Transcription
Another week of uh volatile movements, and the markets have, uh, as is usually the case during any kind of a geopolitical event or even one event that's going on, it's become obsessed right now, uh, with oil and taco. And when is it going to break down and everything? So, I'm going to try to take you guys through this.
For those of, for the subscribers who saw the webinar this week, there'll be another one this coming week. I'm really trying to focus more on the weekends of just kind of going through the overall framework of what's happening. And then now that the paywall is up and everything's going on, I will continue to bring the investment ideas there, as well as a lot of the things on the turbulence model. And I'm building a lot more behind the scenes. Uh, I've really, and I'll show some of it. Uh, but I've moved on into Perplexity Computer, the video series to help any of you, and in particular your kids, uh, really be able to take an idea and convert it into something actionable. Uh, that video series, five videos, uh, went up on the paywall this week. It will go up separately probably within a month, but for the time being, uh, it's on the site. So for those of you who are in that level, you're going to see it. Uh, and it includes a lot of stuff with it.
This is kind of the opening slide from it. And I just wanted to highlight this because for those of you who are, let's say, still getting started with AI, I continue to say the same thing, which is if there's one kind of short phrase, just keep putting in the reps. It's never been more important. Every day, use it. Use it verbally, use it typing, use it every possible way that you can, but you have to use it all day long for everything.
Um, I went through and again, this is what the videos are about. There's five videos. They cover each different part of this. Uh, it also includes all of the prompts that I use during it. So, every single one of them. Uh, this is part of the Claude skill, and you can just see the depth in terms of what's gone on. You can change whatever you want in it, but the main point of it was, uh, people have been asking for a while.
This is the location for those of you who haven't found the site because you're looking at a variety of different places. It is ai.22vresearch.com, uh, or anything on 22V that has AI Macro Nexus, which is all my research and everything that's there.
So, let's get right to it. Uh, obviously Sunday night, every night now is a, uh, an event. Uh, I remember it was like that during 2008. Uh, so as long as the Strait of Hormuz is closed, uh, every day that passes is an enormous amount of oil that's off. We opened around 1:20. Uh, to say I had panic text messages all day Sunday. Uh, some of them calling for the end of the world. We obviously didn't have the end of the world. We're not going to have the end of the world. This will pass at some point. The question that really comes is how disruptive is it going to be, uh, really to earnings and to the economy. And the longer that we sit up here and every day that passes, it really does extend because you're talking about a lot of oil that's not able to get through, that is not there. Um, and, you know, they tried everything this week, a release, a historic release of the barrels, and yet we still saw prices move higher after that. So, insane range and basically ended up kind of where the first print was, believe it or not, for WTI.
Um, gas at the pump continues to move higher, approaching 370. I showed last week when we were at 350 that 370 was already the futures. Well, now the futures are all the way up here at four. Um, so we still have to go. This is overlaid with CPI year-over-year. So, we are going to see a significant rise in, uh, in CPI. Uh, and I'll show you, you know, where that's fitting in.
This is the future. So, this is the second, uh, contract of the Arbob futures for the unleaded gas. And you can see that, you know, the red line here is gas at the pump. Here's where we closed last week, up around $4, uh, dollars a gallon. So, we still have more to go. And this is not just in gas at the pump, which you can see here. Uh, and you see it for every different region. You've got lines building in China. All kinds of stuff. So, this is a global issue, which is why it will hurt the economy. Everyone has a different sensitivity depending on, uh, the strategic petroleum reserves they have, but also how much they're importing, uh, and how much they can control prices.
So, here's jet fuel. Um, I mean, you know, airline stocks have fallen significantly. I think people are still hoping that this just turns around and goes down. Um, I'll go through some of the podcasts that I listen to this week with people. Rather than focus on where it is today, you really want to spend more time on how long is it going to be elevated. That's going to be the impact on things. And again, as I go through this, uh, oil was not the story three weeks ago or four weeks ago when I started really highlighting the risk that was showing up in credit and credit spreads and also in turbulence.
Um, Canantraits is just kind of highlighting what I think is the hope out there by everyone, which is that, um, you know, until oil drops below 70 when the Strait is back in business, I, again, eventually we're going to get there. We always do. Um, the question is how long is it going to take, and during that period that things are, um, higher and we're seeing much higher inflation prints, we're seeing a slowdown in activity. Uh, what happens to private credit? What happens to financial stocks? What happens to the other parts that were already falling beforehand? This is not just one issue.
Um, Marco Papic and geopolitical cousins. This is the one, uh, right off the bat that I want to recommend for you guys. Marco's a friend. We spoke on Friday. Uh, we're going to try to do some things together in the future, uh, just because we have a mutual respect for each other, but also I think Marco is one of the more creative geomac people, uh, in terms of his thought process. He's a great historian, but I think he did a good job in this. And so this one brings in Ed Richardson, who's a shipping and tanker tanker expert. And he makes an interesting point. And rather than again get into when are we going to drop below 70 again and then the S&P can go higher, I don't want you guys leaving this thinking that's the way to think about it. Obviously, anything that suggests that a ceasefire is in play, you're going to see oil knee-jerk come down. We saw that last week on other things that went on. Uh, the real thing is what's going to happen to earnings in the economy the longer we sit at these levels. This is not a minor move in those prices that you saw. Uh, and it's not a minor move in inflation. So, we've taken out rate cuts. We're looking at rate hikes. All of these things matter. Uh, and these are not temporary shifts where you can just fade them. Uh, particularly when we've had this lack of correlation, which I'll go through, between rates and equities.
So what he says that's interesting is they've been hit, meaning shipping has been hit by one disruption after another since COVID, port congestion, uh, the Ever Given blockage, Panama Canal drought, sanctions on. He's basically saying we've been desensitized to geopolitical shocks, and we always assume that they're going to fade quickly and then just go back. And I think that is absolutely true. I think taco is a part of that too. Everyone just assumes that this just turns around the other way. The other issue is what she just points out. Hormuz is arguably the most dangerous choke point in the world. And I think that's what we've seen is it's not just a question of how desensitized people are. The complacency to the, the potential of us being there for, uh, more than a month, two months in terms of having prices elevated and things offline.
Uh, Jeff Curry gave an interview on Bloomberg. Uh, I've highlighted Jeff Curry a lot on here, and regardless of your, you know, views on Jeff Curry, because he's constantly, or most of the time, bullish on, on commodities, at least since COVID, he has been right, uh, in terms of supply demand, maybe not the timing of everything, but he talked a lot about what's going on. And he did talk about a couple things that are interesting. I'm not going to go, you know, uh, go through everything, but he did bring this up, which is, and that we've seen this already going on in in memory. Uh, you're going to see this consistently over the course of the next decade, which is this hoarding behavior. When countries and consumers stockpile oil, demand artificially increases. In the 1970s, oil crisis, hoarding added roughly 2 million barrels per day. Even if there's a day where things are better, I don't think people are going to have, uh, comfort that this is just going to go away. And when you have places like Japan and South Korea that are stockpiling fuel, and you've got lines in China, I just think that the underlying price for oil at this stage will probably stay higher because this disruption is in such an important place, and because we were already in a hoarding mentality of commodities because of what you've seen in some of the parabolic moves.
Uh, on Friday, uh, the US went at Carg Car Island. So, which again, as you read through it over the weekend, uh, you see how important it is. So, all of that's there. And I think what Warren PI's, uh, sent out this week, just as the week went on, this is the curve. So, this just shows that it's starting to be built into these. Each one of these is a longer, uh, duration. So, right here in terms of the, the discount, there was, this was being viewed through the 10th as a short-term disruption, and then as the week went up, we saw all of the backend contracts start to go up more than the front month. So, you started to get a scenario where, as Felix put it, the curve is now taking this much more seriously. So, again, by the end of the week, we had the oil market saying this is probably going to, uh, have elevated prices for longer.
We also saw the same thing in the TIPS market. So, this is now one-year inflation expectations in terms of the TIPS break evens went from down at two and change all the way up to 47. That's one year. You can see where, uh, we have, uh, all of the blue dots here. This is two-year yields across the globe, and basically this is taking the three-month range. So, every single part of the world has now built in that we're going to have higher rates. Uh, so, again, when you're looking at the equity market, if inflation's moving higher, we know that historically has not been good. We watched what happened with oil prices back in 2022, and the fact that they stayed elevated all the way into June when we, and that's when CPI basically peaked. Uh, the question is how long is this going to stay here? If oil stays, you know, higher than 90 for a period of three months, then we're probably going to have to continue to move two-year rates and 10-year rates to this point. 10-year rates, again, same type of thing in terms of the, the movement. So, uh, you're, you're getting this whole scenario that everything is being moved up, uh, across the globe in terms of yields.
So, inflation expectations and then pricing. We don't even have one cut built in before the end of the year. Now we're close, but we don't even have one. And then in Europe, uh, we've seen one of the largest falls in, in from, in six, the contract for Uribbor out to December, but this is the rolling sixth month contract. Uh, so again, a big fall. And you go back and look, this was the QE days when front rates even went negative. Since then, this is the period post-COVID where we get these shocks and they have an impact, and the market has not responded well when we've seen rates kind of move higher, even if it's only for a period of three months. Uh, this drop, uh, had moved up significantly, uh, the move index. So, we saw a 16-point jump in one day in rates vol. That was the biggest move since during, uh, 24, in September. So, we saw rates fall, shoot higher. Here's rates fall relative to the OAS for IG. So, we did see spreads widening in terms of the cash bond market.
And, uh, this is the sixth contract of the VIX, which I've said repeatedly, I said it this week on the subscriber webinar that I expect this to get up to 30 before this is done, um, because of private credit and because of the turbulence that we were seeing way before oil, which I'll go through, but if you add oil on top of it, uh, I don't think people should, uh, should fade this. Um, I think Charlie McGillicut, uh, had a good note out this week in terms of the pieces that I saw. And again, it gets back into inflation volatility. If we start to build in the possibility that we're going to have resumption of inflation, and that's it. It's the possibility we have to continue to move multiple compression into the S&P 500. This year to me was about multiple compression. I've said repeatedly, I think GDP is going to be fine. I think earnings are going to be good. Um, I think in both cases we're going to see it. Even with oil sitting up here at 90, I still think nominal GDP will be strong. The AI trade is going to continue. There will be fears along the AI trade, but I thought this will be a multiple compression year where best-case scenario, uh, you know, we'd we'd go up, uh, you know, whatever, close to close to flat, up a little bit, and most of the multiple compression would be happening in technology. The global market would outperform. And the reason is everything that we're seeing, the disruption from AI, the unwind of the credit, uh, situation from private credit, the fact that the labor market is going to worsen this year as the AI fear, as the AI reality comes in with the agentic AI, and now what you've done is thrown inflation or oil up and inflation fall back into the equation. It's a very different world than when we started the year. So if people are telling you everything's the same and they're saying just step it and buy it, this is just like last year. Remember, Liberation Day was self-created and then it was quickly turned around the other direction, giving the market time to prepare for it. Right now, we don't have that time. This kind of came out of nowhere.
And as I highlighted, this is the end of the year. So, this is December of, of last year. JP Morgan said gross leverage at a 5-year high. Goldman showed global gross leverage at 285. So, the same level as JP Morgan. Morgan Stanley said gross leverage for US hedge funds was only higher 1% of the time during the last 15 years. We are seeing leverage around historical highs on our books. And then UBS jumped in with the same thing. So that's where we started the year. And it makes sense. Everyone expected this to be a great year for earnings, a great year for global earnings because revisions were going higher. PMIs were being built in to go higher. GDP was going to get a surge because of the one big beautiful bill. Again, when you have sentiment that strong and you have leverage coming off a time where we really didn't have much turbulence, um, you end up in a difficult situation.
So, the systematic community for CTAs is definitely getting closer. We closed below some of the medium-term levels, so we've definitely seen this come down. This was from March 10th, uh, on Tuesday, but I think we still have a lot of room for the CTA community to come down. I want to make sure you guys continue to focus on this because I really think people are forgetting about it, and I don't want anyone who's watching my videos to forget about this. Um, I think economists are missing the boat on this. I think this is an obvious trend. This is the year-over-year employment, the white line. We're at zero. No jobs created. Now, this is all payrolls. So, this includes healthcare. Without healthcare, this is a big negative number. Healthcare is not going to be replaced by AI. The AI jobs disruption is happening.
This is nominal GDP. So, you can go back over the course of the last 40 years and just see the relationship between nominal GDP and year-over-year payrolls. Well, this is broken down. It does matter for the K-shaped economy, especially when you move oil prices higher. Just to make sure that people see the other end of the side, because we had all these software engineering job openings and people getting exciting saying that, again, guys, like the job situation has been deteriorating. And just because the unemployment rate is not shooting higher, which I don't think will happen, doesn't mean that if someone loses their job in finance, job openings which are coming down, which I would argue are the most important for the economy, because if job openings in finance are going down in knowledge workers, this is not only important, but this is all college educated. We're collapsing. So, if someone loses their job here and now all of a sudden they're riding through Uber and they're working at Whole Foods and they have a college education, that doesn't mean that just because jobs, uh, you know, people aren't getting fired or they're not claiming insurance. There are jobs out there. We have a labor shortage, but it doesn't mean people are undermployed and upset about it. AI agents could push grad unemployment past 30%. This is the CEO of Service Now, who knows something about AI.
Meta, uh, this morning, Saturday, I'm going to get more into Meta for a while at the end. Reportedly talking about 20% of the company. Uh, this is a cash issue. The reason this is important, this is going to be growing across the market, in my opinion. AI costs are increasing. When you see memory charts that look like that, when you start seeing the hardware shortages, the CPU shortages, the optical fiber, everything that's going on is going to lead to higher and higher costs for AI, especially since a lot of the frontier model companies can't get the data centers built in time. So they're raising price because they don't have enough capacity for everyone that's in. It's going to lead to more AI cost, and the way to pay for that is with labor.
Warren PI's another great chart. AI's impact on the labor market is no longer theoretical. What he did here is go through and take 30 industries that are AI exposed. That's this line. And then 30 industries that are non-AI exposed. So a lot of this is going to be healthcare, but it's also going to be the driving jobs and things like that that are not replaced as of yet by humanoids or autonomous vehicles. So what you have here, this is the gap between the two in terms of how much they've moved away. So again, you have to look at this. AI is having an impact on jobs, and it's having it on a lot of college-educated jobs. It's having an impact on graduates. It is definitely helping to have consumer confidence be at lower levels. But this is before the agentic world. And that's the thing I want to make sure. If there's one thing that everyone has screwed up, it is the exponential nature of the progress of AI. How fast it accelerates. How many people miss the Micron trade, how many people miss the Nvidia trade, how many people miss the Corning trade, how many people, when this stuff goes higher, the dollars are massive, and the moves are exponential. We are entering the agentic world now, which means job losses are going to worsen from here. You have to start extrapolating things that are moving faster and farther than they ever had.
So this is what I want to make sure that I start to fade in. So this is the red line here is the S&P. The white line here is the financial stocks, guys. So when you know, if we go back to Cananteritz talking about oil, a lot of people are starting to talk about, and he didn't do this, he's just saying that right now oil is driving the market, which is true. But what I want to show you is this weakness in financials was happening way before oil. The S&P 500 was here when this broke the 200-day moving average. Now you've extended this even further, and the 200-day moving average is turning down. I've highlighted repeatedly, I did on the webinar, the two webinars I did, how bad it is when financials are below the 200-day moving average and how bad it is when the 200-day moving average is turning down.
So, I want to go back to when this all occurred and just remind you that we also have had turbulence since February 3rd. So, I updated the turbulence model this week. I posted an X, um, that it went into crisis mode. I showed everyone on the webinar this week for the subscribers where it was and what went on, and that was on Wednesday. We're at a point where we've had so many turbulence days that you have to understand asset V is increasing, and that's the main point that comes from the turbulence model is asset volatility is increasing significantly. Remember this measure of turbulence. Everyone forgets this, but this was on February 12th when we were seeing stocks gap down. This is not about oil, guys. Oil had barely moved at that point. And here's what we had. 115 stocks in the S&P 500 had declined 7% or more in a single day over an 8-day session. 115 of the 500. Last time we saw anything like that where we were near the all-time highs was back in do. This is not something to fade and just say this is all related to oil. Here's where oil was during my turbulence model, right there. Here's where it was during the time where we were getting what I just showed you. Financials broke the 200-day moving average in here. Do not get caught in the belief that once oil goes, uh, goes back down for any amount of time, that that means that everything is going to be stable.
Just look at these charts and how brutal they are. This is private equity. This is Blue Owl. This is Jeff. This is Salesforce.com. Don't forget these. This means something is going on. And to me, this is not a bearish, the year is going to be horrible. I do not believe private credit, as we go through this, is going to be in this, uh, mindset, but I do think people are going to get scared of it. And I think the Fed and the central or the or the Treasury Endor are going to have to do something because I've never seen in my career a credit situation that is deteriorating at this kind of pace that just ends on its own. There always needs to be some sort of liquidity facility. And this is during an election year. So as credit worsens and retail is trapped, just remember this.
So Joe and Tracy from Bloomberg, the 2008 meme continues to grow, and people are starting to talk about this more and more. Hartnet talked about it. It's starting to look like it. That's because when we started, when we were seeing financials basically go through the same thing or a similar thing that we're seeing with these credit funds in terms of the private credit funds, uh, we also had oil heading up to $155. Okay, it's different this time, but there's a little bit of rhyme that's going on. And again, this chart shows you how bad it is in terms of just how far it has gone. And again, this was not the case at the beginning of the year. So at the beginning of the year, son, this is all accelerating not because of oil. It's just financials. And it's not just inside the private equity names.
Now, to show you how, uh, I'd say irregular it is for this to happen, for the first time in more than 30 years, financials have been in a 10% plus correction while the S&P wasn't even in a pullback yet. I think a pullback is designed by a 5% fall. Uh, but that, you know, we only, in the entire last hundred years, I think there were about eight that have occurred. This week, Cliffwater had redemptions to 14. They only gave back seven. Morgan Stanley, uh, had requests for only 10 for 10.9. I think think they gave back five. I don't remember what the exact number is. Um, then later JP Morgan restricts private credit lending and it marks down loans. Back leverage is another pain point for funds, meaning the back leverage is where the banks are supplying some things. Before JP Morgan's decision to curve some lending, has the industry on high alert about threats to back leverage. This is what was going on in '08, guys. This was happening more for, uh, margin requirements for hedge funds while the banks were under pressure. Uh, when you start seeing the banks not, uh, provide kind of the back leverage or the ability for them to be there, you end up in a very dangerous situation, particularly when they start marking down the loans on their end. What ends up happening in a deleveraging event where credit is getting hit is when you can't sell something because there's no liquidity for it, you go sell what you can. And just remember when Blue announced that it sold $1.4 billion worth of loans at 99.7, I think if they would have sold them at 85 cents, there actually may have been a little bit more hope. But when you're selling things at 99.7, you assume you're selling good stuff that people will buy, and they won't buy the other stuff. And the more that you sell that stuff, then the other things just continue to go down because eventually you will need to do, or those funds will have it.
So here are the numbers in terms of the exposures to private debt. And again, this is not a systemic number. The private credit situation to me can be controlled, but it's going to take a liquidity facility to stop this trend, especially if I'm right about what's happening with the AI disruption. Deutsche Bank said that came out said they had 30 billion exposure to private credit, and the stock was down 5%. And this is not new news. They've disclosed this in prior years. Goldman, someone you don't think about involved in this. Well, they were down. They're 20% off the highs now. They closed below the 200-day moving average.
So, this is entering a period where each one of these. This was the European crisis. This was the oil crisis. We needed Draghi to do something. We needed the US Shanghai Accord to happen. 2019, believe it or not, this is when the Fed pivoted. So if you go look, the Fed was hiking and then they stopped while this went on and we got COVID, massive liquidity facility. Then you have this where it ends with the Fed pivot here. This was during SVB. This was Liberation Day. Every one of these guys had some sort of help from the government. Just remember that. Every single one of the Goldman Sachs moves. And if I would have taken it back further, we've been in the great financial crisis. OFR, if you want to go read an eight-page report on the size of it, what's going on, this is from this week, measuring the counterparty exposures to private credit. Again, guys, this is a story that's growing, and I wanted to make sure because I've talked about long duration assets. This is not just about private credit.
Ontario Teachers, one of the most sophisticated and best pension funds in terms of of risk out there, um, their private equity was down 5.3%. Lowering the valuations, just remember that we have not just that, we have a lot of different areas outside of private credit on long-duration assets that have been under pressure for a while. Um, this is the thing that I'm surprised people aren't more concerned with at this point.
So, ever since this broke away, so this is going back to the financial crisis. This is financials relative to the S&P in the white line, and you can see it's about to make new post-GFC lows. The yellow line is tenure rates. So, when the Fed raised rates, think about all the times during here. You've seen a Japanese 30-year yield. You've seen a guilt 30-year yield. You've seen all these bonds around the world, and everyone's been talking about the end of the world, the basis trade unwind, and everything else. Well, here we have asset V expanding. We have a serious credit situation that is much different here. This white line here again, SVB rate shot higher. If rates start shooting higher because of inflation expectations at a time when we, this time we were creating millions of jobs here, we are now at zero. The K-shaped economy has created an issue. You cannot forget and put that into context. Yes, we have S&P earnings growing, but again, we're seeing multiple compression. And this is why I say this is not a minor thing to deal with. We have a break going here, and that means bonds right now are not a safe haven. They were heading down until oil went down. And that's the issue is right now we've got a situation with inflation that matters.
Now, with all that being said, everything going on for the week, these are the returns around the globe. So, you had oil trading in a $40 range from 85 to 120, and yet somehow things are not moving. So, that has people, I would say, hopeful. S&P down 1.6%. Not even a move to notice. NASDAQ, not even a new move to notice. Russell, not even a move to notice.
Here's the breath. Now, we are oversold in breath. I highlighted in the subscriber thing a model that I built that basically takes into account the VIX, the VIX curve, um, breath, and we, we're getting closer in breath in terms of the RSI that is over getting oversold. But what we haven't seen is a capitulation day yet. So, this is the net breath for a given day. We finally got a minus 1,000. There's about 1,500 or 16,700. Uh, so we're getting closer, but you can see how many times Liberation Day when we usually make lows. Um, this was the low on the yen unwind back in the summer of 2024 and early August. You can go back over time. All the inflection points have historically used with a capitulation day, which means everything goes down together. We haven't had that yet. I still think that's likely to happen, and that's where you should be looking. Also, the put call ratio, it's not budging. Now, I think the most important thing will be if oil at the end of this week, if the Strait of Hormuz is not open by the end of this week, I'm guessing that one of the reasons this is down here and not moving higher, which is shocking. This is the five-day average of the put call ratio for the market. Um, and I put this up to just show that Liberation Day, we get a spike here. Uh, go through all of the big lows and just see if we get anything normally where you're getting some activity. We're not getting any activity in put call. We've got triple witch this week. That may be the day that starts to scare the hell out of people.
I showed this on the webinar subscriber this week. Factor val much, much higher. So, this is a, a, uh, a composite of factor VS for three separate ones that I keep. It's overlaid with the S&P 500. These is a 60-day V sitting up here. And this is a 60-day V on the S&P. It's just not moving. So, this has really been isolated. Uh, the MAG 7 broke below the 200-day moving average for the first time since there. And I just want to show you, like, this is a big deal. You're talking about over the last six years. You had one break during '22, horrible time for the market. You had another one here in March. Horrible time for the market, until we made the Liberation Day lows. Well, you got the MAG 7, uh, going down, and take your pick. Every single one of them is underperforming the S&P. Every one of them is underperforming the S&P. The hyperscalers have been below the 200-day and they're about to take out the lows that they just went through.
This is where my focus is, and the darling for this week is Meta. So, the Meta chart looks horrible, but I just want to take you through this. This was a big earnings beat for Meta, and they basically peaked, got above it a little bit. This is the, the next earnings report. This one was bad. Stock sold off. Got another good one. Gapped higher. That was the high, and we've continued to move lower. Something stinks in Meteville. Um, just remember they went on a spending spree last year. $14 billion dollars effectively for the CEO of scale, uh, Alexander Wang. Don't forget October 21st, they announce their massive data center is going to be financed by Blue Owl through an SPV. Zuckerberg and Alexander Wang clash over company's AI future. So here he is. This guy's hired for 6 months. Already seems to be some internal problems. Then later, Wang says micromanaging is suffocating. This is later in the month. Then you have one of the most respected AI people in the world, Yan Lun, who leaves, says there's going to be more because the new 29-year-old AI boss is inexperienced and warns of an exodus. Ernst & Young raised red flags on the data center accounting after the last one. You saw what happened on that. Um, they had a big gap higher. Looked good. Auditor raised red flags. Stock hasn't been up at that high since the basically the opening. Uh, on the earnings call, I'm not going to read this whole thing to you, but basically this people were worried about this whole situation as it was. They're not liking what they're hearing. They're wondering what's going on in terms of getting these models out. They're spending all this money on AI compute, and again, they're putting their balance sheet at issue. So you've got a lot of people freaking out on the free cash flow situation. Meta struggles with memory shortage despite launching its own chips. The reason I bring this up is none of these are targeted, meaning they're not part of a bigger story. I'm bringing this up because along the path of it building a top, there's been a lot of stories out there related to the company. It's one of the beauties of having AI and having my AI agents run around and pick a topic and have them go through things and show me all the ones that are fitting a story that could be matching with the price. What's the narrative associated with the price? One thing at a time doesn't matter.
Now, this week, delayed launch. The Facebook parent is considering temporary licensing Google's Gemini models. So, they've been spending all this money, hiring all this talent, and then this morning again, like I said, they come out and say they could be laying off 20% or more of the company. Uh, this would be a free cash flow issue. Now, it's not just them. And I think as the year goes on, if we don't see adoption pick up at a faster pace, if the costs keep going higher, and the data center delays continue, you're just running into a problem here with the free cash flow story. I've shown it before. I showed Dennis to Busher's work, uh, with the cash flow of, of the S&P. I think this is the multiple compression story that people have to pay attention to. Oh, and oh, by the way, in there it mentions stock-based compensation for these companies as well, where their cash isn't as good. I just think you have to pay attention.
At the same time, it's an election year, and one of the stories that's been growing and it grows daily, how much people hate AI in the US. And if you look around the globe, the US hates AI more than anyone. And there's a lot of reasons for it. The jobs, the data centers. Uh, I'm going to keep showing Josh Walsh. Josh Wolf, he's a smart guy. He's really out there basically on the negative side. He sent out a letter, um, that I've highlighted. But he's talking about the massive buildout of AI data centers may be overextended. And I think there is a legitimacy to at least talking about the fact that a lot of the data centers may not be built. And if, if that's the case, you've got a lot of stranded GPUs that are out there right now. One of it is moving to localized edge in inference. And I'm going to be doing more edge inference stuff because I do believe that that is a major investment side away from the GPU-centric side. And there's a lot of companies that I've highlighted in some of my work, uh, that I did, and I included as the opportunities this week on the subscriber one. For those of you who have not subscribed, you should go give it a shot. Go through the thematic ideas. I'm not only trying to help you make money by not blowing up right now as this stuff unwinds, but also being in a position because I think the second half of the year, if not the next nine months, whenever this we get reach this point, I think I already have laid out the playbook on the names that I want to be involved in. Right now, I'm long VIX personally, but I'm also still long semis. I'm still long commodities. I'm buying more of those and Bitcoin. I'm waiting as well. And I put out a substack on this KKR, which got into this, I think, at about $300 million. Um, if I remember correctly, they're selling this at at three billion. Um, you just start looking for smart places that are deciding that, uh, it's time to sell. Amazon was out raising over $30 billion this week. This is the euro portion. Again, why do people hate AI so much?
All right, it was a good week for AI, though. Um, if you haven't used 5.4 yet, it's fantastic. Um, again, I, I say this about every model when it comes out, but I really do enjoy using 5.4 deep research. I think the work that it does is fantastic. Its benchmarking for GDP val is now up to, I think, 80 plus percent. And what that means is of jobs that it can, it's it's up to GDP at 80, 80 plus percent in terms of the ability of replacing jobs with inside the knowledge work side. Uh, it's a big week for these types of products. A big week, a lot of the companies are clearly on the enterprise side. I've had Anthropic brought up to me on multiple calls, and multiple trips in terms of people bringing it into their firm. If you guys still have ChatGPT Enterprise, you have to beg your, your, your tech people to get you, uh, clawed at this point. Microsoft announces Copilot Co-work with the help from Anthropic. So think of this as again an AI agent. So the agentic stuff is starting to build up. Google's, uh, workspace brings in CLI. This is getting great ratings relative to model context protocol MCP for Claude. So CLI, which allows you to do things between your docs, your Gmail, and your sheets. Perplexity takes on Claude Co with personal computer. I'm going to show you something that I built with personal computer this week. Anthropic is in talks with PE firms including Blackstone about an AI-focused joint venture to sell it to their port companies. You're getting everyone's focused on adoption.
Uh, one podcast to listen to this week, uh, Facts and Feelings with Ryan Dietrich. Uh, he had Kai Woo on and Steve How. I've known or I've met, uh, Kai Woo, very smart guy. I was interested in hearing what he had to say, and I thought this was an interesting take, and this is really my, um, issue right now with anyone on the adoption side focused on these guys being able to get the revenues that they want. The timeline of AI adoption is determined less by the technology itself and more by the friction in the real world. Even if AI capabilities advance rapidly, the economy moves at the speed of the institutions' infrastructure and human behavior. That's it from a nutshell. So institutions, anytime there's an AI committee, it's going to take a long time to get AI adoption. The infrastructure, you got to spend a lot of money and you got to build out, not just on the data centers for the cloud, but if you're going to have on-premise to fully use the agentic side, we don't have the infrastructure, and the cost is going higher. The human behavior, humans don't want to use it. They don't want to lose their jobs. That's why they hate it. I've witnessed this firsthand. Several forces create their friction, and here they are. So again, I think this is the case. So there is a large gap between what technology can do. So what AI can do right now, the GDP val says it can do almost every single knowledge work job, especially when you connect it to openclaw or any agentic side that's going to be running 24 hours a day. The level of disruption depends on the relationship between technological speed and the, and these frictions. Um, if technology advances faster than institutions can adapt, and I would say this is exactly where we are, the disruption is severe. Companies cannot adjust quickly. If you don't get the profit margins growing rapidly, the question is, if you have to overspend, will the profit margins actually come down? And that's normally what happens with technology. Or do you fire people quicker, even before it goes through? That's what we're in.
Uh, Frontier Foundry, uh, wrote a good piece on this in terms of the build verse buy. Uh, for those of you people who continually want to tell me about buying software for value, people that need to step in and find some names, I completely agree. There are some names in software that you can go buy. But as a general rule with software, I think you're going to get hurt more than you like.
Openclaw. This is for those of you, and I, I've had people say there's nothing to do with it. Um, so when Stanford and Princeton basically drop lab claw and say it's a full AI co-scientist, we're not just talking about people sitting at home. Um, learn OpenClaw, get it going on your computer. China becomes agentic AI's biggest lab with OpenClaw stampede. The craze sweeps China as authorities seek to clamp down on the state-run enterprises from using it. Andre Carpathy came out with a new auto research that lets you run hundreds of AI experiments a night. You should read about this as well. Uh, this just continues. AI power users are rapidly outpacing their peers. This is a very scary thing. AI power users, which I consider myself. I use this all day long. I can do things with it only because I've used it for so long. You're getting to the point where this number is growing so rapidly for people that just use it as an answer bot. That is the reason why I did the video series. You have to start being comfortable with using it for everything.
Uh, Lisa Su spoke at the Morgan Stanley conference. I wanted to include this this week just because it matches up for the subscribers with something I showed in terms of the advanced packaging side and the names. Um, this is an important thing, and I'm going to write a paper on this, but CPUs are cool. Again, Intel and AMD reporting spikes in CPU demand due to AI agentic AI shortages. Think about Mac Minis. Think about all of this stuff. So when you get into this, this, uh, story from Tom's Hardware, supply shortages are starting for server CPUs. We're also, and that's in China. We're also seeing a spike in demand for high-end Mac Studios and Mac Minis. I have each of those. Um, the rise in popularity of Claudebot is what triggered it. Um, I just bring that up because again, if you're not using it and you're just sitting there waiting for it to be a software tool or waiting for it to make sense and how, how you can use it, I'm going to repeat it again. Go watch my video, put in the reps.
Um, I just wanted to show you something that I decided to do. So, basically what I'm going to do, uh, with my new Mac Pro book is I'm going to try and use, uh, Karpathy's open-source, uh, GitHub auto research that he put out there. And this is basically me asking how to set it up with OpenClaw to be able to run it. I'll be using DeepSeek R1.
I just want to show you what goes on. Oh, this is the personal computer. So, this is one of my thematic baskets, uh, for the subscribers. You've seen it's there. I just want to show you that what I did was I built this in, let's say, five minutes in personal computer and Perplexity. I asked it to build me a dashboard which has real-time prices on it, has all of the technical readout on the names based on the formula that I put in. Uh, it has the recent news, it has the bull bear. I can click on any one of these and it immediately brings this up. This is what you can build in a matter. This chart is live. Uh, so I can go to this site. I can do it whenever I want. I built that in literally five minutes.
Uh, open claw effect. I released this this week. Again, this fits in with the architectural side. We're moving away from GPUs. We're moving more towards CPUs and advanced packaging. This was again to emphasize those names. And the names you want to be buying into any dip that happens here. One of them reported last week. Uh, and again, I referenced that on the subscriber list. This is all related to this paper that went out. So, for those of you who are subscribers, these papers are what you want to read. And if you input these two papers along with the names and basically say which one at this point are the ones to buy based on the earnings commentary or whatever, you can go do your own work on it. That was the whole point of the video is that you can take my research, you can go upload it and you can go through the same process I do. And if you want to come out with different names, come out with different names. If you want to say only show me names in emerging markets, only show me midcap, only show me small cap, you can do that.
So, I'm starting to focus my attention on crypto. This is the final part of the video for this week. Uh, this I released on Substack and I feel very strongly about it. I have waited patiently, uh, for the moment that the network effects would be in gear for Bitcoin, that Bitcoin would separate itself from anything built on code with the Mag 7, the hyperscalers and the software names depressed for reasons that I think are structural. You're going to need growth assets. That's where Bitcoin fits in because the financial guard rails of the future are all related to the digital economy. And I know people don't understand how Bitcoin fits into that. I know they don't understand when I say it is the purest AI trade, but this is it. My whole thing, if you go back over my Substack, is that AI will destroy all public companies. Every single one of them. There is not a single one in my mind that a decade from now won't be disrupted by AI. Right now, it's all based on code. It eventually will be based on humanoids as well. It'll be based on quantum. Everything heading in that direction is getting there.
So, this piece goes through why the private credit side is basically setting you up for Bitcoin. Because every single time that we've had one of those things I showed with Goldman Sachs, I want you to go look at what happened to Bitcoin in the following three months. Here is the chart of the financial sector, which I showed with the beginning to Bitcoin. So, right now, Bitcoin is down here. Financials are going to go through all of the points. Fed pivot right here. Fed injection right here. The pivot here. Sorry, the pivot was here. You can go through all of the times in terms of what's gone on and you guys will see that Bitcoin has moved over time with it. We also on a weekly MACD, we're sitting there. We're now starting to hook up. All we really need is a move higher. Maybe it moves lower before because we get an unwind in the S&P, but it's been acting well for the last, uh, two weeks, and that's because software has found a bottom at this point. Uh, I think the upside in software is limited, but I do think that Bitcoin will continue to trade with it.
The Stripe annual letter came out this week. I highly recommend this as a read. Uh, it's more important than any of the financial companies because the financial companies, you're looking at the old industrial world. For this, what you're getting is companies talking about how stable coin payments are doubling to around 400 billion. That gets into Bitcoin. We're preparing for a world of massive stable coin adoption. Massive stable coin adoption. The five levels of agentic commerce. The idea that your AIs will soon be buying stuff on your behalf. Human beings will not be making purchases five years from now. Agents will be doing the majority of them. You have to be involved and see what that means. It means the total eradication of the middlemen. That's why when these names are falling down and people think that they're buys because you look over history and say it's cheap. When you start building in that three years from now, five years from now, where AI will be, where stable coins will be, what will go on, where agents will be, you go stable coin firms big on AI agent payments that barely exist. These are all news this week. Dollar peg stable coin surged to 313 billion in the risk-off pivot amid the US-Iran conflict. Dubai stable coin demand is going parabolic. I'm in Dubai. What I'm seeing this week is insane. Flooded for this again. You're just at a very different time.
But let's leave it on this note today. If you don't believe me, Stan Druckenmiller said he expects global payment systems to run largely on stable coins within 15 years, calling them more efficient, faster, cheaper than the current infrastructure. If you go back to what Kai Woo said, the problem for the disruption is that all of these organizations that have spent billions of dollars on people and billions of dollars on old technology to get them to be modern and to be able to deal with stable coins, they have to fire all of the people and somehow or another be able to move to the infrastructure that is today. But the problem is, as Kai Woo said, the progress in AI is moving faster. So the benefits go to the smaller companies and in particularly the startups. That is why I keep telling people use AI every single day. Go watch my video how to do that every single day. Come up with new ideas. Listen to things. Upload this and say how do I make money off what Stan's saying in here. You'll be shocked at what you're able to do.
All right, guys. I'll see you next week. Uh, enjoy another volatile.