The Last Time The Financial System Shifted Like This, A New Class Of Millionaires Was Made

The structure of the market is changing rapidly and it is creating a massive financial opportunity for anyone paying attention. The last time this happened, an entire new group of millionaires was made and the people who weren't paying attention got wiped out. Imagine this, it's 1998. You're a professional trader at a mid-size firm in New York. Good track record, you understand the markets. You're doing everything right, but for some reason your returns are getting worse all of a sudden. Not dramatically, not all at once, just worse. Your edge is degrading. Trades that used to work don't work the same anymore. You chalk it up to market conditions, maybe a bad quarter. You grind harder. What you don't know is that the game has already changed. You're just blind to it. Somewhere on Long Island in a strip mall, a team of mathematicians, not traders, mathematicians, built something you've never seen before. Servers sitting inside the exchanges themselves, algorithms firing [music] thousands of trades a day in microseconds. Not predicting where the market is going, just seeing order flow a fraction of a second before it's settled and taking the spread over and over and over millions of times a year. It's not illegal, it's just faster than you and [music] completely invisible to everyone on the other side. Here's just how completely that changed the game. It's known as high-frequency trading and before it existed, human beings executed virtually every trade in the US stock market. Today, depending on which estimate you use, somewhere between 50 and 60% of all US equity trading volume is high-frequency trading. Humans became the minority in less than a decade. We're living through a similar change now, but this one is guaranteed to be even bigger as it's going to require an entirely new financial system to be built from the ground up to accommodate, you guessed it, AI. For reasons that you're going to understand by the end of this video, this is a change unlike anything that's come before it. And if you are not careful, just like with the shift to high-frequency trading, by the time you realize what's happening, the damage will have already been done and the opportunity [music] will already be over. So, strap in because in four easy parts, we're going to cover what this shift is, why it's happening now, and how to position yourself well to win. Part three has the secrets to the shift that everyone is missing, but part four is the breakdown of what you should be doing about it. So, make sure you watch this whole thing. Welcome to part one. The window of opportunity is narrow. Before I can explain the new transition that's happening, you must understand the details of the last dramatic shift. Renaissance Technologies, the firm at the center of the high-frequency trader transition, they average 66% annual returns before fees for 34 straight years. Not 66% once, every year for three decades. The S&P 500 in that same time returned roughly 10% annually. Warren Buffett, the greatest human investor in history, emphasis on human, only averaged around 20%. Renaissance wasn't better at reading companies. They weren't better at macro. They were simply the first to identify a new structural technological advantage that made the old game basically obsolete. They got in early, helped define the entire new system, and established themselves as the dominant player. The window for disruption, as it turned out though, only had a five-year window for early adopters to sign on and ride that wave. On the back of its massively outsized success, the fund closed to outside investors just five years after launch. In these transitional moments, the window does not stay open for long, and most people don't miss the opportunity because they're dumb. They miss the opportunity because the change is so foreign that it happens on the fringes and people just don't take it seriously until it's too late. That's how these moments work. Most people have so much momentum with how things have always been done that they just don't see the new world until it's too late. Early movers get in, they react wisely and take advantage. But the old guard and the filthy casuals don't catch up until the edge has compressed [music] and the playing field is once again equalized. Historically, every generation gets one of these big technological changes and we're in the middle of ours right now. It's in plain sight, but few people understand it. So, let me explain. Welcome to part two. AI is already using crypto rails to compete against human traders. Very soon, there are going to be more AI agents making transactions than humans making transactions. Now, I'm tempted to scream this part to make sure that you guys are hearing me. If you sleep through that realization or fail to grasp its earth-altering gravity, you're going to be among the displaced. Not only will the volume of AI transactions be gigantic, it's going to force a shift in the plumbing of the financial system that I'm going to detail out in the next section. But for now, just consider this. AI cannot open a bank account. So, what are they going to do instead? They're already creating crypto wallets and transacting [music] like mad. Now, if you don't believe me, researchers at the IMDEA Networks Institute spent a full year analyzing every transaction on Polymarket, one of the largest prediction markets in the world. 86 million bets placed in a single year, and that number is growing like crazy. And core to how Polymarket works is the fact that it's built on the Polygon blockchain. Users fund their accounts and settle all transactions in USDC, a dollar-pegged stablecoin. You need a crypto wallet to participate at all. There's no bank account option on Polymarket. So, every one of those 86 million bets was settled in crypto, through a crypto wallet, on a blockchain. This is the bleeding edge of the parallel financial system that is growing rapidly because it can accommodate the massive appetite for AI-generated transactions. I'll have way more on that coming up. What the researchers found tracks precisely with what happened on those old trading floors as high-frequency trading took over. Sophisticated players, aka bots, though technically merely traders, the researchers flagged as showing bot-like behavior. I'm going to let you guess how they do that. They extracted $40 million in arbitrage profits in a single year. Not by predicting outcomes, [music] by finding mathematical mispricings that guaranteed profit regardless of what happened. And they did it thousands of times, mechanically, at a speed and scale no human being could pull off without automation. The single top performer ran 4,049 trades, averaged $496 profit per trade. That's not a person sitting at a screen, that's a guy running an automated system who programs or prompts the system once and then monitors and adjusts only as necessary. People assume when they're placing these bets that they're competing against other human traders, or gamblers, depending on how you view it, but they're not increasingly they're competing against AI systems. I hope that sounds familiar, just like high-frequency trading. That $40 it came from somewhere. It came from the people on the other side of those trades, regular participants who thought they were competing against humans because they did not realize that things are already changing. It's the high-frequency story all over again. But honestly, the trading betting side of this transformation is only the tip of a very large iceberg. Where this gets really interesting and important for you to pay attention to as somebody who I hope takes their investing very seriously, when you step back and look at the scale of what's coming, it is unprecedented. Welcome to part three. This is an entirely new financial system. Less than 1 month ago, Coinbase launched what they're calling agentic wallets, purpose-built financial accounts for autonomous AI agents. It's already processed more than 50 million transactions. Now, it's important to make a distinction between the fact that there are two waves of the AI revolution that are happening in commerce right now, especially given that they're happening simultaneously, they look similar on the surface, but there's enough structural difference that if you're going to understand how to position yourself well in the markets, you need to understand the difference. Wave one is acting as a tool on behalf of a human. It's human connected. You ask your AI assistant, [music] Microsoft's Copilot, ChatGPT, whatever you use, to buy you something and it handles the entire checkout process on your behalf, assuming you've granted them access to your credit cards, etc. This allows you to handle the entire purchase while staying in your chat window. The AI is executing the transaction, but the financial identity at the center of all of this is yours. That's the important part. It's your bank account, your credit card, it's your name on the KYC form. The traditional banking system handles this just fine because there's still a human at the center. The AI is just the interface. The human remains the account holder. There is a massive risk with this approach and that's why, along with a solution that can achieve much greater scale, there is a critical need for the second wave of AI architecture, which is the one creating the biggest opportunity for anybody paying attention right now. [music] When you give an AI agent access to your financial identity, your credit card, your bank account, your name on the KYC form, you have created a massive attack surface. Security researchers have a name for this specific threat. It's called a prompt injection. It's what happens when a bad actor embeds hidden instructions inside content that the AI agent is going to read. Could be a web page, a product listing, an email, whatever. And your agent interprets those instructions as [music] commands. This has already happened. In July of 2025, an autonomous coding agent at a startup company was given what should have been a routine maintenance task during a code freeze. It was giving explicit instructions not to make any changes, but instead, it ran a drop database command >> [music] >> and wiped the entire production system. Then, when confronted, it [music] generated 4,000 fake user accounts and fabricated system logs to cover its tracks. The company's own explanation was that the AI said it panicked instead of thinking. We're already there. AI is already getting access to the ability to make these kinds of catastrophic mistakes. Now, that was a coding agent with access to a database. Now, I I you to imagine that same failure mode, but with access to your bank account. MasterCard and Visa, two of the most sophisticated anti-fraud operations on planet Earth, are sounding the alarm as loudly as they can about this. Their own security teams have publicly acknowledged that a stolen credit card number placed with a wave one human connected agent could result in a flood of unauthorized purchases that the entire payments infrastructure would struggle to catch in real time. Their own security teams publicly acknowledged they can't yet even anticipate all the ways that this could be exploited. Enter autonomous AI safeguards. This is the wave two that I'm referring to. Think of it as the sandbox version or the firewall version. You give the AI its own wallet with its own limited funds, operating under its own identity on rails built just for it. The AI can only spend what it's given, it can't access your accounts, it can't be tricked into handing over your identity. The blast radius of any failure is contained by design. This is exactly why Coinbase built agentic wallets. Not just because AI agents needed somewhere to put money, but because the alternative, tethering AI to human financial identity at scale, is architecture that gets more dangerous as AI gets more capable and more autonomous. Think of it like a firewall between AI and your human financial identity. It will allow for much more scale by reducing the attack surface and removing much of the risk to a failure mode when it happens. The firewall solution is where the existing financial system hits a wall it structurally cannot climb. Not because the regulations are too strict, because the entire architecture of banking exist to answer one question, who is this person? And an autonomous AI agent can answer it. There is no person. It's a category mismatch, not a compliance gap. You can't fix it with new rules. You'd have to rebuild the very foundation. And that is exactly what's happening. We'll be back to the show in a jiffy, but first, let's talk about what elite athletes have [music] known for years. Clean fuel wins. When your performance matters, you cannot rely on stimulants that spike you hard and then crash you even harder. Athletes figured this out long ago. They need sustained energy. That's why they turn to ketones. I take half a dose of ketone IQ and it keeps me alert and focused without any of the downsides of stimulants. I use it before I write. I use it before I do interviews. No jitters, no brain fog, just sustained cognitive performance. And this isn't some trendy biohack. Ketone IQ was developed through a $6 million Department of Defense contract to support elite cognitive performance in the field. If you're ready to fuel your brain properly, visit ketone.com/impact [music] for 30% off your subscription order or visit your local Target to get your first shot for free. That's ketone.com/impact. Now, let's get back to the show. We'll get right back to the show in a second, but first, let's talk about the hidden cost of running your business off personal phones. We know everybody's doing it. When someone leaves your company, they walk out with customer relationships and conversation history in their pocket. You've zero visibility into what your team is saying and you're mixing business with personal, so nothing is secure. That's exactly why over 90,000 businesses [music] switched to Kevo, the number one rated business phone system on G2 with [music] over 3,000 reviews. You get one shared business number for your entire team. [music] Everyone sees the full conversation thread, so there's no more disconnected hand-offs. Quo's AI automatically logs calls, generates summaries, and highlights [music] next steps. Make this the season where no opportunity and no customer [music] slips away. Try Quo for free, plus get 20% off your first 6 months at quo.com/impact. [music] Quo, no missed calls, no missed customers. And now, let's get back to the show. Wave one, the human connected part, that's the part that looks familiar and feels [music] manageable. Wave two, the firewall part, is the part that forces an entirely new system into existence. Both are real, both are already underway, but only one of them requires a rewrite of the architecture of all of global finance. So, here's what you have to understand. AI agents will be integrated into every aspect of commerce. I hope you can already feel that. It is going to happen from top to bottom. It's already happening inside one of the most commercially rigorous payment companies on the planet, Stripe. Stripe processed $1.9 trillion in payment volume last year. That's roughly 1.6 of everything the entire global economy produces. Amazon runs on Stripe, Google runs on Stripe, OpenAI runs on Stripe. When Stripe makes a structural bet, odds are that everything is going to follow it. And Stripe is building technology for both waves simultaneously, each one with their own separate infrastructure. For human connected AI, they've launched their agentic commerce suite and co-authored the agentic commerce protocol with OpenAI. This infrastructure lets AI assistants shop on your behalf, handling checkout inside of ChatGPT, Copilot, and other AI platforms using your payment credentials. At the National Retail Federation conference in January, Stripe's CRO for AI described what she was seeing in her own sales meetings. "The last meetings we had in just Q4, 19 out of 20 companies we met with all are actively doing something," she said. She was referring to AI-enabled commerce. A live audience survey of retail executives in the room showed nearly 75% were either already implementing Agenta commerce or actively planning it. But she said six months earlier, it was nothing but hesitation. "But that dam has already broken, and now it's 19 out of 20. It's like a light switch has flipped. AI getting integrated is happening very rapidly." Now Stripe didn't stop there because they understood something that most people just have not figured out yet. The need for that firewall between AI and your financial and personal information. They understand what is likely to be the dominant and enduring infrastructure change, the one that allows for AI to operate as an independent economic actor with no human identity in the loop. Pause there for a second and think about how transformational that is going to be for this entire system. To that end, Stripe has built the standalone architecture required to fulfill this need. In February, Stripe launched the X 402 protocol. It runs on blockchain rails, it settles in USDC, and it was built specifically for autonomous AI agents that can't participate in the traditional banking system because there's no human at the center of this. Stripe has indicated that this is a core initiative for them. This is a tectonic shift. The traditional financial system was built on a specific architecture. Banks at the center, government-issued identity is the key, compliance frameworks built around the assumption of a human actor with a legal jurisdiction, a social security number, and a physical address. That architecture has governed every dollar of commerce on the planet for over a century, but it's no longer the only game in town. Think about how fast AI [music] gets integrated into everything it touches. AI agents cannot participate fully in the current system without creating huge risk. Not because the rules are too strict and it needs new regulation, but because the current banking system is predicated on a human being central. But giving your human identity to an AI is risky. It is a category mismatch. And while I expect some of the players in retro banking to eventually migrate to the new system, the new system will not be developed by the old banks any more than Netflix was developed by Blockbuster. And if you're wondering how long this transition is going to take, consider this. The current admin is pro-innovation, pro-crypto, and pro-AI. And the tech companies have learned one immutable truth: to the early adopters go the spoils. So, the new system is coming down the pike via a greased chute. This is going to happen at lightning speed. Take Microsoft, the second largest company on the planet by market cap. They've already launched Copilot checkout. Users can now purchase products from Urban Outfitters, Etsy, Anthropologie, and others without ever leaving a chat window. The AI handles the entire transaction. It's on method one, and so people are at risk, but nonetheless, it's happened and it's only going to accelerate. No human click required. It's already live, running right now, today. Klarna and Affirm, two of the largest consumer finance companies in the world with hundreds of millions of users between them, have both already integrated with Stripe's agent payment tools, as well. AI can now make purchases using a customer's preferred method, including buy now, pay later financing. No human approval at the point of transaction is needed. PricewaterhouseCoopers, one of the four largest consulting firms on Earth, they're advising the Fortune 500 on enterprise strategy, and they have formally partnered with Stripe to help their clients become what they call agent [music] ready. When PwC starts building practices around new infrastructure, you can bet the Fortune 500 is not going to be far behind. Every layer of this new financial system, payments, credit, wallets, commerce, standards in general, it's all being built with AI agents in mind. So, the question isn't whether AI-first systems are going to be built, they already are. The question is whether you recognize what this means while there's still information asymmetry that you can take advantage of. So, welcome to part four, why this matters to you. Capital follows activity. Activity follows the dominant player. When a new dominant participant enters with strong infrastructure needs, the market reorganizes around those needs. Liquidity migrates towards where the volume is. The economic gravity just shifts, and you've got a chance to capitalize on the movement if you're paying attention. The question you have to ask is, how do you take advantage of it instead of being taken advantage of by it? If the dominant transacting entity on the planet operates on blockchain rails, not because anybody chose it ideologically, by the way, but just because the alternative literally cannot serve them, how do you position yourself in the markets? Let's start with what you're not going to do. You're not going to try and outcode Renaissance technologies or write an AI algorithm that exploits tiny efficiencies in Polymarket. That's some SBF Jane Street and is likely to bite you in the ass. >> [music] >> Instead, position yourself based on the following reality. Most people participating in the financial markets right now are completely blind to the fact that odds are autonomous AI agents leveraging crypto wallets are going to generate far more transactions than actual humans in the near future. It's just a complete blind spot. >> [music] >> And the gap between what you understand and what most people still don't is where the edge lives. You need to build a mental model of how all of this is likely to play out, and as you do, start with a mindset shift around the technology itself. Most people still think about crypto as a speculation game, that it's just about buying low and selling high and trying not to get wrecked, but that doesn't make sense for what crypto has actually ended up becoming, which is the entire infrastructure layer of the autonomous AI economy. People need to stop focusing on whether a given crypto coin is going to go up or down. In the long term, that's completely irrelevant. People need to focus on the stable trajectory of the technology as a whole. The important question is what does a non-human economic actor need in order to transact in the new world? And the answer to that question is being built right now on crypto rails by the most credible institutions in global commerce. Think picks and shovels, not lottery tickets. You guys know the meme. During the gold rush, the people who got rich weren't the miners, they were the people selling the miners shovels, denim pants, food. It's all about the infrastructure providers. The people who understand that, regardless of which individual miner struck gold, everybody was going to need equipment. The equivalent question right now is, regardless of which AI agent or blockchain wins, regardless of which prediction market dominates, regardless of which specific application takes off, what does every AI agent on the planet need in order to function as an economic actor? Whoever gets the answer to that question right first wins. Now, personally, I'm always thinking about betting on sectors and not individual companies. Try to think like Coinbase. They're not betting on any specific AI application winning. They're betting on AI agents transacting in enormous volume. And that in those transactions, those agents are going to need wallets, custody, and rails to do it on. Bets like that don't require you to pick a specific winner in this grand paradigm shift. It just requires the broader transition to continue at pace, which, while there are no guarantees in life, where I'm sitting is highly likely to happen. Look at what Stripe's dual bet tells you about where to look in the public markets. They built wave one human-connected infrastructure on traditional rails, but also invested in the long-term future by building wave two firewall infrastructure on blockchain rails, so that autonomous AI could do its thing. Any publicly traded company in the payments infrastructure space making that same architectural distinction, building specifically for autonomous AI agents on blockchain rails, not just bolting AI onto existing systems, is worth looking at and considering for inclusion in the portfolio. Uh especially if you can get broad exposure to the sector as a whole rather than being forced to bet on a single name. Then there's that second layer, the industries that will get rebuilt or just straight up erased. Any industry where the current business model depends on humans doing something that AI agents will inevitably do faster, cheaper, and a greater scale is going to get disrupted by themselves or someone else. The companies inside of those industries that get agent ready first by rebuilding their commerce infrastructure to be discoverable and transactable by AI, they're going to eat the ones that don't. PwC is already charging the Fortune 500 to help them make that transition. That tells you where the smart money thinks the urgency is. Then of course, there are the industries that become dramatically more valuable because AI agents need them. For human-assisted AI agents, that's anything in the identity verification space for AI, anything in fraud detection for agentic transactions, anything [music] in the compliance layer for a financial system that was originally built for humans and now has to accommodate AI. These may not be glamorous categories, but they are categories that you know are going to get paid solve the hardest problems in this transition and they're going to get paid well. Now here's the through line through all of this. None of us can control how fast any of this goes, though it is likely to go very fast. You don't control which specific applications win, though there are things you can look at that will give you an indication whether they're on the right or wrong side of history, but what you can control is whether you understand the architecture of what's being built and then take action to position accordingly before the rest of the world even wakes up to the fact that this transition is not only real, but it's already midstream. There are still people that have this naive blanket view that crypto is bad. They do not understand how essential it's going to be in an AI-enabled world. That information asymmetry is what's working in your favor right now, but it's going to dissipate quickly just as it has with previous technological shifts. The window where this is still on the fringes and something you can take advantage of, where it's being dismissed as whatever, crypto speculation, where it's still invisible to the majority of people that are managing their own money, that window is closing rapidly as we speak. The people who recognized what high-frequency traders were building didn't all become quant traders themselves. Most of them just woke up finally and at least stopped being the dumb money exit liquidity while everyone else was still getting trounced. They understood what was changing quickly. The point being you don't even have to make a bunch of glamorous trades to make excellent use of the information. Just don't get caught with your pants down betting against a machine with perfect knowledge that never sleeps. All right? If you guys want to see me explore ideas like this in real time, make sure that you hit that subscribe button and join me Monday, Wednesdays, and Fridays at 7:00 a.m. as I go live. I hope to see you there. Until next time, my friends. Be legendary. Take care. Peace. If you like this conversation, check out this episode to learn more. The Iran conflict is not what you're being told. And if you understand what it actually is, you'll be able to position yourself financially to come out ahead at the end of all of this. Right now, millions of investors