The AI Crash Created a Buying Opportunity Almost Nobody Sees

A couple weeks ago, Wall Street got exactly what it want. Micron exploded, gave us incredible earnings. ASML beat expectations. TSM once again proved that the AI demand is exploding. And these are three incredible companies. And the market, well, it sold off anyway. And that should make absolutely no sense at all. Unless Wall Street really isn't selling AI, it's rotating inside of AI. And if that's true, this pullback may be creating one of the biggest buying opportunities all year. Because while people are truly focused on Nvidia and AMD and Broadcom, and AMD has truly been an outperformer, Nvidia and Broadcom, not so much, but they still get a lot of the focus along with the memory stocks. There's also three companies quietly building the AI revolution that almost nobody's ever talking about. And one of the companies is teaching chips how to communicate using light instead of electricity. Another company makes sure every AI chips actually works before it leaves the factory. And one company literally teaches artificial intelligence before it can even think. So by the end of this video, what I really want to do is you'll understand exactly where these companies fit inside the AI ecosystem. Some bigger than technology. And while I believe the market pullback has given us long-term investors a great opportunity before we can break them companies down, we have to answer one question. Why did great businesses report great earnings and the market still fell off? Because if you can understand that and you can understand why these companies fit in, you'll understand how do we take full advantage of this as we move forward. So before we get into any of that, man, make sure you like, make sure you subscribe, and please share this video out. help us get at least 2,000 likes on this and that helps the channel grow and helps us get in the algorithm. Let's go. All right, let's go back about a month. Technology stocks were making new highs almost every other week, especially memory. Nvidia was kind of getting up. It was moving, but again, the memory stocks, they had the crown. They couldn't do no wrong. And everybody wanted to be in the eye. And then almost overnight with great earnings, everything changed and the headlines kind of shifted. Iran interest rates bombing, profit taken, June swoon, SpaceX IPO, and the list goes on. Anthropic, people getting sued, and just like that, before we saw it, many of the leaders, well, the ones that got us here, they started falling. And not just small falls. We're talking 20%, 30%, 40%. And yet these companies are still up triple digits. Now, if you're a new investor, that can make you question everything because your brain immediately. Did I miss something? Is AI over? Should I sell? The market is not for me. But here's something I've learned through my years of just studying the market, being in the market, and truly just navigating this space. The market and the businesses are not the same thing. The markets tell you how people feel today. The businesses tell you what the company is becoming over the next three, five, 10 years. Those are two completely different conversations. And sometimes stocks will fall while great businesses get stronger. And when that happens, I pay attention. Now, let me say something right quick, man. I took off two weeks. I came from Italy, man. And I just needed to recalibrate because I knew this half of the year. I wanted to give y'all a lot of information. So, strap up. So, here's the big thing the body talking about. Wall Street is not really thinking just getting rid of AI. I think the question that we need to ask is well companies like Microsoft, like Meta, like Amazon, like Google, like Open AI, are these companies making enough money? And are they making enough money to justify the hundreds of billions of dollars they're spending on AI infrastructure? Because that's the answer they're going to have to answer this week as we get into tech earnings, starting with Google. It's not AI, the return of AI. Nope. They're thinking about building brand new airports before the planes ever land. Um, have you ever had to buy land before? Well, they're thinking about building a runway, building a terminal, hiring workers, installing security, laying thousands of miles of electrical cable, building baggage system. None of that makes money in one day. But it requires before the airport can become profitable. And that's what's happening right now. The hyperskalers are building the airport and Wall Street is asking when do the profits arrive because Wall Street is all about money. These companies about building to the future and that's why investors have gotten really really nervous because they're waiting to see when of all the spending going to give us some returns. Now, I know you want to get into these companies right quick, but I got to just show you this because there's a reason why I don't want you to panic. These three companies, I promise you, you've never heard of and they're going to blow your mind. But but we can't miss the idea where we at. We also have an idea that Trump and Iran, we've been bombing each other for nine days. America has bombed Iran for nine days, right? And we got a report out today that said that Iran is saying, "Hey, we want to come to the table." We got a Fed meeting coming up next week, but we also saying this week no Fed can talk. The ideas about do we get a rate hike or not? All of these things come and they're important for what we going at. So, I can't just tell you about these stocks. I need you to understand what's in front of us. Because at investors, all of these things make sense. And I don't want you to think that, you know, it's just about the stocks. Well, we got to understand the landscape. Understanding the landscapes helps us. We also got a midterm election that comes up in late October, early November. This becomes the roughest time to be in the market. Markets are really choppy. I tell my traders, listen, wait for good setups because even the great setups can fail. So, we got to wait to see if we have any followthrough. For my buy and hold investors, you got to wait like just be cautious. Don't be candle have candlestick whiplash, right? Wait to see exactly where this price is going. Now, I'll be honest with you. One of the key things that I like to say is if a great company has great earnings, that has great forward guidance, hits a 200 day moving average, that's the buy no matter what the price is. Simply because the 200 day moving average tells us that the stock has fell 20% from the all-time high on the daily. That 100% is a great buy. Yes, it can go cheaper, but now we just thought that change our cost basis because it is essential for us to build positions in quality companies over the long term. Now, one more thing I want to talk about before we get into it. Years ago, think about the internet. Most information travel through co wires. Today, most of that information travels through fiber optic cables. Why? Because light moved dramatically faster. more information with less heat, with less resistance, but with greater efficiency. Now, imagine taking that same exact idea and shrinking it down. Yep. Until it fits inside a semiconductor chip. That's exactly what one of the companies we're going to talk about do. The company by the name of Tower Semiconductor is helping make that possible. And once you understand that, you'll understand why companies like Nvidia, Marll, and some of the biggest names in AI are paying so much attention to what's called Silicon Photonics. Now, this is what I wanted to get into when I talk about sil silicone photonix. The company is TM. Most investors never heard of Tower. Matter of fact, if you ask 10 people the name, they'll think I said TSM. But they heard of Nvidia, they heard of AMD, they've heard of Broadcom, they've heard of uh Microsoft, they've heard of Google. And again, they've heard of TSMC, but not Tower Semiconductor because if you ever mention Tower Semiconductor, they'll probably think that's a mistake because Tower isn't trying to be the smartest AI chip company. They're solving one of AI's biggest problems, communication. That's right. Communication. AI has more than one bottleneck. And for years, everyone thought the biggest challenge was making chips faster. And for a while, it may have been it. But Nvidia solved a big part of that problem. Every single and every generation of GPU gets faster, more powerful, more efficient. But we've never entered a different phase until now. That problem isn't just how fast one chip thinks. It's how fast thousands of chips communicate together. So imagine this. You're sitting in a classroom with 100 students. Everybody finishes the test at the same exact time, but there are only one printer in the room and now everybody has to wait. The students aren't the problem. The printer is what's called an engineered bottleneck. And AI has the same problem. Nvidia can build the fastest GPUs in the world, but if those GPUs can't exchange information fast enough, then the entire system slows down. the bottlenecks moves from computing to communication. And so how do you fix that? Well, for decades and just just hear me as we go through why TS um this company Tower semiconductor is important because chips have gotten communication through electricity. Electricity served us well. But it comes with trade-offs. As more information moves through electrical pathways, you create more heat, more resistance, more power consumption and eventually you reach a limit. And uh think about the internet years ago. The most internet traffic traveled through again these copper wires. And today they go through the fiber optics. What we talked about and we talked about why because fiber uses light. Just in case you didn't know, fiber optics uses light. And light carries dramatically more information over long distances of time with less heat, greater efficiency. And that's why your internet is faster than it was 20 years ago. Now, here's where Tower comes in. They're helping bring that same idea inside of the chip. That's right. And that's where the term silicone photonix comes in at. Right now, don't let the name kind of scare you cuz I know it can. But silicone photonics sound complicated, but the idea is actually simple. When I read it, I was like, "Bro, they could have came up with a better name than this." Instead of using electricity to move information across the chip, you use tiny beams of light. That's replacing electrical traffic with optical traffic. Think about driving across Atlanta during rough show or LA or New York cars everywhere. Traffic backed up. Now imagine every car suddenly became a bullet train. That's what silicone photonix is doing to data. It's allowing information to move faster with less congestion at a far greater efficiency. And as AI models get smaller, that becomes incredibly valuable. And so you got to think about why this matters to AI. So I want to put it in perspective a little bit. When chat GPT answers your question, it's not one ship working. It's thousands of GPUs communicating and they're sharing the information. They're passing back calculations back and forth, training models together. And every second they spend waiting on one another, it costs time. It costs energy and it costs money. Now imagine an AI factory with over 100,000 GPUs. If communication improves by even a small percentage, those gains are enormous. Training becomes faster, inference becomes faster, energy consumption falls, heat falls, overall efficiency improved. And that's why one of the biggest companies in AI are investing heavily into optical networking because the next bottleneck isn't just computing anymore. It's not just energy anymore. It's communication and that's important. So Tower doesn't compete with Nvidia. They don't compete with Marll. They help make those companies get better. And so Marlli designs these advanced network and optical chips. And Nvidia increasingly depends on optical interconnections as AI clusters grow and Tower manufactures many of those highly specialized silicone photonic devices and it's very different from making a standard chip because building optical devices require manufacturing years of engineering special equipment precision that's difficult to replicate and that's why only a small number of companies around the world can do it at commercial scale and Tower has spent yield building out that expertise. This isn't something that competitors copy the next quarter. It takes decades to replicate. All right. Now, let's talk about the seconding company by the name of Koh, ticker symbol CHU. Now, let me ask you a question. Imagine if Apple builds 1 million iPhones. Would the box, you know, would they box them up, ship them all over the world without ever turning them on? Of course not. Every single phone gets tested. every camera, every speaker, every microphone, every component inside, just everything. Now, think about AI. If AI companies are spending hundreds of thousands, sometimes millions, billions on AI servers, do you think they're plugging chips into those servers hoping they work? Absolutely not. And that's where CO comes in. Now, think about them. Here's what they actually do. Think of a semiconductor factory, right? Every day, millions of chips come off the production line. Some are perfect. Some have microscopic defects. Some overheat. Some consume too much power. Some work today, but fail six months later. See, that's the problem with your eyes. You have to test for them. And that's exactly what CO who does. They build highly specialized machines that test semiconductor chips before they're shipped to consumers. Think of them as quality control for the entire semiconductor industry. Before a chip powers chat GPT, before it goes into a self-driving car before it ends up inside of a AI data center, it has to prove that it can perform. And companies like Coh. So, think of an airport. Here's the easiest way I can kind of break this down for you. Imagine you're going through airport security and everybody walks through the checkpoint. And some people keep moving, others get stopped for additional screening. Now imagine replacing those people with semiconductor ships. That's coo. Every chip goes through inspection. The good ones get to move forward and then the bad ships never leave the factory. The process protects the consumers who spending all of that money. It protects manufacturers and it protects billiondoll systems from failing. And so right now as system becomes faster, smaller, stronger, more powerful. They got to be tested and AI chips are no longer cheap. Years ago, if one chip failed, it wasn't a huge deal. Today, some of the most advanced AI accelerators cost tens of thousands of dollars each. So, a single AI rack contains millions of dollars worth of hardware. Now, imagine a faulty chip slipping through one chip. That's all it takes. And the server could fail. Training jobs could stop. Cloud customers could lose access. Companies could spend millions diagnosing a problem that could have been caught before the chip left the factory. Testing is no longer a nice feature. It becomes mission critical. As chips become more valuable, testing becomes more valuable. And that's why AI is counting on co. Let me bring it to my third company. This company teaches AI to think. N O data ticker symbol I N O D. This may be one that is the most misunderstood in artificial uh intelligence, right? Because when most people think about AI, they think about chat GPT, they think about Gemini, they think about Claude, they think about Grank, and they think somehow AI just became intelligent overnight. That's not what happened. It was taught. And that's where innote data comes in. Think about a child. Let's make it real simple. Imagine you have a 2-year-old child. That child doesn't know what a dog is. They don't know what a stop sign is. They don't know what sarcasm even sounds like. They don't even know what a smile means. So, somebody has to teach him day after day. So, picture after picture and word after word. Shout him is Rachel, right? Correction after correction, eventually the child starts to recognize the pattern. And that's exactly what artificial intelligence learns. Only instead of one child, it's a machine. And instead of one teacher, it's companies like, you know, data. AI doesn't learn from the internet. It learns from prepared data. Now, here's what most people get wrong. They think open AI just downloads the internet and searches the internet and AI became intelligent. It doesn't really work that way because the internet is messy. You and I know there's a lot of bad data out there. Imagine trying to study for the biggest test of your life using books with missing pages, wrong answers, and duplicate chapters. And most importantly, fake information, you'd fail. Before a models are trained, the information has to be prepared, organized verified labeled cleaned and checked. Let me say that again. And checked. And that's where innote data earns its money. Here's what they actually do. Imagine Google gives you 50 million documents. Some are duplicates, some are outdated, and some contain mistakes. Some are missing information. And now imagine training or trying to train the smartest AI model on earth using that mess. You can't. So, in data helps companies clean the data, organize the data, label the data and then verify the data so that the AI can respond and then it evaluates the AI responses and then it approves the training data sets in a simple language. They prepare a textbook before AI ever evens opens the book. So, it takes the garbage, get it out. There's old saying in technology, garbage in, garbage out. And if garbage goes in AI, garbage comes out AI. So if you train it bad information or train it on bad information, you'll get bad answer. And that doesn't matter how powerful the chip is. So what really matters is the information you put in. Even the memory doesn't matter because if none of the information that goes in good, everything else fall. And that's why companies continue to invest billions into quality data because AI is only as smart as the m information that it learns from. And that's why this market is getting bigger and bigger. And here's what excites me. People think AI training is over. They think it just knows what it knows. And I promise you, it's not because every new AI model needs better data. Every update needs more human feedback, even more improvement, which requires better evaluation. as AI becomes smarter and now the standard of data becomes even higher and that tells us that that means that the companies just throw information into AI. Nope. They need experts. They need scientists. They need doctors. They need lawyers. They need engineers. They need um electrical or financial professors and electrical professors and people who actually understand the greatest people each subject. it has to answer them correctly. That's a completely different challenge. And so when you think about who needs this, you should ask yourself this question. Who's building the biggest AI models in the world? Microsoft, Google, Meta, Open AI, Anthropic, maybe yes, Grank with XAI. The larger the model, the more high quality training that the data needs. And so that's going to keep expanding the work. So I want you to look at these three companies, Tower, Kohou, and Inno Data because as the RI AIA gets more competitive, companies are not only getting bigger chips, they're getting better data, and they're going to need more highways. They going to need more training. And so I hope that you take a look at this and understand that this is a great opportunity. So, even though stocks are selling off, this is a perfect time for us to increase and triple our network. It's your boy, the Wall Street Trapper. Make sure you like, subscribe, and share this out with somebody. And tell me below if you like more videos like this. Salute.