New All-Time Highs for Stocks, Gold Cracks, and the Worst Jobs Market in 50 Years

Back again. New all-time highs for the stocks. Momentum shock during the week, especially taking down gold, but other parts of the market. We've got Shei Trump this week. Goldman says it's the worst jobs market in 50 years, and I'll go through why it is and why it's uh going to be this way for a long time.

Uh, most important week for the year is coming up. You've got uh the bulls verse bears and AI verse tariffs. uh concentrated parabolic charts, AI fears continue. We're going to go through the Carpathy vers Altman case, Tesla reported ASML. I'm going to bring you guys uh some more details on something, a stock that I've talked about and just a company and how important they are.

Uh, this will be the lesson one on starting your AI journey. Um, I've been doing this for about a year now. uh my ultimate goal uh was to turn my content into a business both for consulting, research, asset management and that's what I'm doing and that continues to move forward.

Uh, but one of the things that along the journey people have requested is I'm also going to be starting more to help people in terms of how to think about what I'm doing here and again the name of this uh and what I do is the AI macro nexus which means trying to prepare people both for what's happening in the traditional world which is moving in the slow lane that's the bottom end of the K-shaped economy versus the things moving in the parabolic chart side the AI and crypto side.

Uh, and if you still need to go from the slow right lane of the traditional finance world into the left side, the fastmoving lane, I'm trying to help you be able to get in there before those two worlds merge. The people that are bearish, that are perma bears, that are telling you the world's going to end, they are saying that from a position, and trust me, I know they don't need to move over into the left lane. They've already made a lot of money. That's why they're focused on the bare side because it's too easy to make money in parabolic charts even if they correct 40% and then make new all-time highs which is the way Bitcoin moved for whatever and now you've got a lot of stocks doing it.

So I'm going to start lesson one today out of request. The only thing I ask of you guys now the people that have been um I've had a lot of reach out from colleges. I'm going to keep doing that stuff in terms of doing demos. Uh, subscribe. I actually have heard people say this over the year, but I'm telling you as I start going through it really helps me for you to subscribe so I can keep doing this.

Uh, but also forward this to your family, friends, kids. The AI journey is for the people that are going to need to do it. Remember, the world's going to merge into one in about 5 years and it's going to be a very very dystopian five years in my view. uh as we get towards humanoids and we go through knowledge workers and the jobs market is disrupted, you need to learn how to embrace AI and use it and be an entrepreneur uh make your bed is kind of the theme that I'm going be running with. I'll get into that and then I'll talk Bitcoin stable coin and everything happening with Argentina this weekend.

So, the S&P makes new all-time highs. uh I still remain in the same thing which is that can turn right back down again and fall further if she Trump falls apart and the tariffs actually hit in. So just remember um even though the probability is low, I still think this is an asymmetric event uh even though the highest probability is we just have a truce and we kick the can down the road. I do think any tariff thing that gets put on hold for a period of time is going to unleash a lot of buying and a lot of risk assets. And that's because unlike in April and May, we have the spending is already happening. The big beautiful bill is done. This is kind of like 2018 for me for the Trump uh first term which was 17 was the tax year everything ripped. Then 18 was the tariff year and the market fell and was weaker. Even though the market didn't fall this year, trust me, in my opinion, this has been a lot weaker than it would have been given what's going on with earnings and the spending that's about to come, but also the fact that we're cutting rates.

S&P was up 2% last week. IWM was up two two and a half% so it outperformed again. QQQs were up 2% so it was across the board last week. Um, I did mention momentum was hit. Goldman pointed this out. uh really again its momentum. You still had the AI trades working particularly on the semi side but it did impact gold. You had a uh the biggest standard deviation uh move fall in gold. Take it back to liberation day. I do think there's a risk further down for gold uh to retrace the entire gains from the summer if there is a positive outcome in China and the US.

I do think uh people have minimized and again remember every macro person that has been bearish bonds, bearish AI has made money in gold. They talk about it every day. There's not a single podcast that they don't say, "Well, I like gold and xyz." Or if you want to, you know, play that AI is going to disrupt the world. You'd be long gold. The fear in my opinion for everyone who's in this is retail's been involved as well. Retail has migrated there as it does because it's been working. And so you've had retail inflows that have been strong and I think we need to have some at least uh consolidation, but I would say if there was a China US deal that occurs this week that can be viewed as something structural, I think it's much bigger and better than people think. I'm not going to go through all the details on it, but a lot of this has to do with fears over Taiwan. A lot of this has to do with fears that there's no way to solve this, which has been one of the perma bear stories that I've heard repeatedly, even at dinners this week.

Um, for the market as a whole. I just want to make this point because young people gambling, it's a warning sign. It's not a warning sign. It is part of this culture. If you guys haven't noticed, we've got gambling arrests in the NBA. Gambling is part of society. It is an entertainment. It is going to be this way. It is never going away. DraftKings was not allowed in sports here. This is the PCE dollars spent per year in gambling. It is going parabolic just like AI. It is going parabolic. It will not stop. Gaming is coming. Everything is coming. So remember, gambling is part of the market. And retail has assumed something that the government won't allow the stock market to fall. When you have something that goes up over 80% of the years, don't you think retail is going to make money if they just get involved in this? It is the institutional side that has hedge funds and then the long only funds have been replaced by cheaper fee funds like ETFs. So, everyone who's talking about this that thinks there's going to be some kind of correction, the markets have changed and gambling is part of it, but the government is debasing and they're trying to make things go higher. And now AI is making it go higher.

Ups and downs in the Trump she stuff. I'm just going to go through the stories that started, but I will put this. Analysts expect any agreement reached at the summit would simply dial down tensions rather than been a sweeping accord. if anything happens that is viewed more favorably and since there's an 80% chance of a tariff agreement which is different than a trade agreement but let's just say the tariffs to me are the most important thing he has threatened tariffs he has used tariffs all year tariffs have gone higher than what people thought would happen and yet we've not seen the impact I think if any tariff situation comes down lower than where he had them before I I think that's going to be a uh a liftoff point for stocks.

Um, meeting could happen, may not happen. Uh, then we've got uh US considering curbs on software. That's a big deal. So they're still doing tit for tat and still threatening negotiation going in it. She Trump can clear that up. Uh, I do believe it is uh mutual mass destruction. Uh, it's not rare earth wins. This is rock paper scissors. I'm going to bring up ASML. Lithography wins too. uh if China makes a decision to invade Taiwan, it would make sense based on everything I've shown over the last that it happens sooner rather than later, especially at a time when the US is out of ammunitions um and doesn't have any rare earth. So I think any trade agreement people are underestimating the impact that it would have and I think it would be a very very big deal for the market. Uh, China even threatened the farmer leverage as a nuclear option. So they both have the ability to shut each other down and that's what makes it difficult.

Profit margins given concerns in the market about tariffs and higher costs. So again about tariffs and inflation. I've heard stagflation so many times and everyone that said stagflation said, "Yeah, the way it'll show up is in profit margins." You're all wrong. Profit margins are not only sitting above the 5-year average for six trades quarters. You can't even see any change in margins. So I'm sure there's an impact that's happening, but it's being masked by AI.

The jobs market. This is Yan Hatsius. This is the worst. This is this week. This is the worst the jobs market has looked outside of a recession in 50 years. This part I think is just a bunch of crap. Um, on the employment side, Hoti has noted the labor outlooks in surveys such as manufacturing had fallen well below the index point. Agreed. I've shown every one of these charts, including the tightness showing up on estimated unemployment rate, estimated job openings, the conference board labor market differential, and New York Fed finding job expectations to name a few. All of that have shown it. Now, he continues on, America may be in an era of jobless growth. So, jobless growth, the way I'm going to say this is jobless growth, but with GDP. The one thing he hasn't gone through is this. I've pointed this out. The job growth outside of the healthcare industry has turned negative in recent months. Outside of healthcare, we have been losing jobs. The reason I bring this up is every time that you hear the new excuse that no, this isn't true. The Fed shouldn't be cutting rates. They should absolutely be cutting rates because AI is destroying the jobs market. Not firing people, but making it impossible for people to get hired. And I'm going to show more proof of that. But every one of the times that people say, "No, no, no. It's just, you know, between the immigration and all of the stuff that hot it matters or surveys. If there was a situation that we aren't hiring as many people because the supply of labor has gone down, why are temp jobs collapsing? Why are the surveys in terms of hiring collapsing? Everyone ignores those things. Those things have been correlated to the unemployment rate. The reason the unemployment rate isn't closer to five is solely because the labor participation rate has gone down, which means that includes the people who need to work and that's been declining since way before immigration reversed. That was actually happening as immigration was fine.

You have a situation where lending I didn't talk about that one. I got to make sure I bring this up. We have a K-shaped economy. As people lose their jobs and as rates are still sitting up at these levels, we are starting to see the tide of liquidity has gone out, it's getting harder and harder as you're getting these bankruptcies. So, yet another one with inside the car side prima lens this time. This is this week. You're getting more and more signs of this. The delinquencies are going up. This is the K-shaped economy. This is the we need a new type of politics in New York City, so we're going to vote for Mandani. You've got people that are having trouble. They're living paycheck to paycheck. 67% of US consumers are now living paycheck to paycheck according to a payment survey. Carvana stock falls sharply. The auto market is flashing water warning signs. Remember a lot of autos were at the 5-year mark from 2020 when money was handed out. You're starting to see the term show up in the lender's default. Uh, lenders are ramping up bankruptcy protection measures as they position for further distress in the wake of high profile.

So on one side again you have the left lane which is AI and crypto violently moving forward at an pace that is exponential. On the right side you have this slowm moving what used to be a recession that would be causing trouble. The AI situation is hurting the jobs market. The jobs market and the deflationary side of AI is offsetting what is going on with tariffs.

Equipment share is said to weigh US IPO. This is a company that was said to weigh an IPO as of August 20th. Equipment share bonds sync as ousted board member alleynches fraud. This is the same company. This is yesterday. Blackstone says error of bumper private credit returns. This is no joke. The error of lending is over.

I'm not going to go through this today. I just want to say one thing to you. Exponential innovation. 5 years from now, the entire economy is merging with that fast lane obliterating everything in the slow lane. So if you're waiting for a recession, if you're waiting for something that the slow lane has, recessions and credit events are all based on time. We are in exponential innovation. Meaning today by definition is yesterday, every single day. You can ignore it. You can pretend it's not happening. But the reality is it is the credit situation. You will be having bankruptcies. It doesn't mean we're going to have a collapse. It just means whatever returns you thought you were getting in that when you got into it, you're not going to get them. You're going to have periodic defle uh bonds blowing up continuously from businesses that cannot compete in a world of AI, in a world of crypto. You will have more and more disruption the same way you did with Amazon versus all the retail stocks. And name everything that Amazon did in 2013. Amazon was not what it is today. And that's what we're going through.

This is what's coming up this week. We've got the Fed. We've got the Trump sheet. We've got the Mag 7 earnings. The biggest week of the year is coming up. The biggest of the last.

So, here's the irony of this situation. These are the two guys. Look at the way they're dressed. They look. They're drinking. They're both crypto and AI guys. These have been the ones that have killed it. These are not the people that you see on CNBC doing this from the Goldman Sachs, Morgan Stanley, Blackstone, Black Rockck. These are the people that have caught it. And when you listen to them, they are talking about the future. Believe it or not, this is one of my favorite things that these guys are a joke. And yet, they're both really, really smart to listen to about what's coming in the future because they're out there promoting themselves. They're both involved in AI. They're both involved in crypto. My boy Pomp, same thing. He had a great interview this week with, not surprisingly, Tom Lee. Says it's not a bubble and stocks are still cheap. Here are the details from what Tom Lee talked about. Argues AI is sector is not in a bubble. Goes through Nvidia's valuation.

I just want you one time to do the numbers on Nvidia. Brad Gersonner did this on his BG2. He also did this on the all-in last week. It's a very simple thing. Go look at their revenues today and realize that every single data center, one gigawatt costs $50 billion. Of that, Nvidia is between 35 and 60% of that. So, of the 50 50 billion spent, every time Nvidia's chips are in there, it's 35 to 60%. Even if you haircut it, the stock is too cheap relative to the amount of gigawatts that are needed to power AI. So, you're betting on the fact that we won't build the power. you're betting on the fact that some magical um uh efficiency gain is going to come through when there's efficiency gains every day that Nvidia talks about. Their chips are all better than they were a year ago, better than they were two years ago. So, every time you hear some genius talk about the fact that this is going to happen, just remember they have a bias for whatever reason and go back to the incentive thing about why people do things.

AI isn't just labor replacement. It's aentic, super intelligent. Capital expenditures are accelerating. the profit margins are going because they're spending money and they're getting the productivity growth. So on one side for the S&P you're getting the earnings growth on the expenditures, but you're also getting the productivity gains that are coming with the agentic side. This is the whole point of replacing the labor side. This is why it's going to happen. It just won't be fired right away over 5 years where we're supposed to hire say 10 million people will hire nobody. So go to the bottom line. True bubbles, there's not a true bubble. If there was a true bubble, hedge funds will be bullish. we wouldn't have as he puts later uh and I'll get in there historical valuations don't apply in clean markets and that's because again of the profit margins and the roe I don't want to spend time on it but he calls this the most hated rally and this is something that is true the sentiment of professional investors so people that have been doing this for a long time remains net bears over the course of the year it's still net bears but here's the thing when you take the average for the year it's - 11.7 similarly only to 199 1990 2008 and 2022 bare markets we're up 15% year to date and we have monetary easing again we're going to start getting more and he says I expect a 30% correction next year I expect a 20 to 30% correction most years we've done it three of the last five so we've had bare markets guys it's just what ends up happening sorry three of the last six um if you go through and he talks about why it's not just an innovation wave. It's a structural rerating that gets back into the point of if you're reducing labor and your efficiencies are going up, it's a non-stop thing as long as AI continues to progress. So, we're doing agentic this year, we're going to get to humanoids in 5 years in terms of true like seeing the benefits that come into the bottom line. You're fading that. I don't even understand how you can do that without spending your time on AI and seeing if it's actually what it seems to be, which is what I spend my life doing. Almost every single slide you're seeing right now, it's me communicating with AI.

So if AI permanently lifts corporate ROE while making earnings less cyclical, which is what happens with this, we don't have a cyclical economy anymore. The fair multiple for the S&P 500 shifts structurally higher. Possibly from a 16 long-term average to 22 to 25. The reason people don't like this is they still pretend like history is a guide when we didn't have AI during any of this and they say this is just like every bubble. That's what it is. So what I did was say, okay, let's take Dan Ies. Let's let's take Tom Lee. They're both two structural bulls. Give me what they view as differently than everyone else. Now these are two separate people. They talk differently. They look differently. One's a tech person. The other one's a strategist. Just go through the list and look at these things in terms of what's going on. They're telling you the same story. And eventually they get into digital assets. AI macro nexus. the point where AI meets traditional macro together. That won't happen for another 5 years. And when it does, by then, hopefully you've secured your job. Hopefully, you've trained your kids on how to use AI and not told them it's a bubble. Don't spend your time with it or try to send them to a school that doesn't allow it. They need to be entrepreneurs. They need to leave learn this now.

I'll get to that later. This line here is when Chat GPT came in. This orange line here are the S SNP earnings going back to 1990. So temp hiring matches every single time when earnings go except for this time. So here we are. We're back to levels seen. I mean this is the total temp number. We're back to levels. So we didn't even take out the high. We took out the highs here and we've gone down. This has been recession times. And what are the S&P earnings going up? So, let's see. Stop hiring temps. Watch the S&P earnings continue to grow from Chat GPT at that point. Here's another one. Hiring. This is an overlay of Jolts to the S&P earnings. So, again, the S&P earnings going higher, making new all-time highs. They're not hiring. So, the J the Jolts number hiring. Again, go back over time and look, the hiring starts to come in when the earnings stop growing. Well, we still have earnings growing, the hiring is going down. Part of this is that the S&P 497 are in a recession and that the MAG7 keep getting bigger and bigger. This is the concentration side, but part of this again is the adoption of AI. Everyone will be doing what the MAG7 has because remember their profit margins are the highest in the market. That's why the S&P profit margins are so high. Now, everyone's going to get the chance to use artificial intelligence.

This I want to remind people cuz remember as you're listening to these people, go back and see what they were saying back when the tariffs, this was back from April 21st. Look at this. The tariffs are going to be the highest in 8090 years, 14 a half. This is going to lead to a recession if they're there. This is going to lead to inflation. Here's where we are now. Here we are. Fast forward September. We actually ended up there. We're at 15 and change now. Here's the actual number. I'm sorry. 17 16 and a half. This is where we are now. This is what they were saying. This is UBS is still saying US recession probability now at a staggering 93%. This is this week. Um, JP Morgan recession odds 60% chance. This again is right after liberation day. Larry Summers predicts US recession 2 million job losses. Hm. What's US stagflation risk mean for the world? I mean, again, guys, I I don't like I don't know. This is what Tom Lee said on May 15th. E economists predicting a recession without with certainty are dead wrong. Literally said dead wrong. In my 30 years, I've never seen and what he continued to say, an economist predict a recession. I've also never seen a person predict when the whole market is trying to predict a bubble and saying there's an AA bubble. When I came back from Brazil, people were telling me there wasn't a bubble. So, I don't know this situation, but I do know this. If we're in stagflation, profit margin shouldn't be doing this. This is AI, guys. So, yes, there is a part of the economy that if AI didn't exist would be in recession. There would be a part of this if the cloud didn't exist. All of this stuff has happened since 2007. Oh, and by the way, we probably would have had a recession if Silicon Valley Bank was allowed to go under and the Fed didn't bail them out. So we have two changes since 2008. Unlimited liquidity facilities to put in place, which is anytime you worry about credit, it doesn't mean anything. Credit is now allowed to go the way China does it, which is okay, we'll gradually let it go out of business. They just don't allow the contagion. They stop the contagion, but they allow businesses to go out of business. We're going to see tons of bankruptcies now because they raised rates. They said they wanted a recession. They weren't able to get it. And that's because the part of the economy that is the biggest, the technology and digital economy side is growing so fast that it's offsetting everything happening in the other one. So both people are right. There's a bare market somewhere. There's a bull market somewhere. That's pretty much always the situation.

This is from Goldman Sachs about retail. Someone asked the question why retail keeps outperforming institutional money. Simply put, retail has remained risk-seeking while institutions have been cautious. Nets and sentiment have remained low all year. institutional money tethered to valuation and those pesy pesky fundamentals. When they say those pesky fundamentals, when they say valuation, this is history. The same way that when people say when uh defense when interest expense passes defense expense, never bringing up the fact that we never had entitlements in any of these parts in the past. So interest expense seems really big but not when it is such a small number out of the entitlement number. This is a completely new world and these historical benchmarks to do is why these guys actually get it but nobody listens to them and they try to make it hard to listen to them and I love it.

All right, this is what I wrote this week because I started to get the next phase. I come in every Monday and there's something that comes out over the weekend with something about how AI is is a bubble. Capitalism tells fables about the future so it can afford to build it. This isn't criticism, it's description. Every major technological leap arrives wrapped in exaggeration. The steam engine was sold. Blah blah blah. The whole point of this that I wanted to write about was related to Andre Carpathy's interview that was done that got everyone freaked out. There were some things in X saying, "Oh, this is the end. Watch how much Nvidia is down on Monday." Carpathy is an incredibly smart guy. I like listening to him all the time. He was at OpenAI and most importantly to me, he was with Tesla and is the head or was the head of their whole autopilot unit in terms of the engineer on it. So I care a lot about what he said and what he said I actually agree with, but everyone else read it and decided. So I wanted to write something where capitalism always tells fables. AGI is what he was referring to. So when you hear fables, capitalism tells fables. Remember the thing I showed you about equipment shares? Equipment shares was telling a fable. They were trying to raise an IPO and then there was fraud. So everyone immediately jumps to fraud. Sam Alman could blow up. Open AAI could blow up as part of this. But the problem is you are in a part of the phase where you're not seeing any sign of it because demand is greater than supply. And when you're in a bubble, supply is greater than demand.

This is the DRAM price. So remember gold and how much gold has gone up since August. Everyone's investing in that, but they want to fade this. This does not trade on a public market where retail can trade it. This is not driven by stock prices. This is a commodity. This is DRAM. This has gone from here in September to here. Here is the last two years. Think about what's happening. And then go look at the Micron chart. Go look at all of these AI things. We have a shortage right now of compute and memory as I've talked about and what's going higher with it. These are the parabolic stocks that I was mentioning. the parabolic companies. So, you've got Micron in here, you've got Comfort Systems Fix, you've got that DRAM price, you've got Oracle, you've got EOS. Everyone still finds a reason that these are bubbles, that they're not real, that there's something going on. The reality is this is what Comfort System said. That isn't retail. This is about what they have. Their order fix reported its largest ever backlog up $10 billion. Look at that number and then go look at the size of comfort systems company it's up 9 billion 65% year-over-year explained that modular construction this is construction stuff they are seeing explosive growth he described this AIdriven data center trend as a seismic shift for the industrial sector noting that the company has transformed from a midsize contractor that's why I say it was a small company into a core player powering the AI revolution $9 billion backlog for a company that was less than $10 billion not that long ago as a market cap. That's why those stocks are going up. So everyone that's fading it saying, "Hey, it's a bubble. Once it pops, we're all going to suffer."

These are all this week. Data center lending probe by BFA amid AI bubble fears. Surge in AI demand and price increase likely to drive earnings upgrades. So this is the positive side of it. No signs of AI bubble burst this weekend. I also got this. Alibaba cloud cuts Nvidia usage by 82%. Look at the efficiency gains. Alibaba is going through Goldman Sachs sharply raises Alibaba's capital capex outlook to 65 billion. Explosive growth in inference demand. AI efficiency improvements draw drive stronger revenue. The efficiency improvements are driving stronger revenues. Efficiency driving. Oh, that's right. We're getting back into Jevan's paradox. You're moving efficiencies. Price comes down. More people are using it. So you can go through this but the token usage the adoption side inference demand and token consumption are growing exponentially. Bite dance recently announced that its average daily token consumption surpassed 30 trillion in September alone doubling from April to May a level approaching Google's 43 trillion. Alibaba also revealed that its inference ran doubles every 23 months. The reason I want to bring this up remember the Mag 7 are competing with China. China is absolutely dealing with this. Here's what Airbnb said. We're relying a lot on Alibaba's Quen model. They are winning open source. It's very good. It's fast and it's cheap. We use OpenAI's latest models, but we typically don't use them that much in production because they are fast. There are faster and cheaper models. This is why at the end, if you guys are looking for how this works, the multiples of the Mag 7 are going to come down unless they generate tremendous revenues. They're going to make enough revenues, but if they're competing with China, they're not going to get 100% of the pie. This is going to be distributed in a much bigger way, which means price to sales is the new debt to equity. I'll keep saying it again and again.

So, this is the podcast with Andre Carpathy that got everyone bearish. Um, he he's saying AI is he saying he's in a it's in a bubble. No, but he's is basically saying that saying it's the year of agents, calling it over prediction. Okay. He's not saying there won't be AI agents. He's saying in his level, we're not there. Meaning we're not at AGI. They can't do everything it is. And then he goes through this to describe. He says he's extremely bullish on a AI as a tool, but clear that it's not yet a substitute for human cognition. Okay? Anyone that doesn't know that I I of course it's not there yet. So that's why again as I go through it's the realism versus the hype machine and he's calling it out. Dwaresh and Cararpathy are both brilliant, but neither of them are involved in frontier model work. So 10 years to AGI is basically vibe forecasting. Could not agree more. Nobody knows when this is happening. Nobody. It doesn't matter. Did he say that true AGI is not reached until AI can work in the physical world like a human? Yes, I agree with that 100%. So I completely agree with what he is saying, which is that we're not going to have AGI. That's my personal definition that if you ask me what it is, artificial general intelligence means it can do everything that a human can do from a thought process, which also means the ability to walk through a room and see everything and make decisions real time on this. That means autonomous vehicles, that means humanoids. That's what he's saying is that it has to be visual. It can't just be textual. And I completely agree with that and we're not at that point yet. At the same time, the podcast that I recommend the most, if you haven't, if you didn't hear it already, please listen to it. The singularity is already here. AI is solving math. Sora outpaces chatbt and AI is designing chips. This is where you're starting to get into the next phase of AI. And that's what I wanted to cover.

There are two arguments in this. The Carpathy argument which includes Karpathy, Deis Sabis, Yan Lun from Meta, Jeffrey Hinton really one of the fathers of AI, Alon Mus and Dario Modi from Anthropic. They're in one camp in terms of what AGI and what it means. It requires new architecture, long-term memory, multimodality, which means the vision as well. And continual learning must exist before true general intelligence, uh, 10 years. Then you've got the recursive learning side, just a different definition. AGI will emerge once systems recursively improve themselves. Meaning, they're learning on their own. You can have this point going on already and this point not there yet. That's why they're both right. In my opinion, there's recursive learning that clearly is happening based on everything that people are seeing. It's just the degree of what level. So it's kind of like going from 25 to 75. You don't just go from 25 to 75. You go from 25 to 30 to 35 to 40. That's what we're going through. And that's why these arguments in these bare markets and taking something that is continually moving and turning it into a binary is just not right.

So Musk versus Carpathy on AGI, a system that can see, hear, move, and reason and embodied intelligence indistinguishable from a human's capability. I agree. That's his definition. A system that can remember, perceive, reason, and act across modalities. A digital employer agent with persistent cognition. These are very similar. They're basically on the same page in this. They might have different timings as to where it is. Now, Elon Musk posted an X saying 99% of input and output for all a models will be photons. Okay? And this is a critical part when it comes to Tesla and it comes to humanoids. So, they're both again why you need to have pixels. It's the only way to scale. In short, Musk is forecasting a photoncentric future for AI to unlock godlike scalability where light becomes a default language. This gets really important because this is why to go from text to this, this is going to take a while. The next two stages of AI, we finished stage one. We've consumed all human knowledge and now we're starting to get garbage on the the textual side. So they have to get good at ignoring it. This is why Tesla's robo taxis are so important. This is why I wrote this. This is why everyone again that has a perma negative view on Tesla. These guys in my opinion, due to Alan Mus, due to Carpathy, these two guys have worked on this and that's where we are now in terms of making that next step. We need to have this to get to true AGI and they've been working on it. Not Whimo, Tesla. You can be positive on Whimo as a car, but Tesla is making the move to get into both.

So, in the Q3 earnings from last week, Elon Musk expressed 100% confidence in achieving unsupervised full self-driving at a safety level far out exceeding human drivers with plans to remove safety drivers from robo taxis in large parts of Austin by year end. as soon as they make it through this where there's safety drivers are gone. Okay, where he's saying and again he he might be off by 3 months we're definitely heading there but the second that you are at year end where the safety drivers are gone in my opinion the investors will start jumping on this fact. So the fact that the price has moved, remember price leads narrative. The narrative for investors is going to be starting to look at the full TAM for not only you for not only robo taxis for that you can just go read Dan Ies and go see what he says but it starts to merge in Optimus because the brain for if the brain is good enough to drive around with it's not too far for this but also he confirmed that a prototype of the nextG Optimus version 3 will be unveiled in February or March 26th. Now, what unveiled mean? I don't know. He also has the shareholder investor event coming up on November 6th. So, the fact that he said there's going to be a that he's going to be removing the drivers. Let's go. If Tesla remove safety drivers from robo taxis by year end and demonstrate safe sustain autonomous operations, the valuation pol policies contemplate both bear and bull. So, bulla ba base case here and again this is revenue not valuation. the bull case becomes very very large. Goldman Sachs and ARC forecast 150 to 200 billion in annual revenues from mo from autonomous mobility by 2030 with margins nearing 40 to 50% due to lower operational costs and high software. So the price targets in the bull scenario reach 3,000 to 4,000. Again, I'm not getting into where the stock's going to be. You guys can do your own homework on what you think is going to happen. I'm just saying that that's how important this is. And a lot of people especially that cover the stock or that have a view just like they do with AI, they're not even spending time on it because they don't believe it's going to work.

What would happen in terms of the typical cost versus Uber and versus Whimo? Will you get significantly lower fairs because of the cost of the car? The reason that's important, will it be difficult for other cities, think New York, to not allow Tesla robo taxis without the safety driver if it can save riders 50% compared to other rides? You already have problems in cities with the inflation side. This is going to be something because of where we are today where inflation is a massive political risk. If it's working in other cities and it costs $10 to take uh a robo taxi and $30 to take an Uber and in New York they're not allowing it, it's going to be pressured. This is the way AI and this stuff is going to spread in states that are trying to regulate things because eventually if people are having trouble with jobs, the pressure on payments living paycheck to paycheck, every dollar counts and that's what you're going to see. Now the other thing with it is as they start driving they accumulate more real world miles. The resulting data enhances the training of their large-scale AI models. Remember, as they get more and more autonomous vehicles, which will happen if the auto autonomous driving is allowed, if robo taxis are allowed without the safety driver, then people can start to get in the business of having their car be an Uber without needing to be in it, which means they can be making money. That collects more data on its own. you go through it, they're driving more often, you start to get more driving miles, and just like you do with an LLM, you actually get a better model. So, a successful unsupervised robo taxi driver uh prepares the readiness of the brain for Optimus. Tesla's FSD and robo taxi software represent one of the most advanced vision-based AI stacks in real world navigation, perception, autonomous decision-making. So all of these things go through the economic implications for it. It's super important as the beginning to the timing of this. Yes, Tesla robo taxis play a critical role in the timing and path towards reaching AGI until you get to the point where they can learn more. It's going to take longer to get to fullblown humanoids. Humanoids are already happening. If you haven't seen this video, again, figure 03. The difference for this is this is a video, but it shows it in the home. So, if you want to go see what's actually going on, remember, is Figure 03 the Model T moment for robots. Is the humanoid robot industry ready for its chat GPD moment? I will say that Tesla's robo taxi when we get the first ability for it to have be unsupervised and it spreads even if there's accidents and even if they try to shut it down the data is going to start to be captured and the the humanoid chachi moment will be looked back in my opinion as when robo taxis went.

So now jumping back to what was said in the singularity piece for moonshots. We're already living in the singularity. AI adoption is eight times faster than the internet. GPT pro solves math again. All of these things AI and data centers drive 92% of US GDP. You guys can go go through it. I am also in the camp that people just don't realize we'd be having a recession right now if it wasn't for AI because the profit margins would also not be good. And if the profit margins wouldn't be good, the stocks would be down, the stocks were down, there'd be no consumption because we already know that the job market's weak. The only way people are spending is that of savings. And the savings that they're able to spend that is at the high net worth individuals, which is dependent on asset prices. So asset prices go up. The new economy will be defined by abundance. I'm not going to talk about this in great detail today, but I did start to talk about it for a while in here. If you want to start to hear the way that I think about the way the world will look in 15 years, which I have not talked about on this video, uh, let's just say it all has to do with abundance has to do with not needing people to do anything. So, you get into a scary thing of identity, but that's where we're headed, which is the jobs market is changing.

Now, for Dark Cash, he did put out something very good this week on the capex overhang thoughts on the AI buildout. And again, I'm not going to go through all of this. I think it's worthwhile to go and he did it on Sam Walton says he wants to create a factory that can produce a gigawatt of new AI infrastructure every week. So what would that mean? They question whether it's physically possible and that's the point. Can can the power grids materials and supply chains scale fast enough? And that's the question we're going to run into. Everyone's worried about the money. They're worried about the bubble. The question is the demand is so great right now that if they could and they could snap their finger, they need all this. That's not what was happening in the bubble, the dot bubble. We didn't need the supply because we didn't have the applications. They already have the demand for it. That's the problem. The difference between the two. We're playing from a position where the supply can't actually be built because this came out of nowhere and the demand is going too fast. So the compounding is happening every 6 months. That fast lane is moving at an exponential pace. So you have exponential demand. We want to cure cancer. can't do that if we're giving everyone porn and we're giving everyone all these tools for videos and everything which these guys need to do to generate revenues. So the problem they're running into is they can create all these tools. They can't let you make a video for an hour and a half so you can go make a movie and try to sell it. If they could, like I've said before, Disney would buy all the piece. So they go through all of these things or he goes through all the buildout. He goes through the complex parts.

So what I did was okay show me now the companies in the stack behind the infrastructure that he referenced in particular uh that would that is the the biggest problem. So these are all companies that you should be looking at. This is all the part that he's saying is going to have massive supply issues. Most of these are semiconductor companies. Um, and that's the issue. The green one I want to highlight here is ASML. The monopoly in EUV lithography the single most irreplaceable company in the AI chip production. This is why I keep saying that ASML not only benefits from the growth of this and why I like it as my favorite stock for next year or from this point on for the next year where Micron was my favorite for this year. Focus the ASML way. It's a book that goes through an unparalleled analysis of ASML's extraordinary rise and the geopolitical place they put themselves in. They are at the center of a geopolitical storm between the United States, China, and Europe. If you don't know about ASML, it's a cool thing to go look because of where we are right now. Um, in systems terms, ASML represents the choke point in the architecture of modern comput computation.

Without its machines. Um, it's worth going through these machines are about $200 million. So, think about how how expensive. 200 million, not two million, 200 million. Um, they're not easily replaceable.

So when you go through this, what I said is, given how many companies are trying to build new chips for placing AI in the entire industrial and consumer infrastructure of machines, does ASML benefit from the next stage of AI for NPUs and edge devices? So again, I've I've mentioned them as one of the most important. When you're constantly trying to come up with new chips, you need to use their machines for every chip that's being made. So instead of just using the same thing over and over, if you do a new one, well then you're going to have to use a new machine from the stuff that I've learned. So the demand for them goes higher. So you get all these parts. They're also involved in another part of the MPUs. Basically, we're going to need so many different types of chips that they're going to be brought into it.

Now, quantum. There was a breakthrough. Again, I didn't get any calls on Bitcoin this week on it, but Quantum is making advancements. That's another reason why I'm going to continue to remain on the optimistic side that China and the US are going to come to an agreement because I believe it's mutual mass destruction. Uh, and the only way I see this not as an option is if China really wants to invade Taiwan now before the US can build up some munitions and before quantum. So the China-US thing to me is a major inflection point. Anything comes out of it that comes to an a trade agreement where the countries are going to work together. Even if it ends up breaking in two years, I think the stock market is just going to run because I think it's being held back like a Kentucky Derby horse that wants to run.

X-bankers get paid to build the bots. Former financiers are earning $150 an hour to train OpenAI's models to do the junior banking analyst jobs. This was out as a story this week. It's a big deal. Um, and I, when I say it, Mker, I told you this is what everything is doing. They're bringing experts in to train the models so that they can give the models or sell the models to the banks so they can get rid of the people. This is what's happening. Humans are augmenting and they're being involved because as they get fired, they can go get a job with OpenAI, making some money in the intern, but also learning how to use the thing. You learn how to use it, then you can get a job. OpenAI just said the quiet part. They aren't building chatbots anymore. They're building an automated researcher, a system that can discover new ideas, run its own experiments, and push progress in every science. Recursive learning. We are getting to the recursive learning point.

Now, this port, this New York Times had an article um that was leaked. Amazon is planning to replace roughly half a million jobs with robots in the next few years. Half a million jobs, according to a report. Now, some of those would be new hires that are not happening. They employ about 1.2 million workers in the US, second only to Walmart. You're talking about replacing. So, this was a leaked report. If you haven't seen the robots, they do. It's coming, guys. AI and robots will replace all jobs. Elon Musk working will be optional, like growing your own vegetables instead of buying them from the store. I know this is hard for people to do. This is the world that you have to start thinking about, not just for you, but as I said, I'm getting more into helping the kids because there's a lot of people that aren't going to change. Reminder, the job losses started. ChatGPT earnings are growing, hiring is going down. This is where I want to start the training for people. So this is lesson one. Um, if you haven't seen this commencement speech, uh, from Admiral McCraven, he gave it in 2014 at the University of Texas. One of the best, if not the best, definitely in the top two for me, along with the Steve Jobs commencement speech. Um, he put this book out afterwards, Make Your. That he goes through in there, and everything that he talked about, in my opinion, is the way that I am going to help people with artificial intelligence. This is what is going to be part of my job. Whether it's businesses through the consulting side, or whether it's creating videos to show people I do things and trying to get people to think. If this said, "Make your bed," which is what he went through, I'm going to convert it into how you use AI. So, empowerment in the age of AI, lessons from Admiral McCraven. Again, me brainstorming with AI, converting his 10 lessons that he mentions.

"Make your bed every day." Start your day with discipline and success. Get a win to start your day. Even when it's hard, even when you've had a bad day, you make your bed every day. Simple things. Start small every day. Use AI daily. Even for five minutes. Small wins compound. I'm telling you, that is a fact. I've learned so much doing it. The more you continue to use it, even if it's for five minutes, and start your day that way. I use it all day long. And when I say all day long, there's almost no hour, just like I don't go an hour without walking 250 steps minimum. There's not an hour that passes where I don't use artificial intelligence. I don't have hour meetings. And if I'm in a meeting, I will eventually have it run something while it's going on.

"Don't be afraid of the circus." Hard moments build resilience. Embrace the learning curve. AI mastery comes from persistence. You get better as you keep using it. And if you go backwards, you ask another question, you come back, you will get better. "Be your very best in the darkest moment." Keep your composure under stress. Stay calm in uncertainty. Use AI to organize thoughts and build clarity. Before I do any presentation, I did this one one this week for a pension plan. I always prepare using AI. I prepare organizing my thoughts, just like something like this. But then I also, once I have this, I go out and talk to it about this. It quizzes me on it. It goes through it. As I'm walking, I'm speaking to it. I walk down Man or down Park Avenue. Nobody knows that I'm not talking to a human. They think I am. I'm talking to OpenAI.

"Don't ever, ever ring the bell." Never quit. Stay curious for life. AI learning never ends. It compounds. I can't say it loud enough. That is your first lesson. I will expand upon that in videos. Um, the team on the weekly roundup for forward guidance. I just wanted to bring up. I like these guys. I've told you they are the best macro thing to listen to. Again, the one thing they never talk about, or at least if they do, I haven't heard it. So, these guys are unique. They're very, very good Fed watchers. They're very good macroeconomic people, and they're excellent crypto people. The part they're missing is, to me, the most important part. It's the bridge between the two, which is artificial intelligence. The only way you can truly understand what's happening in the traditional f traditional macro world because it's being disrupted by artificial intelligence. It's having the job impacts. It's having the profit margin impacts. So if they say stagflation, I say stagflation, yes, but not when you have AI. I say jobs, yes, they suffer. That's the right lane. That's the slow-moving lane. The left lane is where the profit margins go in. That's where the building is happening. It eventually all ends up in Bitcoin.

So what I did was say, here's my view, which again, I'm not going to go through, but the abundance era in the way that I describe it, which OpenAI or ChatGPT has, Gemini has, Perplexity has, because I've put my endgame. I'm a systems thinker. I think about what the world is going to look like in 2045 and I walk backwards. And then I watch AI as it's moving. I compare the two, and everything that shows up, we're very similar on this. The big thing is exponential innovation compresses time. The traditional finance world is all based on time. The reason there are no more cycles, business cycles, recession cycles, which I've been preaching since 2013, is because time is going to zero. Today is yesterday. By the time I do this, so many more things have happened. So many models have happened, so many advancements have happened. There is something going on that's different. The entire fiat system is based on slow-moving time. So when time goes parabolic, the entire system breaks. There will be no leverage in the future. Zero. The entire system is built on leverage. How do we go from a leveraged world where there's debt and deficits to an unlevered world? How does that happen? That's part of the thing that makes this fun to think about and to go through.

So, here's the winner at the end. My opinion, Bitcoin. So, Bitcoin is still basically right around where it was a year ago. So, we hit here in December. We haven't done anything. Even though year to date we are up 20%, last year we were up a ton. It's fine. Uh, we're starting to get a little technical MACD buy signal on the dailies. Uh, we've got V all the way down here. So I'm very comfortable knowing that when we peaked at the four-year cycle, and this is the 90-day V. So this is a little backward looking. Whether it's here or whether it's here, we were pretty much at one-year highs and VSS. Right now, if we're peaking near the all-time highs, slightly below it, it's not happening with the lowest V there. Bitcoin has actually become boring. It's also very negative. The space is horrible. There's really, it's, it's been a brutal year for crypto relative to what you'd expect. And yes, I know Bitcoin's up 20%. In fact, 19 as of when I did this, it was at one 11. Yeah, 111 and a half. Um, but after two years of over 100%.

Now, Galaxy reported earnings and they highlighted the sale of more than 80,000 Bitcoin, representing, this was public, representing a $9 billion notional value as part of a broader estate planning strategy for a single client. This was one of the largest Bitcoin trades ever. So, if you're trying to figure out why Bitcoin is not going higher with all of the buying we're hearing from retail, all the buying we're hearing from uh these treasury companies, possibly governments, other firms and analysts confirmed that long-term holders, OG whales, have ramped up disposals on-chain, showing 240,000 sold by long-term holders just in the past month. This is an October thing. So, this doesn't include the sales that we were talking about there. Um, so you're talking about now you're getting into the 30s, 40s of billions, maybe even more. So OGs are getting out. And then you've got this thing which goes through. Bitcoin have sold over 240,000 over the past 30 days. Largest wave profit taking since January of 25. And this is all with the market at highs again. You've got the four-year. The biggest problem to me, I could give you a bunch of reasons why people will be bailing out of Bitcoin right now. Uh, people are scared of the four-year cycle. I've heard that. I talked about it last week. I think the other thing that is going on is people may think that Bitcoin's lost kind of its ability. It's just not sexy right now. You've got gold going up faster. You've got the AI trade which has driven attention. Without retail, the institutions, I, you, they don't, they're not panic buyers. They don't push things. They're just there. On the sell side, we have, according to Mike Novogratz, when he spoke uh on the Raul Powell episode, he not only had sellers, he constantly has overwriters. So, there's been V compression. You've got people selling call options. You've got the Bitcoin miners that are continually selling call options. If you don't have retail involved in the space, it's not surprising to me that it's sitting there. So, right now, you have all of these people looking for yield and people basically thinking that Bitcoin has reached a point where they've got better places to put their money. And the buyers, they're just there and they come in every day in small amounts. The question is, when are we going to break out?

So, the part that I care about is that every day there's more and more crypto visions being laid out by the banks because of the regulatory side. That's the part that matters to me. The regulatory push is coming. It's a slow-moving thing. We haven't gotten the market structure act done, but it's happening slowly but surely. I've talked about comparing this to the time of the internet era, coming out of 2002 to the the dot bubble and coming out of it. We are, in my opinion, next year is going to be like 2007 or 2008 from when the iPhone came out. You're going to have an investment boom that's going to go on in these things because of AI agents, which I think is the trigger point. The payments MCP. This is Coinboy is launching its payments uh model, Context Protocol. Easiest way for agents to get on-chain. So the merging of agents and crypto is going to start happening. You've seen all the things on stablecoins, and that's why the timing. I wrote this in Substack this week. If you haven't read it and you're curious about stablecoins, and again, if you're in the right-hand lane of the slow-moving tradey world and you haven't wanted to get into the fast-moving left side, I highly recommend reading this. I tried to simplify it in a way that made it easy, and I used two podcasts, both from Moonshots, that were crypto-forward. You can go listen to both of them.

I think there's an important event going on as I finish this up in Argentina happening this weekend. Uh, there are elections. And if it ends up going badly for Malay, where he basically, the disapproval of him shows up, I think Argentina is going to be the first one of why Argentines are turning to crypto in the latest peso crisis. So you've already seen this go on. But remember, here are the Argentine bonds. The white line, Bessant comes in to give money. If the election goes badly, I would expect the bonds to break down below the lows, and this becomes another story, and Argentina goes through another phase of this. If somehow or another he wins, the US will give him $20 billion. You can stabilize and maybe he can work through this. But I think the only way he's going to work through it is if he moves into what this orange line is, which is El Salvador bonds. So look how correlated they were all year. So two countries that historically have had some problems with their bonds, and oh wow, right around here, Argentina starts fall off, but El Salvador just keeps going. Bitcoin, fiat. So I think this is a big story. But I also think it's a big story for another reason. Why is China spooked by stablecoins? US rich, uh, US or EU is worried about stablecoins from a risk management basis. Systemic bank run risk hangs over stablecoin bills in Congress. Remember Silicon Valley Bank, the second biggest bank failure in US history, took less than 48 hours. Do not be surprised if Argentina does not run into the same problem after this election. If their currency starts falling faster and their bonds are falling. Right now, the locals have the ability of moving into stablecoins quickly, and you could see, just like what happened with Silicon Valley Bank, you could see the locals have a run on the Argentine system. I believe that's what's going on anyway. So, in the US case, they want dollar-backed stablecoins. So, by giving $20 billion or saying they're going to, if they don't do it, they're kind of winning either way. Either they go towards a dollarization or the money comes back through the door in stablecoins. This is why you have to read what's going on. This is why Europe and China are so worried about stablecoins because at this point, they're just a twin for the dollar. Standard Charter estimates $1 trillion could could exit emerging market bank deposits for US stablecoins by 2028. This was something they published not that long ago, earlier in the month. I think this is an important story when it comes to Argentina.

That's all I got for this week. Uh, again, uh, subscribe, forward to your friends. Let's get the subscriber numbers up. I'm working on the videos to make sure to help you, but more importantly, your kids. And for any of the colleges out there who want to reach out to me, I will try to squeeze it in over time. Uh, jump on the phone for you an hour, kind of give you some stuff to to work on and some things to think about and do a broader presentation than this. Uh, thanks guys for everything, and I will see you next.