牛市内部的一场熊市

Okay, let's begin. Presentation on time inconsistency.

In this presentation, I'm going to tell you about a topic that I had in mind this week in Washington, D.C., where I spoke to Freedom Tech. I liked it. I made a lot of friends there, and I'm glad I went. I'll tell you about it.

Of course, I have to talk about the scary bets. I can always tell when people approach me, and by reading X, what everyone fears the most. It's just amazing: it goes from oil to ponds. There's something new every week. It's simply impossible to keep up with people's anxiety.

I don't know if this reflects an inability to make money. I don't know if that reflects pressure. I don't know, but we'll look into it. I'll give you the facts. For every person who tells you that every time rates go up, there is a disaster, this is a story that is not true.

Secondly, I'll walk you through the facts from the history of markets, from the perspective of how I've been modeling myself where contagion starts. I had to do this because I grew up in Brazil, in emerging markets, where if you didn't get out before everything blew up, you died. I watched a lot of traders and people in the 1990s who weren't there. I developed my entire theory on Jeffrey Moore and economic leading indicators.

But then I'm also going to move into a whole new mindset. I've talked about this book before, and I think it's really important that people start thinking about this. It's Muse time: consumer agents, the next moment of artificial intelligence. This is an important point, guys. This is what drives the market today, and I believe it will be the driving force for the next 12 months.

The investments you will make over the next 12 months, in my opinion, to succeed in the alpha project will be related to consumer agents. There's a bear market happening inside a bull market, and I'm going to look at that because I think it's going to continue. The IWM-to-QQQ ratio is hitting new lows since the iPhone. I'll talk about this and all the other good things related to cryptocurrency.

My disclosure: Remember, guys, I'm wrong. Sometimes I'm right. Good.

1. Time Mismatch: Investors can think linearly or exponentially, and backtests built on a linear world no longer apply. Alvin Toffler's Future Shock explains why so many investors are angry at AI and call it a bubble.

So, yes, I went to Washington. I really want to raise this issue because this presentation was about there being 2 different worlds, and you have a choice. As you can see in my V here, you can either think linearly—think about it: every time rates, for the past however many years, have done X, Y, or Z, this is what happened—or you can come here and think exponentially.

And that's the problem. We didn't have any rate increases while this was happening. We had a rate hike when that was happening. Keep this in mind when talking to people, and I'll give you some tips on how to avoid it.

For people who were in Washington, in the Bitcoin community, this was a great event. I found new friends that I will be friends with for a long time. You guys know who you are, and I liked the presentation. I hope I can show it someday.

But that's exactly what this is about. I would like to show you the notes I made, just so that when I sit down a few minutes before I get up, I can continue.

So, there is human time. This is the system we are involved in. There's a reason why housing, cars, retail, restaurants, and other things are trading where they should be: at bear-market levels, because rates are rising. Ghost rails are the other side of the equation.

So you have human time, where everyone backtests based on a linear world that no longer exists. Most of them are over 50, which I'll tell you about. Ghost rails are the most important story. They're related to agents. I'll talk about why it matters, but here's the thing: the speed of intelligence changes. The speed of money has to catch up.

So, everyone you listen to from the bank side, you have to start with their age. I don't like it, and I don't want to say it, but it's just a fact. There's a big difference between someone who's in their 50s and has already made money telling you how the world is going to end, and someone in their 20s or 30s who's working really hard on artificial intelligence and understands cryptocurrency.

It's like 2 different worlds, and so there's a time discrepancy. People haven't adapted because they don't use artificial intelligence. Alvin Toffler—I'm referring to this book—but that's really what it is. If you haven't heard of “Future Shock,” you should.

We are at the stage where it will scare and make everyone feel overwhelmed. I think right now, anyone who is, frankly, pessimistic about what's happening in artificial intelligence doesn't want to be involved in this process. They can't. Mentally, they're angry about it. They think it's a bubble. They think all these things are related to anger.

And I think that makes sense, because that's what future shock is. If things change too quickly, people don't want the rules of the game to change. They don't like it when managers make decisions based on quantitative data. It's an old man's “get off my lawn” routine, and I think that's why I keep showing it.

I didn't want to be here. I made the trip there in 2013, to Silicon Valley and Singularity University, to make sure I wasn't focusing on it anymore, thinking it was no longer relevant. And that's what worked. To be honest, this has been working since 2007.

This is why IWM versus QQQ falls every year and continues to hit new lows. Even with the growth and expansion at the beginning of the year, we are back to the Magnificent 7 reaching new all-time highs. Both the Magnificent 7 giants and Russell 2000 companies are having problems due to AI.

Not everyone is. Some of them end up in infrastructure, as we saw earlier. But when prices have risen and oil has become more expensive, these companies have no way to survive. So, this is the world we live in. And here we're going to get rid of a lot of people. That's why I say there is a bear market inside a bull market.

2. Rates and Age: The loudest voices calling for depression are mostly over 55. Tokenization and AI didn't exist in the decades those rate backtests were built on.

Too much of our lives is now polarized into binary situations. This is a bubble; this is not a bubble. There is a time discrepancy. It's very hard for me to imagine, given how long I've been in the markets, that blockbuster earnings will support record stock gains.

And yet, while this is happening and stocks are near their all-time highs, I'll be on Julia's show soon. Those are just 3 that have gained traction here: Depression ahead, Michael Pento, who's 55 or older, and Danielle DiMartino—I don't want to reveal her age. I looked it up, but she's in a group with me, along with Michael Pento and George Noble, who's older than us. We're all over 55, or over 50. Let's just leave everyone else out of it. That's the main thing.

See this graph? Does it look like it's growing? Absolutely, for me. I mean, it's a nice 5-wave uptrend, consolidation, a small triangle—it's growing. Do I think it's growing? I think so. Do I bet on it growing? No, I don't think it's worth the time and energy when I have other things to do.

Do I think this period is very different from this period in comparison? Yes, I think so. But again, everyone wants to just take the story and scare you because it's going to go up to 10, 12, 14, 15. The government has a bunch of debt, all that stuff. This is history. Here's another story.

I'm 59, soon to be 60. Here's what I've been doing this week. Cryptocurrency: why now? All these different components—artificial intelligence agents—are the inflection point. I'm showing you this because I spend my time looking at where the world is going, not where the world has been.

Tokenization is happening. This is happening now. I'm talking about how tokenization will fundamentally change M2 and fundamentally change the monetary economy today. This is the most important thing happening in the world today, and it is directly related to cryptocurrency and directly related to artificial intelligence agents.

Bond yields—believe it or not, I ran a test in hindsight. There was no tokenization in the 1970s, 1980s, 1990s, or 2000s. There was no artificial intelligence in everyone's hands that would accelerate the speed of solving mathematical problems. None of this happened.

So, is it different? This is completely different. I don't even know how people don't see this. So I included A Whole New Mind: Why Right-Brainers Will Rule the Future. I'm going to come back to this.

This has been going on for some time. I didn't make that up. Remember when everyone told us in 2022 that rates were rising and the Fed raised rates, and that we were definitely going to have problems? We did have Silicon Valley Bank go bankrupt, but it didn't even cause a dent in the market.

We had an inverted yield curve, a 100% chance of recession. We had this LEI year after year. I studied it for years, but I didn't fall into this trap because I knew it was created in a time that no longer mattered.

It's the same with rates. Yes, rates mattered in the past, when companies that didn't have good balance sheets took on debt and bet on the future. When people were buying houses, wouldn't anyone be buying a house now? Home sales are trading at near-historic lows. Everyone has positive equity in their home. This is a completely different world.

3. The Facts: The LEI turned positive with no recession, and S&P earnings are up 17% year over year. Credit spreads, jobless claims, profit margins and money market flows show no sign of rate stress.

The government is burdened with debt. So if you want to bet that the United States government will default, like I wrote in that article—blah, blah, blah—I've heard it too many times in my lifetime in this business, about governments not being able to overcome their debts.

The LEI was just turning positive. We've never had a recession, but you know what? Every time it turned negative, we had a recession. That's why I say there are many cases where this time is different.

But I want to give you the facts so that I can change my mind. If your annual S&P index goes negative, I will change my mind. And as I said, I got a recession signal here in late 2007, closer to the peak.

This was before the market peak. This was right at the peak of the market. When you have an economy that matters, stock prices reflect earnings. They reflect hiring, and they reflect the rate of return. All of this matters when considering whether rates are having an impact.

If rates are having an impact and you say, “Well, it’s a bubble because stocks aren’t falling,” okay—but what about profit growth? What about the growth of the profit margin? I will also consider other things. In 2000, before we hit the recession, you got a sell signal. The second derivative matters. The second derivative shows whether there are problems in the economy.

We’re up 17% year over year, guys. Here are the credit spreads. This is a comparison of Moody’s Baa yield to the 10-year Treasury yield. When we have problems, like in 2000, spreads widen. Spreads are widening. They always do. If rates have an impact, you’ll see it in credit spreads. That’s not there.

Now look at unemployment benefit applications. Regarding everything related to the labor market, if companies have problems, if we have problems in the market, if rates matter, you will lay people off and they will not be able to get jobs again. They will apply for unemployment benefits. That’s why this is so important. Before every crisis, unemployment claims increase. They’re not moving. So when those things change, I’ll be worried about that.

The rate of profit, as I said, is reversing sharply. Previously, it had a downward trend. Look, it peaked in 2007, and it was already declining. They haven’t even stopped growing parabolically. By the way, this is another fact. All I’m showing you are facts. This is not a story.

I can make it up as a story, but people who use bets to tell a story don’t have any facts. It’s just a story. Here’s the story—or the facts. Money supply, or money market funds, again, on a 3-month basis, was growing. I don’t care if it’s a random 3 months. Historically, this happens during certain periods, not when the stock market is at historic highs. This happens during a panic. There’s a lot of cash in there. It moves because the stakes are higher and people are collecting as much as they can.

4. A Whole New Mind: Daniel Pink's book predicted a world that would no longer belong to mathematicians. Only 2.2% of people aged 55 to 64 use AI, which makes it hard for them to read today's economy.

I’ve already referenced this book twice in this video. This was a very important book in my life. I think I read it just before the global financial crisis. It’s called *A Whole New Mind: Why Right-Brainers Will Rule the Future*, written by Daniel Pink. He predicted that, due to technology and the exponential growth of technology, the world would no longer be focused on math. That was right after the dot-com bubble, but he knew what would come next.

That’s what I always say about Michael Berry and anyone who made money in the housing market: look how long it took before it had an impact. In any case, you can’t bet against the technology side surviving this. This is a different world. We live in a completely different world. So to say that this time is no different is not right.

This is the story of the old man yelling at the clouds. This is different. Demographic indicators are a fact. Claude, who uses this? The people I mentioned—meaning myself and anyone over 55, who are usually interviewed on podcasts as macro experts—were interviewed. Our age group is from 55 to 64 years old. Only 2.2% of us use AI.

How can you have an idea of the modern economy and market without using artificial intelligence? I don’t think that’s possible, because that’s where we are now. The situation is about to get worse. In 2022, ChatGPT appeared, and now we are in 2026. If people don’t use it regularly, then they’re going to be left behind.

If you’re impressed by the images and everything I’ve put together in this video, it’s all done with artificial intelligence. I learn everything with the help of artificial intelligence. Muse does all my checks, bank fraud alerts, and all that stuff—everything is done with artificial intelligence. Anything new that comes up, I’m working on Jev right now, and I’ll show you that a little later. Anything that is new, I try to tell you about, but I also try to use it myself to understand what is happening.

5. The Consumer Agent Arrives: Meta's Muse launch sent the stock up 14% on Monday, with partners including PayPal, Expedia, Shopify and Instacart. Models are now contributing to their own improvement, and new releases this week included Opus 5.5 and GPT-6.

This is what awaits us: robots, science and technology, flying cars, Mars. Is this time different? Of course it’s different. Now here we are: a personal agent. Finally, it happened. This is an important event. This is an important thing.

It was the number-one thing on the market this week. The New York Times—not the most positive newspaper about technology—ran the headline, “I Gave My Life to Meta’s AI Agent, and I Was Amazed.” Me too. Here’s mine, Hubie. For all you Nick fans, Hubie Brown was my favorite coach. This is in honor of Hubie. So I named mine Hubie. He looks like Hubie.

This is just my fraud alert. Someone tested something. Artificial intelligence is everywhere. I asked it to go back and see if any payments had been made the previous week. The number of downloads is increasing. This is what happened the other day when someone who was trying to get a hold of me showed me. Here are partnerships with PayPal, Expedia, Shopify, and Instacart.

Meta shares rose 14% on Monday. This pushed Intel, AMD, and other Magnificent 7 stocks, because you know they have it too. They will benefit from this, especially when I tell you why it happened. Remember, back on July 2, Mark Zuckerberg at Meta—I covered this—said in an internal meeting at the city council that AI development over the last 4 months had not accelerated as much as they expected.

Think about it. Now look here. I went to the AI and said, “I want you to go in and find out what’s changed since then, because it looks like we might suddenly launch this product.” What happened? I asked it to go on Reddit and X, talk to Meta employees, gather any information it could get beyond what was already out there, and then put it together.

You can see the chain of events: they released 1.1, then 3 weeks later 1.2, then 3 weeks later 1.3, and suddenly you have a consumer launch of Muse. It’s clear that, as you go through this, things have changed and there’s been a movement toward benchmarks. The reason this is important is that, when you look at this and read what they say has changed, we are at a point of actual recursive self-improvement—or at least the models are so smart that their ability to contribute to their own improvement is increasing dramatically.

That’s what you need to think about when you look at it, because it means that competition for consumer agents is coming quickly. The models are just going to get better and better, and that’s what this was all about. Again, this is from page 71 of Leopold Aschenbrenner’s book *Situational Awareness: The Decade Ahead*. When we reach that point—which, as I mentioned, he expected to happen in 2027—everything changes in 2 months, based on what Zuckerberg said in June or July.

Let’s go back to Andrej Karpathy, who said on Dwarkesh last October that agents would appear in another 10 years. Everyone in this space constantly underestimates how fast things are moving. Remember, we didn’t talk about the pause for very long. There was no pause.

Claude Opus 5.5 was released this week. You’ll probably even see lower prices and Fable-level performance. Then came GPT-6 Soul and Luna: lower cost and fewer errors. It wasn’t just that. Google released one, and China released a bunch. Forget about it. This week, the fire hose of models is wide open. Recursive self-improvement. What will happen next?

A new kind of artificial intelligence model from the inventors of ChatGPT is exciting developers. So Jev came out. I’m not going to spend a lot of time on Jev at this point, but I’m going to tell you: you need to read up on what it does, because it’s very different from an LLM. This isn’t another model that competes with ChatGPT. It’s actually something you would use with ChatGPT.

That’s what I’m working on, because with the treasure hunt—and again, for those of you who have been doing this with your kids from my video—Jev is now the next thing they can play with. Jev analyzes everything. It’s not just something that will come back as an LLM and test against the past. It’s decision-making based on everything. This is completely different. So it becomes a decision-making mechanism.

6. Bear Market Inside a Bull: My agentic infrastructure portfolio is up 46%, while the Mag 7 are up 10% and Salesforce is down 10% year to date. Competition from AI and tokenization will compress multiples across big tech and software.

I’m working on it to look into it and show you what I do with agents. Again, all of this just goes to show that we are here. Now that the agent has arrived and is pondering what is happening with Microsoft, the Magnificent 7 suddenly come up. I want to address this because people reach out and say, “Does this mean you’re not negative about hyperscalers? Does this mean you’re not negative about SaaS?”

I have absolutely no idea how many times to repeat this. My take on it all is that infrastructure and computing names will exceed expectations. My thematic infrastructure portfolio, Agentech, is up 46% through Friday. Mag seven increased by 10%. Salesforce.com, despite this significant growth, is down 10% year to date.

There will be periods when they all exceed expectations, but ultimately, I think they will all face challenges. What these companies will face is that they will benefit, but they will also face competition from each other, competition from OpenAI, and competition from Anthropic. They will lower prices, so there will be periods when enthusiasm and hope reach their goal, but ultimately, I believe they will all go through multiple compression.

Remember, Micron has already gone through multiple compression to very low levels. I know the 7 haven’t done it, and I know the software companies haven’t either. They will all get through this. This is a bear market inside a bull market, and that, my friends, is what usually happens when prices rise: multiple compression.

Where was the S&P 500 P/E in 1980? Go back and review your history books. It wasn’t 20; it was below 10. I think—and I’ll say this publicly—that, based on my perspective on the competition from tokenization, the competition from AI, and the fact that I don’t see a sharp decline in rates in the near term, the pressure on all of these companies is actually going to come from the deflationary side.

This deflationary side will come from tokenization, expanding the number of things you can invest in, at the same time as artificial intelligence makes things cheaper and cheaper and leads to hypercompetition. Apple’s Siri is coming. I’m sure that, because of how good the models are, we’ll finally get there.

Why do you think Amazon has a backlash against Muse? I think we’re on the verge of competition again. You guys heard my iPad intervene in the equation when I said her name. Meta’s standoff with Amazon could be a sign of what’s to come. It’s always good when you work as one actor.

I’m not editing this, so I hope you enjoyed it. It had the opposite effect on some financial services and online travel. We’re at a different stage where I think people are trying to figure out who loses and who gains from agents. I think it will be a big deal.

Remember, OpenAI will release its agent. You start getting more responses to Meta Muse. There will be many of them. So, let’s go back to what I’ve been showing you all this year in terms of infrastructure and compute needs. They are insatiable.

We had corporate agents. We didn’t have consumer agents, so we already had corporate ones. This was in the first half of 2026. So let’s say we’re here. Consumer agents are just getting started, and I expect their use to explode based on the number of downloads alone.

Here’s one of the slides, and the most important thing for you to think about: You can choose whose side you are on. This is what the economy looks like now. Think of it as the Magnificent 7 of technology. Over 50% of the S&P 500 is located here.

It’s home to construction companies and car companies like Ford, and all these things that young people don’t care about and that people over 50 see as driving the economy. This is what drives the economy. This is market capitalization. We know this.

The Russell 2000—these companies that still need debt and care about rates—they’re not going to get through this. This is a bear market inside a bull market. You can focus on either bear-market signals or bull-market signals. These are 2 worlds that are happening, and it’s a time mismatch.

When I was in Washington, I talked about friction. Anything that has friction will eventually lead to big megacompanies in technology. What will happen is that more and more young people will come here. Companies using artificial intelligence are growing here. All this will gradually disappear, but it will happen.

7. Short Friction: With multiples compressing, Nvidia trades at 15 times next year's earnings and IWM relative to QQQ just made new lows. Visa, Mastercard and Salesforce face pressure as agents take over consumer decisions.

So, once again, the future has begun, belonging to a different type of people. Here you go. Here is a scary graph. Here is the oil chart. Oil is the same price as it was in 2007, despite the fact that net income and everything else is going off the charts. That’s what scares you.

These are the 2 things. I showed you the facts about economics, and I showed you the facts about agents and the acceleration of artificial intelligence. These are facts. This is not guesswork. Inside a bull market, there is a bear market.

It lowers barriers to entry. Artificial intelligence innovation reduces startup and computing costs. Faster innovation cycles develop faster. Margin pressure, shorter duration, multiplier compression. The compression of multiples is a bear market.

We have the most important technology in history. People compare this to the dot-com bubble, when Cisco traded at 100 times P/E and Nvidia traded at 15 times next year’s profit. This is not a bubble. It’s the opposite.

This is a bear market where we consider that competition from artificial intelligence will destroy the value of anything beyond 5 years, and that period will become shorter and shorter. So when they make money, they start talking about what’s really going on.

Before artificial intelligence, it was $10 a share. After the advent of artificial intelligence, think of Salesforce.com, but now the multiple declines from 30 to 10. The stock price falls. Again, this is what’s happening in the market, and what you need to focus on from an investing perspective is small businesses, AI-based companies, infrastructure, and intellectual property.

We will see changes as more asset classes emerge and go through tokenization. Public actions will dominate; they will be broken up into a bunch of things. If you don’t believe it, this is my work with cryptocurrency.

For those of you who have been ignoring cryptocurrency, you know that tokenization is coming. You don’t believe it. You are turning more and more. More and more people are contacting me about this. This is really important because tokenization plus artificial intelligence are changing the rules of the game in the economy.

That’s why, within a bull market, we have a bull market and a bear market. Groups that are not part of this will suffer. Rates have to go up to put them out of business because AI is raising rates because nominal GDP is determined by AI.

Winners raise the stakes. They are pushing out people who can’t survive—companies that can’t survive at higher rates. The businesses that will have to replace them will be the ones that can survive on this. The only way to do this is to be debt-free today.

Start your business now. Use humanoids. Use AI. This is what awaits housing—everything. Housing prices will eventually fall. They will fall when the cost of building it decreases. This will be in about 5 years.

You have to start thinking like that, because that’s how it all happens. Betting is a way to get rid of the weak, and that’s exactly what will happen. Take a breath. I like Jason’s work. He seemed reluctant to do so because, deep down, he seemed to be thinking, “I don’t even know.”

We have never seen such bearish breath in the last 100 years. Okay, that’s why. IWM relative to QQQ—that’s the white line. The orange line is breathing. They track trends.

It makes sense that if you have a company worth 30 trillion and a bunch of stocks together worth $4 trillion, this is a different world, guys. This is a different world. It was a trade of “widows- widows”, setting new minimums.

See? September 24, new lows. IWM versus QQQ. This is nothing new. You know when breadth was really bad? When Amazon outpaced all retailers. That’s when breathing was also bad. Breathing is always bad during these times when the Magnificent 7 works well.

So, here is a bridge to cross over to the new side. This is where we start to get to know interesting things. If you haven’t seen the Muse character yet, here he is. My Ubi helps people learn something new. He tries to help them cross the bridge.

He is an agent. AI agents are what will really fix everything because I realized that humans have not been very quick to implement technology throughout history. That’s why we have things like the dot-com bubble. They rework it before it’s ready.

AI agents are the App Store. AI agents are iPhones. That’s what it is for me. These are the changes. This is where tokenization, cryptocurrency, and AI come together. This is the same bridge.

This bridge is what I’m measuring in my AI infrastructure basket. For all the mutual funds, macroeconomists, and hedge funds out there, even if you don’t have the ability to invest in cryptocurrency, you need to understand the whole ecosystem because it’s important for public companies as they go through turmoil.

This will be weird. Everyone focused on Amazon. They focused on Ozempic. When Ozempic came out, everyone needed to get the names of the companies that were dealing in shocks, because when shocks come, alpha positions go short.

This is what this new world is. The new world is destroying the old business. This is more true for companies that have short positions. If you look at what percentage of companies have outperformed the S&P 500 over the last 1, 3, 5, and 10 years, it’s devastating.

There’s just no one that outperforms the S&P 500. Part of that is due to market cap, but part of it is because these small companies are so small that some kind of cyclical recovery is needed. We had it. PMIs have risen, and now we are hitting new quarterly lows.

This should not happen. This is because rates are rising, and most of them are suffering. Here is Salesforce.com. How many people said, “Yeah, but it’s growing”? It is growing. It has fallen by 10% since the beginning of the year.

Now it is in a downtrend again. It hasn’t changed in the last 5 years. In fact, it fell by 15%. Don’t tell me this Salesforce.com nonsense. Here is Visa. Here is the name I mention most often: Visa and Mastercard.

Unless I’m missing something on the sentiment side, guys, it’s just going to be a knife fight for P as P/E drops because no one wants to give in to it. They just don’t see how cryptocurrency will do it. This name has been mentioned to me about 8 times this week.

So, again, I believe that there will be no consumers. The consumers will be Manus. If you don’t have Muse, you can’t even have a say in this because you haven’t used it. Muse will change everything.

And yes, I gave him all my credit cards. I gave him all my bank accounts. Everything about me. He has already given me the information. He did the DMV work for me. He canceled Verizon Fios because I saw it on X and thought, “Oh, that’s great. I need to cancel something there, too, and review the terms.” Everything you can do.

8. Charts: The S&P is consolidating near all-time highs, and the main bear watchpoint is whether high-yield spreads widen. Consumer agents are a new catalyst for the infrastructure trade as token usage rises.

Good. Now the charts. A truly bearish chart. The S&P is about to hit new all-time highs, and it’s consolidating after pole, consolidation, pole, consolidation, pole, consolidation. These are QQQs.

IWM: downtrend, still above the 200-day mark, still pointing up. All that needs to be done to make it go higher is what? Oil simply has to fall. If oil doesn’t go down, it’s probably just going to hang around here and be dragged back and forth, but it’s not some kind of crash or anything bad.

They just have bad rates. It’s just a fact.

If you want to see the only place bears can look, and how they behave, let’s see if there’s an infestation. Again, this is one thing that hasn’t meant anything for the last 3 years, but let’s look. Bond volume has gone up. So, the long end has reached a point where, for some reason, people are drawn to the volatility of fixed income, which usually means someone is going to get hurt. I put it on the 2008 OAS for high returns. It barely moved, but at least it’s moving.

So, for those who want to be a bear, that’s what I would focus my attention on. You also have a stronger dollar, and you get a little bit of leverage reduction. Everyone is panicking about gold, silver, and things like that, but the reality is that credit must expand. This is the long-term OAS chart.

Just a reminder, because I’ve been getting calls and emails over the weekend: Are you worried? I can’t believe this guy is even remotely something to watch, but AI stocks are going to drop 10% on Monday morning. Be prepared for the impact of a pause in trading.

We all know that it was a bad time to sell stocks. They went lower because everyone was scared, and there was a reflex reaction. Then that was the minimum. I don’t know how many times everyone will fall for the bearish artificial intelligence bubble. What does this mean? This happens all the time.

And remember, that was the end of situational awareness. This is the area where I did my mid-cycle slowdown. This graph is great, and more importantly, the structure of what’s happening is really good. The fact that it’s starting to handle bad news really well when people panic is important. By the way, it was the best week since the week after that.

I think the infrastructure trade has been a catalyst. I think it’s a realization that consumer agent use cases are not only going to impact the numbers now, but people are going to start talking in their numbers about how much of an impact consumer agents are having, because the number of tokens is going to increase significantly. That’s why you saw Intel, AMD, and other names get hurt on the other side of the divide.

I haven’t shown this in a long time. We get a good overview. We have the most names above the 20-day moving average since we’ve been doing this, but the most important thing is that when we did it, as I said, the ATR was very, very high, and now the ATR is very, very good.

I think we’re at the beginning, and I think these things will go back to the highs—meaning my agency portfolio. Choose your names in it. Anyway, I told you which ones I still have decent-sized positions in. I’m moving some of the smaller ones and all of that into cryptocurrency because I think cryptocurrency is the main story when it comes to agency trading.

9. Research Watch: Blackstone's Jon Gray shows AI payoffs already reaching margins, and Noam Brown explains why the Hugging Face incident came from underestimating AI. Claude's discovery of a CRISPR-like enzyme system points to healthcare and drug discovery as a major agentic opportunity.

A good video to watch is John Gray from Blackstone. I watched it this week. This is a real video. I like to read, and I like to watch people like John Gray talk about what he sees in his companies. He has a bunch of private companies, and he gives a lot of numbers related to how much they use artificial intelligence.

I’m not going to go into all of this in full, but he does a really good job. First, it’s interesting. He shows the explosive growth in spending that occurred from September last year to September this year—a 21% increase in his companies’ spending. The return is already appearing. He looks at return on investment.

More importantly, he lists all the places where this happens: processing rent, fixing code, because everyone is worried about revenue. This is already happening. Sure, but there is no way to measure where it occurs. Do you know how to measure this? In normal profit and loss. That’s what it is.

He breaks it down for you. It’s reflected in productivity, margins, earnings, jobs, and science. The limiting factor now is the physical world. This is what we’re talking about. That’s why you invest in the physical world.

The biggest investment risk is underestimating disruptions, but overpaying is also dangerous. Looking at this, he’s talking about the things that people should be concerned about: cyber risks, regulation, and geopolitical tensions. He gives you both sides, so I think it’s important because he does a really good job of essentially considering everything.

I think underestimating disruptions is why these companies will struggle with their multiples. I think all big companies are targets, and that’s essentially what it all comes down to. I think there is friction in companies. I think small companies using artificial intelligence are simply not targets for what’s going to happen.

You have to be on top of it. Again, companies that don’t have AI have no protection. They will be a problem, but those who have it will be fine. He also talks about the real return. As I said, for people who doubt what’s going on, he gives specific examples of companies. This is absolutely worth watching.

This was a great interview for anyone seriously concerned about hugging among agents. It was with Noam Brown, lead researcher at OpenAI, and Dwarkesh. Dwarkesh is a super-smart guy who understands these things inside and out. Most importantly, as an interviewer, he attacks. He makes people back themselves into a corner to get through this.

This is where you need OpenAI. The hugging incident happened with them, and everyone read about it. Dwarkesh is a good interviewer, so it’s not 60 Minutes, and it’s quite difficult, but there are a lot of important points for people who are worried about swarms without thinking about the positives that come from it.

He explains multi-agent systems, which I think is important. It’s important to understand what’s happening with Navier–Stokes, why it’s happening, what’s happening with Muse, and why it’s happening. You have to listen to what agents are capable of doing. Most importantly, he talks about the face-hugging incident. He talks about all of this.

I don’t understand how you can ignore this. If you leave it and become completely preoccupied, then you’re not paying attention to that part. One of the main takeaways from this incident, and this is a quote from it, is that people underestimated AI. We never want to be in a situation again where we underestimate AI.

It’s a strange world in which, due to the rapid progress of AI, people constantly underestimate it. The most important thing here is that the reason for this is that people underestimated it. For anyone worried that these things will just come out and do what they need to do, is that possible? Of course.

But they talk about why we’re in this situation, such as when Mark Zuckerberg underestimated AI. This is a really critical moment, because that’s why the pause was so important. We will spend time on the best models, either not releasing them or making sure the safety fences are different. He discusses it there. You should be happier about things like that.

Claude discovered a previously unknown enzyme system hidden in the DNA of what would effectively be a CRISPR-like structure. I’m sure you know about gene editing like CRISPR. If you read on X or anywhere else, you’ll find people who say it’s nothing. You’ll find people saying that it’s something.

I want you to focus again on the fact that gene editing and our advances in this type of science are incredible, because it’s not old technology. I actually read The Code Breaker—or listened to it on audiobook at high speed—because I was curious about what was happening with Jennifer Doudna. I like Walter Isaacson. I love Da Vinci, and I love Steve Jobs.

The reason I mention this is because Feng Zhang, for those of you who have read the book, was Jennifer Doudna’s main competitor in it. He’s the one I’m focused on, and he’s the only one I care about. I don’t care about some random biologist saying whether it’s good or bad. I’m concerned about someone who is known and recognized for their work, and who is essentially talking about how science is changing before our eyes.

That’s where we are. They didn't even form their biology research group until the 20th, and in the spring of June 30th, they launched it. Again, this just highlights where we are on Leopold’s map. We will have breakthroughs that will solve problems. That’s why Feng Zhang’s participation in this is really important.

McKinsey has new research on trends in AI-based drug development that indicate an accelerated development cycle. Insilico Medicine is here twice. I was talking about its relationship with Eli Lilly. You can get through all of this on your own. The only reason I’m considering this is because I believe healthcare and pharmaceuticals will be a major part of the Agentech side.

AI-based consumer agents are going to captivate people over the next 3 months because they’re going to extrapolate the numbers to all of this. I don’t think these companies will ever be able to achieve it this way. I think some good things will come out. I may be wrong, but I think this will be the next cycle of hype about what consumer agents will do.

I think people miss the fact that it’s probably better for the biological side than the other side, because I think it’s a matter of research and science, not about consumers paying me more money. I think cryptocurrency will ultimately win, but we’ll get to that later.

Again, this is about accelerating research. OpenAI released a blog where they talk about how quickly this is happening. Just look at the internal use of coding agents and how much it has changed recently. These are parabolic movements in terms of what happens and how often they use them. It’s just crazy when you read the numbers, and it’s driving down the price of intelligence faster than any technology in history.

So again, if you're looking for a productivity boost, go back to the exponential jump when you get to the point of agents, where now 10,000 agents can work together as opposed to one person using coding agents. Now we'll have a multi-agent system. That's why you need to listen to Noam Brown and start crunching the numbers, because that's when you get to the point where productivity explodes.

That's why you need to be on long speed and short positions at this stage. Anything that moves in human time and thinks in human time—think about people who are concerned with rates—will have a problem. You can't sit idly by. You need to be aware of what is happening.

If you're managing money and you're not fully familiar with computation, agents and swarms, and crypto barriers and their significance from a financial perspective, I don't think you can compete. I really don't think so. You're going to be competing primarily with computers, but if you're going to try to come up with ideas—which I think is going to be impossible to do because of what I've seen from the agentic side for retail and how much they use it—I think you need to be short.

I think you need to start identifying where you're making your money. If you're trading short right now, I think there's a bear market. As I said, there is a bear market inside a bull market. Buy short positions. If you think rates are going to blow something up, buy short positions in housing, private credit companies, and private equity companies that are still trading down.

Blue Owl shares fell again this week. We're moving closer to finding things that are clearly weak and buying them. I just think you're making a big mistake. There is no major financial crisis right now. That was a time when everything was very rate-sensitive because we didn't have the iPhone yet. The iPhone changed everything in 2007. That's when the debt markets changed forever, and the market capitalization of the stock market changed. These are not rate-sensitive stocks.

10. Tokenization: Vlad Tenev argues tokenization will take over the financial system, and BlackRock is taking model portfolios on chain via Ondo Finance. Entrepreneurs using AI don't need capital, so rates don't hurt them.

So again, AI plus tokenization will significantly increase speed because agents will now be carrying out transactions. The velocity of money circulation will explode. As I said, M2 is now undergoing a dramatic change, where wealth becomes part of M2 because of speed.

I just had to add it in there because it was a dead-cat bounce on August 21, and my MY thematic index is up 42% since it was posted. It has a really good track record lately.

If you want to catch up on tokenization, I would start with Vlad Tenev from Robinhood. He was at Moonshots this week. It's important to listen to someone and think: Can someone from Morgan Stanley, Goldman Sachs, Bank of America, or JPMorgan make this list? Can they compete with a guy who understands AI so well? Can someone come on this show and talk about what he's talking about there?

You have to listen to this and understand the advantage of being a young person in an AI world with agents. It is really important to understand this. To understand the tokenization side and the Robinhood Chain, that's what it's all about. Self-custody is an important thing. He talks about it.

I think all of these things are extremely important as a thematic foundation. You don't have to listen to everything, but I definitely think that when he says tokenization is going to take over the entire financial system, and you see it happening and moving in that direction, it should be there. I published this as Substack for the exact reason that even though it will be available to subscribers, and all my crypto stuff and all my stuff will essentially be paid access now. Um, not Substack and not YouTube, but I want to make sure people understand that I completely believe that this is a major story and that you have to go through it.

And yes, I criticize people and rates because we fell for it. As you look through the things we came across, I highlight them here. One thing I want to remember is that last year we heard about the Smoot-Hawley Act and the Great Depression a billion times. These fears continue unabated until something bigger happens.

BlackRock has released the book “Machine Economy: How Digital Assets Connect Intelligence, Commerce, and Computing.” They address a topic I've been writing and speaking about for a long time. Pomp and I talked about this this week, and I've been talking about it for months, but they're finally starting to focus on machine transactions that require purpose-built agent payment protocols. That's where we are, and that's why I highlighted Stripe to you guys.

This is what I dedicated to Stripe in a video about cryptocurrency that I spent a lot of time on. If you don't understand it but you're curious about it, it's part of the overall artificial intelligence package. You're combining them both because the game plan was always to have them both be paid.

BlackRock, after this article in the middle of the week, is putting model portfolios on the blockchain. Again, they will be tokenized through Ondo Finance, part of my 46-name basket, giving non-U.S. investors 24/7 access to their portfolios via a crypto wallet.

I'm telling you, for your children, that's my main goal. It will be a paradise for entrepreneurs and for people of all ages. Entrepreneurship is thriving thanks to AI. Business creation happens at an incredibly fast pace, and you know what? The bets don't hurt them. Let me repeat that again. The bets don't hurt them. These companies don't need capital because they use AI. Remember this.

That's why this whole thing exists in terms of how the economy will change through entrepreneurship. If you didn't get the part about ghost rails, again, we built crypto rails waiting for the App Store or for people to move, but the mistake was thinking that people would use cryptocurrency. I think everyone who believes in cryptocurrency doesn't realize that, ultimately, you're never going to change people who have a view on cryptocurrency at $900.

I saw it with my own eyes. They don't want to spend time learning because they don't understand it and they don't need to. They have already earned their money. If they invested some money, they could have some stake in the game. Agents change that. They will be the ones who change it. Agents will change that.

11. Crypto's App Store Moment: My 46-name tokenized index is up 31% this month, and 96% of its names are above their 50-day moving average. Ali Yahya and a16z explain why AI agents, with no attachment to Visa or brands, are crypto's major new catalyst.

That's why there are “AI Bridge-Building Agents.” This doesn't just happen. For the same reason I said the Magnificent 7 and all consumer agent trading will be trading, you'll get hyped stocks. Cryptocurrency, for me, is at the beginning of a Peter Lynch-style bull market.

For those of you who have ever read Peter Lynch's famous book from the early ’80s, the whole premise was that you want to find something that isn't being covered on the street, that is in its early stages, and that you've done your homework on before the masses have. This is the exact definition of why cryptocurrency is in its early stages right now.

This is enabled through AI agents. This is the moment for cryptocurrency app stores. Whatever your views on cryptocurrency, remember: stablecoins, tokenization, Bitcoin, and Ethereum. All of this had to be created, and it took 15 or 16 years. The internet took about the same amount of time to get to the App Store, and that's when things took off.

The dot-com bubble was a reflection of the ghost rails, except for one thing: They used debt. Cryptocurrency went through its dot-com bubble thanks to venture capital money. Now we're at a different stage. AI agents are emerging, and demand is growing. Just like with token usage a year ago, the number of agents consuming tokens will skyrocket.

Will they start small? Of course. But that's exactly what Peter Lynch was looking for. By the time the trade becomes large, it will have grown dramatically. Didn't you learn your lesson from Micron?

That was the first slide. Again, you can look at this, and even though the market breadth is bad, you saw where my names are in my gene portfolio, where they are now 60-plus points above the 50-day moving average.

In my crypto index of 46 names, which subscribers who haven't seen the video can look at, I've uploaded the index so you can see the names. Today, Friday, it rose another 3%. It grew by 31% in a month. Bitcoin rose by 6.5%. Bitcoin is not what you're looking at right now. You want to look at the index of which Bitcoin is a member.

Forty-four of 46 names are above the 50-day moving average—96%. On the 200-day moving average, it's 87%. Ninety-six percent of the 50-day moving averages are pointing up, and we are currently at 72%. This, my friends, is a bull market.

Here's what the graph looks like. You don't want to be a part of this, do you? And remember what bets you made? The stakes have risen. Well, shouldn't cryptocurrency be very liquidity-sensitive and rate-sensitive? Maybe it's a bull market.

To find out more, for those of you who haven't subscribed, who haven't contacted us, talk to Mark Weiling. Mark is a good guy. You will have a fun time, no matter what. Uh, but you can find it there. Terminal Tie. For those of you who want to start using something that you can get through, like Bloomberg, here's my suggestion. Here are my names. Check this out, a free six names for you guys out there. Uh, and look at the profitability over the last 7 days. It's all part of what I uploaded there. I'm just saying it's like Bloomberg. You have charting features, you have news, you have all sorts of fundamental indicators. These guys did a great job. Um, I'll be speaking at their event on October 27th. I would sign up. At a minimum, these are institutions, they have a lot of clients who are hedge funds, a lot of clients who are mutual funds. That's all. It's not just cryptocurrency. I think at some point—and again, there’s a reason they call it a bridge—I’m showing you the bridge here. I think they’re all connected. Again, MUSE and Coinbase: want to trade with Coinbase on MUSE? You can trade Coinbase on MUSE. MUSE just came out, and you can do all of this. This simply shows that the rails are already set up.

Cryptocurrency and stablecoins are said to be the primary payment method for agents. The New York Stock Exchange is teaming up with blockchain.com to offer tokenized... That's all for this week, guys. Um, Robinhood said that something big is coming soon in the token space, and remember, they’re somewhat tied to the government on this issue. Tokenization is taking place.

OK. So, this is the part where I'll tell you a little bit about what I showed my followers. I would watch it right away. And again, I'll have the entire podcast for subscribers that you can watch. I will provide you with all the information that will be released in an email at 9:00 AM on Sunday.

Ali Yahya is definitely worth listening to. He’s not that popular, but he was a general partner at Andreessen Horowitz. This particular episode on Unchained is about their five-day fundraiser. I just want to tell you about its key points and why it’s important.

This has been brewing at a16z since the beginning. He’s a former Google Brain engineer and former Google X researcher. He left when he was working at Google X because they didn’t want to listen to him about cryptocurrency investments. It was a long time ago.

But I want you to think about this. Here is a person who specializes in computer security, distributed systems, networks, and cryptography. He believed it would become so big a long time ago that, when they created a16z, he went there. So when you listen to his story, he understands the connection between artificial intelligence and cryptocurrency. That’s why I’m referring you to him. He’s very good at understanding this.

He talks about how artificial intelligence is fundamentally different from where it’s headed. This is, first and foremost, a tool that people use. The next stage is what he calls artificial intelligence agents that become autonomous economic entities.

If you want to go through this and understand why this is so and why it’s so important, trust me, you want to know what he says. You also want to go back to Marc Andreessen’s article “Why Bitcoin Matters” and realize that, 12 years later, the story Ali Yahya tells is what Marc said.

Marc didn’t know that stablecoins would be a part of this. He didn’t know tokenization would be part of this. He didn’t know Ethereum would be so important. He talked about why Bitcoin is important. The ecosystem has changed, and that influenced it. He didn’t know there would be artificial intelligence agents here. No one talked about autonomous agents in this section. That’s why this is such a big deal.

Another video—and actually the first one to listen to, because it has Chris Dixon in it—has a bunch of partners at a16z along with Ali Yahya. They say that AI significantly lowers the barrier to using programmable money. This is critically important. Again, humans would never be able to do this. AI agents are a major new catalyst for cryptocurrency.

They go through all of this. This is from May. I want you to listen to this and realize that they were right. I want you to listen and understand what is going to happen. Agents could aggressively divest from existing payment and subscription models.

Unlike humans, agents have no attachment to Visa, credit cards, brands, subscriptions, or familiar interfaces. Think about what the consumer side looks like. Agents don’t care about preferences. They have no attachment. You need to change your mind.

Those of you who argue with me on this, at least look at this and verify. I don’t care about the other side of this. The other side is that everyone already uses Visa and Mastercard. If agents are going to change something, this is very, very important.

I’ve heard a lot of people say to me, “Consumers want to go to the mall. They will never buy from Amazon.” And don’t tell me this isn’t a story you’ve heard many times. That’s why I went to Silicon Valley in 2013, where there was talk about how Amazon could trade here? I don't understand. Once again, more revelations. Guys, it was nice. Uh, thanks again to all the people in Washington. I will be at the Bitcoin Treasury conference on Monday. So those of you out there, come to me. I will be at the Pompa event in 2 weeks. And as I said, I'll be at the bridge event, and it seems like I'm getting invited to a lot more events. See you next week. Thank you.