Top 10 Stocks I'm Buying to Get Rich in 2026 (Without Getting Lucky)

It's almost the new year, which means your feed is probably flooded with videos claiming to have the top stocks for 2026. And if you've watched even a few of them, you already know the pattern. A random list of companies, a couple of cherry pick charts, maybe a bold price target slapped on top, and that's about it. This video is different, not because I magically know the future, but because every stock on my list actually works together. I'm not betting on one company getting lucky. I'm owning the winners across five massive, fast-growing AI markets at the same time. And that's how you build real wealth without relying on luck. Your time is valuable, so let's get straight into it. Before we talk about 2026, we need to talk about accountability. Anyone can sound smart talking about the future. What actually matters is results. This is my third year putting out an annual stock list. So instead of asking you to trust me blindly, I want to show you exactly how my past lists performed and let you decide whether the rest of this video is worth your time. Here's the list I published for 2024, along with each stock's performance over the year and how it compared to the S&P 500. And 2024 was not an easy benchmark to beat. The S&P 500 returned over 25%, which is one of its strongest years in recent history. But the stocks on my list returned 77% on average. That means we more than tripled the market during a year when the market itself was already on fire. Now here's my 2025 list. Once again, a strong year for investors overall, with the S&P 500 returning about 18% year to date. But the stocks on my list returned 36% on average. Put those two years together and we're up roughly 140% in just two years without leverage, without day trading and without chasing hype. That's what I mean when I say getting rich without getting lucky. And I'm not here to drag this out. Here is the full list for 2026. I've included timestamps so you can jump around if you want. But before you do that, there are a few important things you need to understand so you don't completely miss the point of this portfolio. First, this is not an all or nothing list. This is the foundation of my portfolio for 2026 and beyond. And I'm sharing it for educational purposes only. You don't need to buy every stock on this list. Buy what you understand, buy what you believe in and ignore the rest. Second, I'm not going to rehash the same surface level talking points you hear everywhere else. Instead, I'm going to show you how these companies fit together to cover five critical AI markets. Models, chips, data center infrastructure, AI native applications and enterprise software. Owning the leaders across all five is how you win without having to guess which single company dominates everything. And finally, full transparency. I'm not a financial advisor. My background is in electrical engineering and data science and I've spent over a decade using AI in real world systems. I invest for the long term. I stay inside my circle of competence. And that's why every stock I cover today is directly tied to AI and the hardware and software that power it. This video will stay high level because there's a lot to cover. But my channel goes deep into the science behind these businesses. If a company catches your interest, chances are I've already done a full breakdown on it. So with that context, let's build the foundation. Every year, I say the same thing because it's still true. Beating the market is simple. It's not easy, but it is simple. If all you did was buy and hold the NASDAQ 100, ticker symbol QQQ, you would outperform the S&P 500 over the long run. The NASDAQ is heavily weighted toward technology companies and technology compounds faster than almost anything else in the economy. Right now, it's finishing the year up about 22% compared to the S&P 500's 18%. That may not sound dramatic, but over time, that gap compounds into life changing money. The problem is that the NASDAQ also holds companies that don't really fit the AI focused future I want exposure to. You're still owning things like Costco, Pepsi and T-Mobile. Great businesses, but they're not where the AI upside is concentrated. So in 2025, I found a fund that lets me have my cake and eat it too. Vanguard's Information Technology ETF, ticker symbol VGT. VGT holds 322 companies, which puts it right in between the NASDAQ 100 and the S&P 500 in terms of diversification. But unlike both of those, it only includes companies focused on semiconductors, hardware, system software, and applications. In other words, it's almost perfectly aligned with the AI revolution. No consumer staples, no telecoms, no dead weight. Even better, the top 10 holdings make up about 60% of the fund. That includes Nvidia, Microsoft, Broadcom, Palantir, AMD, and Micron. Companies that sit right at the center of the AI value chain. On top of that, VGT has lower fees than SPY or QQQ, which means you're not just outperforming the market, you're paying less to do it. And that accelerates compounding even more. If your brokerage doesn't offer VGT, QQQ is still an excellent alternative. Either way, this kind of fund forms the backbone of a long-term portfolio. But the real outperformance doesn't come from index funds alone. It comes from owning companies that dominate specific layers of the AI stack. And that's where AI itself comes in. The global artificial intelligence market is expected to grow nearly 19-fold over the next nine years, representing a compound annual growth rate of about 38.5% through 2034. That's roughly three times faster than the historical average return of the S&P 500. You don't need perfect timing when you're riding growth curves like that. At the very top of the AI stack are models. Models make the rules. They define the APIs, the features, the pricing, and the capabilities that everyone else has to build around. If you control the models, you influence the entire ecosystem. That's why Nvidia, Google, and Meta platforms are at the top of my list. Nvidia now has Nemo and Neutron, a full family of reasoning and agentic AI models released with open weights and integrated data tools. Meta has Llama, which has become one of the most widely adopted open-source large-language models on the planet, not just inside MetaZone apps but across enterprises building co-pilots, internal tools, and customer-facing products. And Google has Gemini, which is deeply integrated into Search, Workspace, Gmail, YouTube, and Google Cloud, giving it unmatched access to usage data, feedback loops, and monetization pathways. And speaking of data, you already know how important protecting it is. VPNs encrypt your traffic and protect your privacy, but they can also save you serious money by letting you shop from different digital locations. That's why this video is sponsored by Surfshark. Surfshark is fast, easy to use, and works on unlimited devices with a single account so you can protect yourself and your entire family. They also offer Surfshark Alert, which notifies you if your email, credit cards, or personal information are leaked in a data breach. And because many online stores adjust prices based on location, Surfshark can help you find better deals on flights, hotels, and rentals. If you want to try it, you can go to surfshark.com ticker to get up to four extra months free, backed by a 30-day money-back guarantee. Now, while many frontier model companies like OpenAI, XAI, and Anthropic are still private, owning NVIDIA, Google, and Meta gives you exposure to some of the most powerful AI models that are already deployed at global scale and generating real revenue today. They also have something almost no startup can replicate, distribution, billions of users, global cloud platforms, and deep integration into everyday workflows. And here's where things get really interesting. The companies that build the models also influence the future of AI chips because chips are designed and optimized around the most widely used architectures, which means owning the model makers also positions you perfectly for the next layer of the stack, the hardware. That's why the next group of stocks focuses on the picks and shovels of the AI gold rush, the companies designing and manufacturing the chips that make all of this possible. And that's where we're going next. Because once you understand that AI models shape the rules of the ecosystem, it becomes obvious that the real leverage sits underneath them. Models don't run on magic. They run on silicon, memory, networking, power, and manufacturing capacity. And every time a new breakthrough model is trained or deployed at scale, demand explodes across the entire hardware stack. That's why the next layer of my portfolio focuses on the companies designing and building the chips that power the AI era. Owning NVIDIA, Google, Broadcom, Taiwan Semiconductor, and Micron means owning a massive portion of the global AI hardware supply chain, from accelerators and custom silicon to networking, memory, and fabrication. This is where the get-rich-without-getting-lucky idea really starts to click, because these companies make money no matter which AI application or model ends up winning. Let's start with NVIDIA, because it's impossible not to. NVIDIA controls roughly 92 percent of the data center GPU market, and that dominance didn't happen by accident. Their hopper architecture reshaped AI training, and Blackwell is doing the same thing for both training and inference. These chips are the default choice for almost every major AI workload on Earth. When companies talk about building AI factories, they're usually talking about racks full of NVIDIA hardware. But NVIDIA is not just a GPU company anymore. They're building an entire AI computing platform, from CUDA and software libraries to networking, interconnects, and system-level designs. That's why their acquisition strategy matters so much. NVIDIA recently announced a roughly $20 billion acquisition of Grok's assets, marking the largest deal in the company's history. Grok was founded by the original architects of Google's TPUs, and focused on ultra-low-latency inference for large-scale workloads like recommendation systems, search ranking, and real-time AI agents. This is a defensive and offensive move at the same time. NVIDIA knows that inference is the next major battlefield. Training gets headlines, but inference is where AI becomes embedded into every product, every workflow, and every service. By absorbing Grok's talent and technology, NVIDIA is shoring up its position exactly where future demand is headed. They've done this before. Their $7 billion acquisition of Mellanox in 2019 added InfiniBand, Bluefield DPUs, and high-performance networking, which turned out to be one of the most important moves in NVIDIA's history. Grok fits that same pattern. At the same time, NVIDIA doesn't operate in a vacuum. Google is one of the few companies on Earth that can challenge NVIDIA in specific AI workloads through its Tensor Processing Units, or TPUs. TPUs are custom accelerators optimized for Google's internal AI models, including Gemini, as well as external workloads through Google Cloud and Vertex AI. What makes this particularly interesting is that Google has openly discussed selling TPUs to external customers, including companies like Meta. That's a huge signal. It tells you that Google isn't just using AI hardware defensively. They're actively trying to take market share in accelerators. And from an investor's perspective, that's perfect. Owning both NVIDIA and Google means you benefit regardless of whether GPUs or TPUs win more inference workloads over time. You're not guessing. You're collecting. Then there's Broadcom, which quietly sits at the center of nearly everything. Broadcom is the undisputed leader in custom AI chips, designing Silicon for Google, Meta, OpenAI, and Anthropic. These custom accelerators are tailored to specific workloads and deployed at massive scale. Broadcom also dominates networking. Over 99% of global internet traffic touches a Broadcom chip at some point. That matters because modern AI data centers are increasingly Ethernet-based, and Broadcom owns that ecosystem. Networking is one of the biggest bottlenecks in AI infrastructure. You can have the fastest GPUs in the world, but if data can't move efficiently between them, performance collapses. Broadcom's deep entrenchment gives them an almost unfair advantage. New entrants aren't just competing on performance, they're competing against decades of integration, relationships, and deployment. And none of these chips matter without manufacturing. That's where Taiwan Semiconductor Manufacturing Company comes in. TSMC is the backbone of the entire semiconductor industry. They control over 70% of the global foundry market and roughly 90% of advanced node manufacturing. If you're using a cutting-edge chip, there's a very high probability it was made by TSMC. NVIDIA, AMD, Apple, Google, Broadcom, Microsoft, Amazon, they all rely on TSMC. The barriers to entry here are almost incomprehensible. Tens of billions of dollars per fab, years of process development, and an ecosystem that's nearly impossible to replicate. If something catastrophic ever happened to TSMC, global technology would be the least of our concerns. From an investment standpoint, that level of strategic importance creates a moat that only gets wider over time. Then there's memory, which doesn't get as much hype, but is absolutely critical. AI models are memory hungry. As parameters scale, so does the need for fast, high-bandwidth memory. Micron is the only U.S.-based company with a position in high-bandwidth memory, or HBM. Their HBM chips are used in NVIDIA's Hopper H200 and Blackwell B200 GPUs, as well as AMD's Instinct accelerators. Memory has historically been cyclical, but AI is changing that dynamic. This isn't consumer PC demand that comes and goes. This is structural demand driven by model size, inference workloads, and data center buildouts. As long as AI models keep getting bigger and more widely deployed, Micron's relevance only increases. When you put all of this together, you start to see the pattern. NVIDIA, Google, Broadcom, TSMC, and Micron all have deep moats built on capital intensity, intellectual property, and supply chain dominance. They don't need to pick the winning app. They profit every time someone trains a model, deploys an agent, or scales an AI service. That's the kind of exposure you want if your goal is compounding, not gambling. But chips alone aren't enough. AI hardware needs an enormous amount of supporting infrastructure. Power, cooling, racks, networking, and physical space all become constraints at scale. That's why the next layer of my portfolio focuses on data center infrastructure. Once again, NVIDIA and Google show up here, but for different reasons. NVIDIA's reference architectures and AI factory designs are shaping how modern data centers are built. Google, through its hyperscale infrastructure, sets the bar for efficiency, power usage, and deployment. But there are two other companies that deserve special attention. The first is Vertiv Holdings, ticker symbol VRT. Vertiv builds the power and cooling systems that make high-density AI data centers possible. Traditional air cooling simply doesn't work for modern GPU clusters. Vertiv's liquid cooling solutions are designed specifically for AI workloads, and their flagship systems can scale up to 600 kilowatts per unit. That's enough to liquid cool multiple NVIDIA Blackwell racks simultaneously. What makes Vertiv particularly compelling is how integrated their systems are becoming. These aren't bolt-on solutions anymore. They're being designed directly into AI-optimized data center layouts, including NVIDIA's own AI factory-style deployments. As power density increases, cooling becomes mission-critical, not optional. That gives Vertiv pricing power and long-term relevance. The second company is Iron, which takes a very different approach. Instead of trying to compete with hyperscalers as a full cloud platform, Iron focuses on building AI-first data centers and renting out bare-metal GPUs at scale. They partner with companies like NVIDIA and Microsoft instead of fighting them, which is a much smarter strategy. Iron controls roughly 3 gigawatts of secured grid-connected power across regions like Texas and Canada, with plans for massive campuses capable of supporting up to 2 gigawatts of liquid-cooled Blackwell-class deployments over time. That translates to hundreds of thousands of GPUs' worth of capacity. Cheap renewable power, owner-operated facilities, and long-term partnerships make Iron one of the cleanest ways to gain exposure to the physical backbone of AI expansion. At this point, you have the physical side of the AI revolution covered. Models at the top, chips in the middle, and infrastructure underneath. But AI doesn't create value in isolation. It creates value when it's embedded into real-world workflows, decisions, and systems. That's where the digital infrastructure layer comes in. This is where companies like Palantir and CrowdStrike start to matter even more, because they sit at the intersection of AI, data, and enterprise operations. They're not just adding AI features. They're rebuilding how organizations operate in an AI-driven world. Palantir is becoming the operating system for decision-making across enterprises and governments. Through platforms like Gotham, Foundry, and AIP, Palantir integrates fragmented data, builds a shared ontology, and allows AI models to interact directly with real-world processes. Manufacturing, logistics, finance, defense, these aren't toy use cases. These are mission-critical systems, where accuracy, traceability, and trust matter. As AI agents become more autonomous, the need for a secure, auditable layer between models and actions becomes non-negotiable. That's exactly where Palantir thrives. If companies want AI co-pilots that don't just talk but actually act, this is the infrastructure they need. At the same time, every one of these AI-driven systems expands the attack surface. More automation, more agents, more endpoints, which means security becomes even more important, not less. And that naturally brings us to CrowdStrike, because AI-driven enterprises need AI-driven security to survive. And that's where the final layer of this portfolio really starts to lock into place. CrowdStrike sits at a very different but equally critical choke point in the AI era. As companies deploy more models, more agents, and more automated workflows, the number of machines making decisions explodes. Every endpoint, identity, workload, and data store becomes both more valuable and more vulnerable. CrowdStrike's Falcon platform ingests and correlates trillions of security events every single day across endpoints, cloud environments, identities, and networks. That scale matters because modern cyber threats move faster than humans can react. What makes CrowdStrike especially powerful in an AI-driven world is that security itself becomes a data problem. Falcon's massive telemetry advantage feeds directly into its machine learning and generative AI layers like Charlotte AI and Falcon AI detection and response. These tools don't just flag alerts, they automate investigations, contextualize threats, and stop attacks in real time. As generative AI lowers the barrier for attackers, it simultaneously raises the value of platforms that can defend at machine speed. That's why CrowdStrike benefits structurally from AI adoption instead of being disrupted by it. Now step back and look at the competitive landscape. The biggest competitors to Palantir and CrowdStrike aren't small startups. They're Microsoft, Amazon, and Google. And that's exactly what you want as an investor. Google is already on our list. Microsoft makes up more than 12% of VGT. Amazon was intentionally swapped out of the individual stock list this year, but it still has indirect exposure through cloud demand and chip manufacturing partners. In other words, no matter which enterprise software stack becomes dominant, this portfolio captures the upside. You're not betting on one company's execution, you're owning the ecosystem. That's the through line of this entire strategy. Every stock on this list reinforces the others. AI models drive demand for chips. Chips drive demand for manufacturing, memory, and networking. Those chips need power, cooling, and physical infrastructure. And once AI is deployed into the real world, companies need software platforms to operationalize it and security platforms to protect it. This isn't a collection of disconnected picks, it's a system. And systems thinking is exactly how you avoid luck. Luck is required when you need one company to be perfect. Systems compound when multiple winners grow together. That's why I care so much about adaptability, which is the final column on my list and one of the most overlooked factors in investing. Adaptability isn't fluff. It's the difference between companies that survive transitions and companies that get disrupted. Look at what's already happened. Google saw inference becoming a bottleneck and decided to commercialize TPUs instead of relying solely on NVIDIA. NVIDIA saw specialized inference chips emerging and responded by acquiring Groke's assets and leadership. Meta realized closed models could trap developers, so they spent aggressively to make LLAMA open and ubiquitous. Micron walked away from low margin consumer memory and poured capital into high bandwidth memory for data centers. Every company on this list has demonstrated a willingness to pivot, invest, and defend their position aggressively. They don't wait for disruption, they move toward it. Markets reward that behavior over long time horizons. And this brings us back to why I structured the portfolio the way I did for 2026. The foundation is broad AI focused exposure through VGT or the NASDAQ 100. That gives you diversified upside across hundreds of technology companies that are already benefiting from AI. On top of that foundation sit targeted positions in companies that dominate specific high growth layers of the AI stack. This year I made three deliberate changes. I swapped Amazon for Micron because I wanted more direct exposure to AI hardware demand rather than consumer and logistics heavy businesses. I replaced Arm with Taiwan's semiconductor because manufacturing dominance matters more than licensing models when advanced node capacity is the constraint. And I swapped Fortinet for Iron to gain exposure to the physical infrastructure and power layer that increasingly defines AI scalability. Those changes weren't about chasing returns. They were about tightening the portfolio around the most structurally advantaged parts of the AI economy. And it's worth emphasizing something important here. None of this requires perfect timing. You don't need to predict when interest rates change, when the next model release happens, or which app goes viral. These companies are positioned to benefit from AI adoption over the next decade, not the next quarter. That's how you let compounding do the work for you. Now I know this video has been long, but that was intentional. Building a real foundation for 2026 means understanding why these companies belong together, not just memorizing tickers. To me, investing isn't about finding the next hot trade. It's about aligning capital with the fastest growing, most durable trends in the economy and letting time do the heavy lifting. So once more, let's ground this in reality. Here's the list I published for 2024. It returned 77% on average during a year when the S&P 500 was already up more than 25%. Here's the list for 2025. It returned 36% on average, roughly double the market again, bringing total returns to about 140% over two years. Those results didn't come from leverage or speculation. They came from understanding where growth was going and owning the company's best position to capture it. And this is the list for 2026. It covers AI models, chips, manufacturing, memory, data centers, enterprise platforms, and security. If one company stumbles, others benefit. If one layer compresses margins, another expands. That's what a resilient portfolio looks like. Generative AI isn't slowing down. Every breakthrough creates new demand elsewhere in the stack. More capable models require more compute. More compute requires more power and cooling. More automation increases the need for software orchestration and security. This feedback loop is still in its early stages, and the companies on this list are the ones building the rails it runs on. If you've made it this far, I appreciate it. And if you feel like I've earned it, hitting the like button and subscribing really does help the channel grow. I focus on the science behind these stocks because understanding how things work is the fastest way to build conviction and avoid emotional decisions. I'll continue breaking down these companies throughout the year. As new data comes out, technologies evolve, and competitive dynamics shift. Investing isn't static, and neither is AI. The goal isn't to predict the future perfectly. It's to stay aligned with it. So there you have it. My top stocks to get rich without getting lucky in 2026. Not because I know something magical, but because I know how these systems fit together. And until next time, this is tickerall you. My name is Alex, reminding you that the best investment you can make is in yourself.