If You Want To Create Wealth In 2026, Watch This (24 Stocks)

The next uh next opportunity is here. And even though everybody thinks that the AI cycle is finished, it could not be further from the truth. We haven't even started yet. The next AI super cycle is about to hit, and smart investors are going to do very well while 90% of investors will go down the drain complaining and moaning and blaming everybody else. Today, I'm going to give you 24 stocks that I'm watching for this next super cycle, and everything will be given to you for free. Nothing is behind a paywall, so don't click nothing, don't smash nothing, don't buy nothing, just listen and pay attention. Okay? Let's get to it. So, a few years ago, I was having a conversation with a lot of people about Nvidia. And people were explaining to me that Nvidia is just so darn expensive. Oh my god, Nvidia is 20 times sales. It's impossible to buy anything at 22 sales and make money, Tom. Don't you know that? It's way too expensive. Guess what I did. I was looking at the fundamentals. I was looking at the revenue growth. I said, "Well, look, 22 ain't cheap, for sure, but if the company actually explodes revenues, then it won't matter. The stock price will go up because in the long run, the stock price always follows the revenues, every single time. Okay? Guess what happens. 2026 is here, and earnings blew right past that multiple. Earnings have led the stock price to 5x from 2023, late 2023, to right now. Why? Was it luck? Is it market irrationality? No. Just look at the revenues. Look at the screen. 61, 131, 216. Revenues have gone up from 60 to 220, the price of the stock followed. It's that simple. Over focusing on multiples is a mistake that rookies do when they think they are the next Warren Buffett. Having multiples is half the story. It's like having a map without a compass. By the way, you can get a great compass through Stock MVP. Stock-dash-mvp.com. It's a platform I built for you guys to analyze every single stock in the world within 30 seconds with access to calculators, to insider buying and selling, to Wall Street analysts, every piece of information about a company you want in seconds. And you can customize it, you can make it your own. It's absolutely a product I could not be more proud of. So, if you take a look at Stock MVP, this is a little uh screenshot. And you look at Nvidia numbers. You could see it's not market irrationality. Look at what's going on with the revenues. Look at what's going on with the operating income. Look at the free cash flow. The stock price just followed the fundamentals like it does every single time. So, multiples are important to understand, but they're not the whole story. And the same thing happened with Palantir. I hate to bring it up again, but it is. People were telling me, "Well, Palantir is so expensive. It's 17 times sales in 2023." Guess what happened? Revenues doubled. The price 20x. Okay? Now, it's at 58 times sales. So, if you stayed away from Palantir at 17 times sales, now you're probably saying, "Oh my god, what a mistake. I just missed the 20x because I was over-focusing on multiples and not looking at the fundamentals." It is never a bad time to buy a great company. Never. Look at this. This is from my group, from Roic Academy at pekingnology.com/don-nash. This is a list that includes all the companies that I like, my top stocks list. And as you can see, we've added Nvidia in October of 2020. It is up 1,300% since. We have added Palantir on October 2020. It is up 1,200% since. So did Arista Networks, for example. It's not that massive game changers, and they will create generational wealth for investors. This presentation is going to show you those 10%. It's going to explain to you how to avoid the 90% of pretenders. And again, I have nothing to sell here. Everything will be laid out for free in this presentation. I'm not going to leave you hanging. So stay till the end because at the end, I'm going to give you the most interesting part. But check this out. The framework you need to have when you going into a supercycle, like agentic AI, is this. You want to have the story and the potential and the upside, right? But you also want to have already established fundamentals. You can't just invest in the story with the hopes of the fundamentals one day catching up. Both need to be there already. You need proof of concept, okay? So, the agentic shift is going to make a lot of money. Find the misunderstanding, make the money. I would say that. I think a lot of people still don't understand how big agents are going to be to AI, okay? Right now, think about how we utilize AI, even on the corporate level, right? A chatbot answers a question, okay? There's a GPU usage spike, and then it dies. There's no context. Everything is a one-time thing. With agents, it's literally 24/7, but it builds layer upon layer upon layer upon layer of itself. There's endless demand for context, for CPUs, for GPUs, and it's 24/7. It continues. It's non-stop. It's a whole different ballgame, which needs a whole different infrastructure. And Jensen Huang literally told you this in the Q4 earnings of Nvidia that the agentic AI inflection point is already here because it's way past that point. Now, think about it. Agentic workflows burns 10 times more of the tokens than chatbots because they have to do all the work and they have to keep in mind what happened the day before, the month before, the year before. And it's non-stop. Also, think about the total addressable market. It's going to be huge. Right now, who uses Claude mainly for heavy lifting? Developers, software, coding right? If agents become a thing, then everybody starts using agents. We're talking about Excel Word PowerPoint Outlook informational stuff. We're talking 50 times more total addressable market and 10 times more usage in tokens. So, we're talking about 500x more AI demand that we have today. And today, it's already pretty explosive, right? We're talking about 24x of today's global capacity just by 2030. It's a whole new level of demand, and I don't think people understand how insane this is. Once this becomes cheap enough, the demand goes up, demand goes up, it becomes even cheaper, demand goes up even more. It's a Jevons paradox. It's going to go into this self-feeding loop, into the flywheel effect, and people will miss out again and again, they'll be too late. And it's not just about the GPUs anymore. GPUs are important, but agents, they need to work 24/7. They need to hold a massive amount of context. So, CPUs are important, memory is important, networking is critical, cybersecurity. We're going to cover all of that today, don't worry, okay? Now, if you think chatbots were power-hungry, and they were, right? Agents are going to be 10x more power-hungry. We're talking about 24/7 base load, that's massive. If you think the current grid isn't good enough for data centers today, think about the future of agents, how underprepared we are, how much more upgrades we need. Okay, we're going to talk about the opportunities in the energy space as well today. A lot of them out there, okay? We'll talk about all of that. But I want us to start with this, okay? And I'm taking time off of my vacation, trust me. I'm in a very beautiful place, let me just show you what I have outside in the pool okay? I'm taking time off of my vacation because this is important. I don't want you guys to miss out, I'm giving everything for free. So, I want you to pay attention, okay? Stay with me here. There's five layers of the AI stack I want us to focus on, okay? There's the compute layer, the cloud and the power layer, orchestration above it, and embodied AI at the top, okay? That's physical AI. Of course, there's the consumption level, but that layer isn't that interesting to me because it's saturated, low margins, it's commoditized, it's very crowded. I'm more interested in layers 1 through 4. Okay, these are the picks and shovels of AI. This is the infrastructure I want to invest in. Okay? So, let's start with compute, layer number one. These are the engines, the memory, the roads. These are also the machines that are building the machines that build semiconductors. And these are the top companies in this layer. Starting, of course, with Nvidia. The mac daddy of this layer, Nvidia is the entire stack's engine. It's a massive ecosystem. They're not just GPU producers, they're also software makers. They have a whole ecosystem impossible to get out of, okay? Every developer, every lab, every data center, everything is built on Nvidia. Moving from Nvidia takes years and billions. It's impossible, Herculean, okay? And also, Nvidia has multiple capex cycles ahead. Every time we're going to complete a capex cycle, new technology will come out because the demands of AI go up, new redeployment of new GPUs, and new equipment. It's not telecom. Compute gets old very fast. So, Nvidia has at least five or six capex cycles ahead until it even starts to slow down. Look at the numbers it has today. 220, almost 220 billion of revenue growing at 65% per year, a company of this magnitude. 75% gross margin. And operating income growing at 60% per year. This is insane. Of course, to complete Nvidia, you have to get Broadcom. Broadcom is the perfect completion to Nvidia. These are custom AI chips, and this is the network that connects the GPUs. So, Broadcom actually makes custom AI chips for data centers. So, Nvidia basically gives you the computing engine, and Broadcom helps customize and connect all the infrastructure around it. It's just as critical as Nvidia, and guess what? It's just as gold standard as Nvidia. Look at their fundamentals. It's going to blow your mind. AI semiconductor revenue year over year is up 143%. Operating income up 89% year over year. Net income is up 292% year over year. This This is the company you want to own going into the super cycle. And of course, there's AMD. Now, AMD it ha- it has basically two things I want you to remember. Obviously, it's the second sourcing of GPUs. And even if you have the third the fourth and the sourcing of GPUs, they would all sell out. There's plenty of work in the GPU sector. But also, they're massive CPU producer. And now CPU became a thing again because of a genetic AI, because of context, because of memory. That means that they have a double thesis both in GPUs and CPUs, okay? Massive. Look at the numbers. Net income is up 164% year over year. Operating income is up 94% year over year. Free cash flow is up 180% year over year. Wonderful numbers. Out of the park. Of course, there's ARM, the company that Nvidia tried to buy but was denied because it would create too much of a monopoly. They literally own the blueprint. Every single time somebody builds a chip, somebody makes a semiconductor, ARM gets paid. They have the ultimate toll booth model on the AI industry, okay? The genetic AI basically adds more and more royalty income to ARM. Wonderful business to own, okay? ASML the literal monopoly of lithography machines. These are the machines that build the machines that build semiconductors. There's only one company that makes them. They're out there in the Netherlands. Probably all the employees bike to work. I don't know, I'm just guessing, right? Nobody wants to talk about them because they're boring. Dutch company. There is no alternative. They're booked They will be booked for the next 5 years full capacity. Wonderful business to own. Of course, there is Micron that used to be a commodity and used to be an under the radar business, completely ignored, but now, because they have capacity, because they they have the scale, because high bandwidth memory has become a thing, all of a sudden, the commodity part of it goes out the window because their moat now is the fact that they can make a lot of memory. To produce memory out of thin air right now is impossible. If you want to build a facility to compete with Micron, it's going to take you 5 years and billions of dollars, okay? So, we're talking about massive demand ahead. High bandwidth memory is going to be huge. The gigantic shift is going to take Micron even higher. But guess what? Even after the huge spike in price they've seen, they're still trading at five times forward earnings. Their operating income is up 650% for the past year, and their free cash flow is up 1,300% and this is just the beginning for Micron. Now, Arista Networks is a company I actually added to my top stocks list a few years ago. It's up like 340% since we added it. It basically connects the different GPUs inside the data centers. They're the switching gear and all the networking gear, which becomes even more important with agents, obviously. Solid company, it's going to grow really well in the gigantic AI. Revenue is growing at 30%, income is growing at 23%. Wonderful company. Now, Cadence is an EDA software, basically electronic design automation. What they do is they simulate the whole process. Let's say you want to create a new semiconductor, a new chip, and you want to run it through the whole process and simulate the whole production line. Find mistakes before you commit billions of dollars into the production line, to the tooling, and you waste time. It's a critical tool to have, and agents will require more chips, massive demand ahead. 42% free cash flow growth, 22% operating income growth, 31% operating margin, wonderful business. Now, Lam Research, it's the global chip etching equipment producer. What do they do? Let me explain. And they are a gold standard, They own 50% of the market. They're the premier one. So, ASML basically prints the chip pattern, then Lam builds a machine that carves these patterns into the silicon. So, they completely um uh integrate with the ASML equipment. They're just as important, right? They're going to be a massive massive winner of the agentic AI phase, okay? 40% net income growth year over year, 24% revenue growth, and 40% operating income growth year over year. TSMC, a monopoly, yes. They're in Taiwan, a disputed potentially geopolitically problematic region, but they are a monopoly. They are a monopoly. They manufacture nearly every chip on Earth. You can't go into the agentic AI without having some TSMC. They're sold out into infinity. Operating income growth is at 48% and they're trading below 20 forward PE. Now, let's go to the second layer, power and cooling. Now, this is super important. Obviously, you have to talk about Amazon here. AWS is the toll booth booth of AI. They're the biggest cloud producer and they're absolutely underpriced right now. Everybody ignores them. Everybody thinks they lost it. But, the net income growth of the past year is 31% and they're currently trading at 3.6 price to sales and 25 forward PE. They own the largest toll booth on the highway of AI. You can't go wrong with Amazon. Same thing with Microsoft. Might be even worse as far as pricing and misunderstanding. They own the cloud, the business software, the data, and all the tools required to run these AI agents inside the organizations, okay? 20 time earnings, 47% operating margin, and you can buy them right now at nine times sales. And the stock is actually down recently. Google with the TPUs have completely broken the mold and have created vertical integration, at least in some part, with the TPUs. So, basically, by not needing only GPUs, they have improved their margins insanely. That's why they're the fastest cloud growing right now. 63% annual growth in cloud, definitely the fastest. Microsoft is growing at 40, Amazon is 28, but they're absolutely insane. 32% net income growth year-over-year, and you can only pay 23 time earnings right now, forward earnings, to buy them, which is a good price. Vertiv, a company I've added I think a year ago, we've done very well with Vertiv. Obviously, cooling is super important. The more density you're going to have with the genetic AI, the more thermal management you're going to need. Obviously, Vertiv is going to do very well. Look at these numbers. When I added this a year ago, I knew it's going to happen. 170% net income growth, 67% free free cash flow growth, and 38% operating income growth for the past year. Absolutely insane. Now, remember when I told you that the grid is going to need upgrades? You said, "Well, well, GE Vernova, GEV, they build the turbines and the grid hardware behind all of this, okay?" They cannot expand data centers without the major major grid upgrades, and GE Vernova is the go-to, okay? They build all these stuff that are needed to upgrade the grid, but we're talking about 76% operating income growth. 214% net income growth and 62% operating profit margin growth just for the past year, because this is already happening, folks. Now, Ciena is a company I've never talked about, but they're the optics between the data centers, okay? Agents, they actually multiply traffic, but not just inside the facility itself, but between data centers. Ciena makes the high-speed optical network that basically connects all of this. Look at the numbers. 47% income growth. 63% operating income growth. 76% free cash flow growth. All that is just in the past year. Bloom Energy I talked about before many times. There's a lot of hoopla around it, short reports, etc., etc. Look at the numbers. Onsite power is critical. When the grid is like that, think about how important it is onsite power. Almost at 40% revenue growth year-over-year, 217% operating income growth year-over-year, 72% free cash flow growth year-over-year. Phenomenal numbers. Now, of course, they might be just lying to everybody. In that case, could be, maybe. But, if they're not lying, oh my god, these numbers are monopoly numbers. You have to get in on Bloom Energy. Good. CEG, Constellation Energy, a little bit slow growth, but it's nuclear. And nuclear is so important, okay? A gigantic AI shift needs 27, sorry, 24/7 reliable power. They need sustainable, cheap, reliable power non-stop. That's nuclear, okay? It's not speculative, and it's cheap. Three times sales today, 125% free cash flow growth year over year. Wonderful company. Digital Realty. They are the landlord. Never talked about them before. They own all this real estate that's being leased out for these data centers, okay? As demand for power data centers actually goes up, boom, DLR makes more money. And guess what? Look at the numbers. Net income growth 117%. Free cash flow growth 520%. EBITDA margin 61%. These guys are basically selling the land everybody's going to farm. Can't go wrong with that, okay? Now, let's talk about layer number three, orchestration, okay? This is the bridge when you connect AI and business value, and money, okay? Agents are going to need software that governs, watches, monitors, and secures these agents, okay? I'm going to show you a few elite companies here. Obviously, Palantir has to go here into the thesis. I mean, I'm not going to go again about Palantir. It's my top conviction, obviously. Look at the numbers, okay? Everything you got to say about Palantir being eaten up by chatbots, which is ridiculous. 85% revenue growth. 145 rule of 40, which is four times highest than Daily number. And 360% operating income growth year over year. Enough said, okay? Except Palantir, we also have MongoDB. They are the database. If you want good output from your models, you need good database, okay? MongoDB is the gold standard, okay? 314% free cash flow growth the last year, 23% revenue growth. Context is super critical for this cycle and MongoDB is on that bottleneck. CrowdStrike is just as important because with these agents working independently, cybersecurity becomes super important. Think about all these agents, they have access to your information, to your codes, to all all the stuff that can really hurt you, right? Look at what's going on with CrowdStrike now. 120% net dollar retention, 22% revenue growth, 23% free cash flow growth. Wonderful company for cybersecurity. AI net, obviously a company I talked about multiple times. They're in layer one and in layer three where we talked about them in today's presentation. And now we're getting to embodied AI. Now, embodied AI, think about it. I think that the 2030 to 2035 is going to be the robotics decade, okay? And I think one company is positioned to dominate robotics by 2035. Think about it. Embodied AI robotics is going to become the largest layer of the AI economy, okay? Physical labor is way bigger than the economic pool than software, okay? Right now, as it stands, we can replace 13% of current US working hours using technology we already have today. Okay? As these robotics, they get better and better and better, every single warehouse factory delivery uh supervision, every repetitive household work, everything becomes robotics. Okay? We're talking about 3 trillion of economic value by 2030 just in the US. And that's why Tesla already is leading this massively because they're already shipping robotics at scale right now. Optimus isn't their first time of creating a robotics product, okay? FSD is a robot. It's a robot that sees the world through cameras using neural network and making decisions on the fly. The only difference, it has wheels instead of arms and legs. I mean, that's it. FSD is humanoid robotics on wheels. Think about how easy it's going to be for Tesla to take the lead here. The expertise in electric motors, check. Expertise in vision and sensors, check. Expertise in batteries, check. Expertise in supply chain, check. They're the world-class leaders in manufacturing process, check. Data, check. 10 million real miles every single time. There's nobody that can compete with them, okay? Think about when Elon Musk was debating why lidars is a bad idea. It was easier to put lidars on these cars, but why did he did he insist on cameras only? Because he was building the the robotics division already at the time. You can't manufacture human robotics, you know, with a lidar on his head. >> [laughter] >> It's not going to work, okay? That's why he wanted cameras. Eight cameras, it's cheap, okay? Think about it. Think about all the miles he's creating every single day in the physical world with FSD, okay? Tesla also actually has good fundamentals even though they're pivoting. By 2035, I don't know if Tesla even makes cars, but still the numbers are super solid even as a car company. 74% free cash flow growth with past year, 44 billion in cash, plenty of money to do this, and 95 billion of annual revenue just from selling cars. So, they have the money, they have the capital, they have the incentive of Elon Musk, they will win the robotics race, okay? I want you to focus on infrastructure over applications, folks. Focus on modes over hype, orchestration over shiny wrappers. That's how you win. So, how do you actually do this, okay? Make this note, very important, okay? The most important thing you got to do before you invest in any of these companies, make sure you have an emergency fund for 6 months. If you don't have it, the moment happens and when the feet hit the fan, you're going to be selling stocks at 20 cents on the dollar because you're going to have to pay rent. An emergency fund comes first before investing. Then, never break these three rules. The S&P is critical, you got to have at least half of your portfolio. Bonds are important, DCA is important, DCA X2 investing system is important, and trimming is just as important. So, the BCAX2 system I teach is very, very simple. You deploy 50%, 50% you wait. Every time the stock comes down, you deploy double. 150%, then when the stock goes below or above the threshold, you adjust accordingly. So, you buy at all times, but the buy rate changes according to the price of the stock, okay? And then, you also have the trim rules. Once the stock goes up 50%, 100%, 150%, 200%, you trim according to the system and use the same proceeds to continue to DCA into the stock, okay? It's super important to follow these guidelines and this is exactly why I always say, never sell because the stock went up, never sell because the stock went down, never sell because of macro. Focus on the fundamentals and that's exactly what we teach on Patreon, on the Roth Academy. patreon.com/don-nash you can get in, you can get my top stocks list, but I don't want you to join because of the top stocks list. Join because of what you will learn. Join because you will become a better long-term investor for the next 10 years. Join because of the community, because of the lectures, not because you'll get some stock picks. If you want to be a part of this, we have 25,000 members. If you want to get in on this, I invite you to join the academy, become a better investor and learn how to do this yourself, how to be a thinking person, not how to buy stock picks. Thank you so much and I'll see you in the next one. Peace out.