes from, where spending will increasingly accrue to it's to these areas like advanced packaging, it's to these areas um it's it's how you stack chips, it's it's the testing to ensure that these very complex chips work. And so I'm gonna be looking at Anto Innovation, which plays in this area. I'm gonna be looking at Camtech, two stocks that are kind of derivatives, right? So if if you don't want to buy memory because you're afraid by this historical norm, the thing that would reduce memory prices actually has very clear investing opportunities. And in many cases, they haven't run up in many of the ways semiconductor stocks have. So well, it feels like I mentioned this kind of turn beneath the surface of the biggest companies are are staying very cheap and the law of the market's getting historically expensive. There still is always areas of opportunity, and this is actually one of the top areas of opportunity I'm personally most interested in right now.

SPEAKER_00 26:05

Yeah, I want to go something, go back to something you mentioned on ACM research, which is you said China doesn't have access to the lower nodes, like five and three nanometer. Do you see any opportunity for you know the larger node manufacturers, not just China, but broadly speaking, as an opportunity, as we're getting to a point where like any chip could be put to use, does the size of nanometer like matter as much? You know, it and we we had seen Intel selling out their old CPUs as an example of like old stock is now suddenly valuable. Like, is a global foundries interesting here?

SPEAKER_01 26:37

Or um oh I mean I'm gonna look up global foundries has been on an absolute tear here.

SPEAKER_00 26:45

Well, they they've they've doubled they've doubled since like February or March, but oh but the five-year run is completely muted compared to everything else we're seeing. And I'm wondering if like in the spirit of like what we saw at Intel, where even old tech is you know able to command a premium, does something like Global Foundries become more interesting?

SPEAKER_01 27:05

Yeah, I mean, also United uh microelectronics is is up, it looks like 200% year to date. So Austin, that's that's been a great trend. It's it's something I looked at, I wish we had pulled a trigger on it. You look at some other areas um like optics that don't require something on the cutting edge. You don't need three nanometers for for optics. So a lot of the companies in this kind of older fab space, they they've seen incredible returns. It's something I've looked at. I I wish we had pulled the trigger on, but it definitely shows how dynamic these spaces are and how unexpected that you've got something like AI that you'd expect to be absolutely on the cutting edge. And in a perverse way, um, buying the stuff with the older facilities has outperformed Taiwan semiconductor this year. So, you know, that's Austin, that's what's always kind of humbling about this. Uh, there's there's many ways to win and many unexpected ways to win.

SPEAKER_00 28:02

Uh speaking of ways to win, the the reliable one has been seeing what technology is coming down the pike in this AI build-out and then buying the critical pieces that power that infrastructure. So you have talked a lot about the 800V transition that's going to be happening, particularly driven by NVIDIA's rollout of Ruben. And this is basically this is behind your power management thesis, right? Like we are going to be changing how we manage power delivery coming into the building and getting it into these chips. And you need all of this different equipment to manage that. So can you give a quick overview of the 800V thesis and why we need to be paying attention to it? I know you've touched on it a few times, but for any listeners who haven't heard those prior episodes.

SPEAKER_01 28:45

Yeah, I just wanted to be really brief on this because what I'm planning on is the next episode. We're going to do a much deeper dive into it. But semi-analysis, they just posted a very long, it's probably 80 pages, um, deep dive into the 800 V transition. We're seeing a lot of movement on stocks in the space. We're also seeing, again, if if you go into looking at what kind of stocks are being talked about on Twitter, which you which again, you have to review because you have to see uh where this interest level is. I predicted, you know, we first talked about in March, I predicted it would be the next big transition in AI. We talked about it two weeks ago. I predicted it was gonna see a lot more momentum. We've seen that across recent weeks. It is a fever pitch. But what semi-analysis did is they really broke down how this is gonna move in phases from phase one, which is beginning right now, basically in 2026, to fourth phase that's gonna be in the 2030s. And, you know, Austin, one of the things is this this is an opportunity, but it's also a threat because we're gonna have a lot of components in the data center, some of which are built by companies in the portfolio, which which are going to see um just a lot less sales, right? And this is the thing about AI, right? It's very dynamic. So I want to make sure that the portfolio is adjusted, not only with some of the smaller, more speculative plays, but also that your stalwarts are really well positioned for this. So next episode, we're gonna really dive into this opportunity, but I did want to just mention this briefly in this episode because I'm sure if anyone's out there um looking at what stocks are getting the most attention, you're gonna see a whole new group of names. And it's all coming from this trend. And uh, you know, like I said, I saw one account the other day with 1400 followers mentioning a stock and it was up 40% a day. Uh so that's that's kind of where we are on the mania level. And we will explore some of the smaller opportunities, but like I said, we're also gonna make sure this is a space where you can also really find stall wars for your portfolio, those kind of stocks like Eden that uh John Rotanti was talking about next week. And we'll cover both ends of the spectrum.

SPEAKER_00 30:57

Wonderful. Well, I know our listeners are here for the picks, so they're gonna have to wait until that next episode to get the 800 V opportunity. I know they're all excited about that. But let's uh let's look at earnings a little bit. So you wanted to talk about Semtech, right? So this is another recommendation that was made at that March Nadir, which boy, that there are a couple repeating patterns. I want to draw attention to this one, right? You bought some stocks in the Deep Seek sell-off, you recommended some stocks in this most recent March sell-off, and those end up being the biggest winners. And as we have talked about repeatedly, it's always sexier to be a naysayer. It's always, it always feels good to point out the holes in certain theories. And when the market sort of subscribes to that, it can be really challenging to make investments, but you need to be willing to look through those clouds and you can pick up some absolutely fantastic opportunities, as you did with Semtech, which has now doubled since March 31st. So, what's going on with their earnings? And what did you see that made it easy to recommend in March when a lot of people were in a bearish mood around AI?

SPEAKER_01 31:59

Yeah, I'm looking at March. We issued four recommendations that month. Uh, air test systems on semiconductor, both in the power management sleep, ACM research, which we just talked about in STEM tech. That's a good vintage.

SPEAKER_00 32:13

That's a good vintage right there.

SPEAKER_01 32:15

That is uh that's a fine line right there. Um, yeah, up over 125% average STEM techs now, after earnings up 103%. But it's a company we haven't really followed up with a lot. So I did just want to mention them again because again, you missed that podcast. We had a lot of people in that period that were probably low-tuned out because it's it's harder to be interested in the market went down. Contrarily, though, that's normally when you have your best opportunities. And Austin, this is a company I I really like them on the simple thesis that we look back at the smartphone boom, and uh well, Apple was a winner in that one, it should be obvious to everyone. Actually, across the first five or six years of it, um the companies that saw the best returns were the companies that were essentially cleaning up the 4G signals within iPhones and putting putting the chips in to do that. And there is a corollary to SemTech and what its business is. And essentially, what it's gonna do, we're gonna move to higher and higher speeds in AI networking. So we're gonna move from 800G all the way up to 3.2 T. We'll have 1.6 in between there. And in this transition, they're gonna have an opportunity to increase revenue by tenfold. So what we're seeing right now, Austin, is we're seeing a big scale into 800G. We're gonna see 1.6 T ramping into the second half of 2026, and we're gonna see a lot, we're gonna see a need to push faster into this next speed of networking. Um, and that is going to give them this opportunity to see 10 times the content. Often, there's another thing that came out of these earnings. What's interesting is we've kind of had this debate happening of do you go for optics or do you go for copper? This company wins on both. They're a leader in active copper, and that's a market seeing massive acceleration. So this is a stock, I don't want to cover it massively, but I did want to say it's probably one we haven't circled back on since we recommended it. It's up more than 100%, and it's doing that largely because there is so much momentum behind it, and it's not just playing trends, it's it's not just in the trend, it's seeing a massive acceleration of revenue it gets from the trend, right? So you don't just have optics rising, you have a tenfold opportunity in terms of the revenue that they can get as we continue increasing the speed in optics, and they're going to benefit from copper at the same time. So, Austin, just a stock I'm really impressed with. I'm really glad we were able to get into the portfolio. And and I hope some listeners out there were able to add as well, because um, it just seems like a tremendously well-positioned company. And final point on them we are reporting at 352. I I don't think we'll be able to get this live, um, but Marvell is reporting in eight minutes. They will have reported by the time people listen to this. Semtech should be very positive read through to Marvell, but hey, they are Marvell. They've been known to disappoint in spite of everything else.

SPEAKER_00 35:21

Let's not give them too much credit. Let's not give them too much credit. You need you need to have uh you need to have that debt in the portfolio. Well, hey, I know you want to get uh out of this quickly so you can check out those Marvell earnings, but I wanted to land with just a couple listener QA's if you've got time. So I've got two of them. Question one came from Stefan on Spotify, and they said, great episode. I'm not usually a big fan of ETFs, but for a diversified play on the AI power infrastructure buildup, I can recommend Defiance AI and Power Infrastructure ETF, ticker symbol AIPO. It has basically all of John's top picks as its largest holding. So, Stefan, of course, here was talking about your uh episode with John Retonti, which was fantastic. If people have not listened to it, they should download it and go back and listen to that. What do you um what do you think about the Defiance AI, power and infrastructure ETF?

SPEAKER_01 36:12

Yeah, I just pulled up on the screen I'm looking at right now. It's got about 710 million in assets under management. Uh expense ratio is 0.7. That's kind of what you'd expect for an actively managed ETF. Austin, I do think one thing I would really like our listeners to know is we're gonna release a 45-minute breakdown on AI ETFs, but that's only going to be on our YouTube channel. It won't be a podcast. We're releasing that next week. I'm also planning to do an hour plus on type semiconductors 101. And we're also gonna do an investing in AI for beginners. So if we have anyone out there who's listening to this episode who's saying, whoa, this is this is a lot of stuff coming at me, we're we're gonna have some resources for people to go back and kind of start from that like first principles of of learning about how to invest in AI. But again, these will be YouTube specific. We we might add some of them into the podcast, we'll see. But we'll put a link to our YouTube channel in the show notes. For anyone out there, if you're not subscribed, you know, if you love going and watching podcasts on in podcast players, I listen to most of them in my podcast players versus YouTube. I just want everyone out there to know we're we're going to have some really differentiated stuff coming. So I would go subscribe to it, I would set notifications. Now, second Austin, one of the big areas when I'm thinking about AI ETFs and the advantages that they provide is basically that they can have foreign exposure. You buy QQQ, you're not going to get SK Heinex, you're not going to get Samsung, you're not going to get Taiwan semiconductor. Most AI ETFs have outperformed versus the QQQ in the past year. And that's less about superior selection, and it's more about they get these foreign companies that have been big winners.

SPEAKER_00 37:58

Did they get some of the Korean memory stocks in there or anything in that supply chain?

SPEAKER_01 38:03

Yeah. I mean, Taiwan is now a bigger stock market than India. Um I think Taiwan has a stock market's up 50% this year. So you look at the US, whoa, it feels hot. It's up 9%. Taiwan's up 50%. Taiwan's got, you know, it's not just um, it's not just Taiwan semiconductor. You get um custom chip plays, you get a lot of in-the-rack plays. So these are things you don't have exposure to. But Austin, this one should perk your ears up a little bit. Um, Defiance, which makes the ETF that Stefan had asked about, which was the uh AIPO, they just filed today for a China robotics ETF. So when we think about, yeah, one of the challenges to investing in robotics is when you look at a lot of the supply chain, it's very difficult to find companies that aren't in China. Well, you're gonna be able to go out and get exposure there. So even if you don't like ETFs, I would tell people a good reason to look at ETFs at this point in time, and we discussed this on that kind of 45-minute YouTube, is they can be a way to get exposure without needing to have an interactive broker's account and spend sometimes $100 on transaction and needing different currencies. So ETFs as they relate to AI, it's actually one of the strongest spaces to look at ETFs because it's going to give you some exposure you'd otherwise have a real tough time getting.

SPEAKER_00 39:32

I love that point. And also, you know, particularly for robotics, I know you and I are both excited to invest in that trend, but there's an extra layer of expertise that can be hard, right? Like the the hard engineering aspect of that might be hard to punch through and really find the insights in the way that you have in the AI and chip space today. So while I know you and I both believe in that trend, and I expect we will both be invested in that trend, and you've already made a recommendation in Oustur and we'll likely make others. The nice thing about a robotics ETF is that it can it can give you that basket exposure without necessarily having the engineering or manufacturing background to understand the material science differences or the benefits of different actuators, which is just not a background that you and I have. So I also like the vehicle to get access to these trends without necessarily needing that depth of expertise in the underlying product. Uh moving on, one more question. I'd like to close it out here. This one's from YouTube. We've got Chia Chia Chang, and they said, uh, thanks, Eric. Really great insight here. I've just started to play closer attention to macro cycle risks, and you mentioned them immediately. It certainly feels like a good time to prepare for another painful cycle of ups and downs, similar to 2022. Before you respond to this, I just want to note this is also always the case. And one of the lessons that you know I have learned and come to appreciate over investing as many years as we have, there is always some pain around the corner. And investing is a lot easier when you just expect that. And it is also extremely humbling when you try and incorporate macro level risks. If you told any investor a year ago that the United States would be at war with Iran, 20% of the energy would have been, you know, shut down through the straight of horror moves. The US had $100 per barrel oil and gas was up about, you know, $450, inflation was still not tamed, which has been like a five-year challenge. If you told any investor that a year ago, they'd be like, oh my gosh, the market's going to melt down. And yet, here we are. So while it's always important to pay attention to the macro, and I think it is always important to be prepared for a macro downdraft, not just be prepared, actually expect it because they are always around the corner. It is incredibly difficult to predict. And it's much better to just expect it. And when it happens, invest through it. That's my advice. I'm curious what you say.

SPEAKER_01 41:52

Oh no, I I completely agree, Austin. Um at the end of the day, you can create scenario analysis that makes it almost impossible to invest in any environment. So the most important thing is, like you mentioned, staying invested for the long term. And there will be times that we will talk about on the show. You can go back, listen to the show. There's been many occasions, perhaps a half dozen. We talked about wanting to put your finger on the scale, and this is the time you add to stocks. Um, and those have generally been very good, right? Because over the long run, the market does rise. It does tend to compound around 10%. So if you can give yourself the advantage of lower valuations, it's just generally going to be the best time to invest. But in between, you don't see us hopping out of the market, right? That's just not how we invest because you only need to get it wrong once. If you jumped out of the market during the tariff camp drum in April 2025 and you kept waiting to for a moment to get back in, you're cooked. You're done because you're not getting in right now and you have missed out on all those returns before. And I've been through a dozen instances like that in my time investing.

SPEAKER_00 43:05

Yeah, I uh can't speak for you here, but I feel pretty confident that you and I both have adopted this. I I think in a way, the the buy, sell, hold framework, which people understand and appreciate, and it comes from Walty Research Reports, and it's it's just part of the common zeitgeist now, has actually done a lot of investors a disservice where they feel like at every moment they need to make one of those three decisions. And I think you and I have a much different framework, which is we either have two gears it's invest or invest more. So invest is like we always we're always trying to invest. As I mentioned, I forget which episode we talked about our own personal portfolio approaches. I invest in the index every single month just to make sure I am constantly investing through all markets, including markets where I may not have great ideas. And that has been a wonderful behavioral change I've made. And then when you get those big opportunities, which might be a world class stock where you have an insight, it might be the market selling off like it did in March or like it did in Deep Sea. Or any other any other combination of things. When you have those golden moments, that's when you simply invest more. So yeah, I try not to think in terms of buy, sell, hold. I try and think in terms of I'm always investing. It's just at what point am I investing more aggressively? I'm not sure about you. Agreed. No, I think that's a beautiful way of putting it. Okay. Uh well, let's close it out there, Eric.

SPEAKER_01 44:20

Uh, we are sticking to Oh, wait, I want to get to the second part of his question. Oh, okay.

SPEAKER_00 44:23

Well, then never mind. Never mind. I'm sorry to uh prematurely cut you off there. Go on.

SPEAKER_01 44:32

Oh, well, he had asked, uh, he said the interview with John was fantastic. However, I was quite surprised that he didn't mention Parker Hannifin, unless I'm overestimating the degree to which Parker Hannifin, which is Ph, will benefit from AI cooling and utility infrastructure. I would love to hear an expert perspective from John's view. So I emailed John Rotanti and I got an answer on this question. And I folded up here. And John said, Parker Hannifin is a high-quality industrial that I think is getting better, and it's in my investable universe of about 50 sucks. So I do hold in high regard as a broad-based industrial. Data centers currently only account for 1% of revenue, but C C business is growing fast and getting me interested. I hope that helps. So there you go. We we reached out to John. We got continuation from there. And the last part that was asked in this question was whether or not I'm considering allocating capital to physical AI or robotic companies such as Rockwell Automation or Teledyne. And the answer is yes. And next week we're gonna, I I I am hoping I'm gonna be able to touch on some of these broader companies that's going to continue expanding some of the space that we have so that it's just, you know, not just semiconductor companies. We're continuing to kind of give people kind of new trends and also kind of new profiles of companies, right? Um, you know, there there's something to be said that an 800 V transition could be uh hugely beneficial for a company like ABB. Uh that's that's not going to be the company that is jumping 100% three months. So uh I don't think it will, Austin. I I guess I can't discount. Stranger things have happened.

SPEAKER_00 46:09

Stranger things have happened. Micron added Micron added 400, 500 billion in in two months. So they're it's possible. Yeah.

SPEAKER_01 46:17

But I was pleasantly surprised with the reaction, especially to John's stocks last week, because these aren't stocks that were promising massive returns. What they are is they're things that can diversify and form the ballast of your portfolio. And I was really heartened to hear the positive reaction to that because, you know, like I I've said many times in this portfolio, if the goal was just swinging for the fences, I would have been doing options on all these companies. I will have that's never what the goal's been. The goal has been for people to go through a transition like this and be able to invest above the market at a time, I believe, will be one of the greatest times to you know build your wealth, but is going to come with risk. It's going to feel uncomfortable a lot. And you need someone, you know, walking you through these situations a lot. Because often, as we we've seen, we've we've had several uncomfortable situations. It hasn't felt as uncomfortable recently, maybe in marketed, but but you know, how investors are able to build a portfolio that's resilient. Um, I I know I kind of keep springing back to what I'm seeing in social media investing right now, but often it just reminds me so much of what I saw in 2021. And almost all the accounts that were beating their chests, playing these games. Well, I don't want to say games, but mentioning stocks, watching them pop 30% a day, and everyone got so excited. Well, someone held the bag on the way down, right? Someone held the bag on the way down. And it's it's not about making money on a single stock. It's not about making money a single month. It's about making money across the next five or 10 years and setting up a system that actually provides everyone out there with a better life. Um, because, you know, hitting a jackpot in Vegas doesn't give you a better life. Building your life so that you have a good job, uh, you know, a good system, all those other things that provides you with a good life. It's no different than investing. I I I'm sorry if I've gone on this tangent a few too many times, but I do think it's how's the altitude on your soapbox there? That's thin air. Thin air up there?

SPEAKER_00 48:28

No, it's a it's a wonderful point. It's a wonderful point. You know, this we you and I both agree. This is the greatest game on earth. It's a lot of fun, and it has the wonderful side effect that when you play it well, it can also be life-changing um, you know, for you, your family, your friends, your community, however you feel like uh spending those winnings. So it it's it's important to it's important to remember this is not about getting the highest return in the shortest period. That's often a recipe for disaster. This is about profiting from companies that participating in the great innovation cycle that has created so much wealth and prosperity. Okay, well, let's leave it there. Um, I like leaving John the last word on Parker Hannifin, a world-class guest. I expect he will be back again. I know our listeners enjoyed that. So, John, we're gonna give you the last word here. Uh, Parker Hannefin is in your investable universe, therefore, our listeners should give it a look. And hey, listeners, we are going to be back in your ear next Thursday, as promised. We're doing episodes every week and we're gonna try and hold ourselves to account there. So until then, Eric, thank you so much for your time. The AI Investor Podcast is for educational purposes only and should not be considered investment advice.