Contexte
So, now it's time to go on to the four companies I've identified that I believe are right in the middle of helping solve that shortage and should see outsized upside over the long run. Number four, Cororeweave, ticker symbol CRWV.
Contexte
Number three, Apply Digital, APLD. Now, Apply Digital builds AI factories and leases them to hyperscalers on 15-year take or pay contracts. It's basically a landlord.
Contexte
Number two, CleanSpark, ticker symbol CLSK. CleanSpark is a Bitcoin miner sitting on about 1.8 gigawatts of power, land, and data centers that just signed its first AI lease, its first proof of concept.
Contexte
Next number one, Nvidita semiconductor NVTs. Now Nvidita makes the power semiconductors gallium nitride and silicon carbide that let electricity get from the grid into an AI chip at the voltages the next generation of racks requires.
I'll break down the company and why you may want to put it on your radar and begin your due diligence on it.
Contexte
At the end of today's video, we have our sponsored segment on XCF Global, ticker symbol SAFX, on the NASDAQ. This small renewable fuels company just started producing fuel at a Nevada refinery permitted for 38 million gallons a year... I'll break down the company and why you may want to put it on your radar and begin your due diligence on it.
Transcription Complète
Folks, you better get ready because there is a massive shortage in one specific area of the AI trait. There's a massive structural shortage that few people are paying attention to right now. And if you could position yourself correctly to benefit from said shortage, well, you might find that you're very happy with the result. In today's video, I'm going to break down what exactly the shortage is, why it's such a lucrative opportunity, and then the four companies that are specifically targeting said shortage. And then at the end of today's video, we have our sponsored segment on XCF Global, ticker symbol SAFX, on the NASDAQ. This small renewable fuels company just started producing fuel at a Nevada refinery permitted for 38 million gallons a year, and it is doing it into a diesel market that has run up more than 85% since the start of the year. I'll break down the company and why you may want to put it on your radar and begin your due diligence on it. And as always, if you're the one taking the ultimate risk, you got to be the one doing the ultimate frisk. Always do your own due diligence and all ideas presented. Okay, so for starters, the context here is that every stock follows a cycle. You have three stages. Stage one, the early stage. This is when the seed is being planted. You have high promises, but very little is proven yet. It's cheap precisely because of that risk. And the stock goes up and down like crazy. Sure, but overall it's still very cheap compared to the potential. Stage two, the building stage, the company's actually building and executing on the vision. You're getting mixed proof of concept. Some things are working, some things are still unproven. there's some failures and so on and so forth. And depending on where we are in the market cycle, these can be priced dirt cheap or very, very expensive. Stage two is where we like to find companies for the optimal risk versus reward setups. And then stage three, you have maturity. This is where most people like to buy companies because it's a lot safer. However, the stock has become a mature company at this point, meaning that most of the extreme growth is behind it, but the proof of concept is undeniable. So, your risk is very, very low relative to the other stages. Mature companies can absolutely still deliver above average returns, especially if you buy it during a big dip period or a market panic period or if you're buying in a sector that overall is earlier stage. But if you're going to break it down, every stock falls into one of these three categories. Now, within the AI sector, some of the best opportunities are in the early parts of the building stage. How do we know that they're actually building something useful? Well, we want to find companies that are building into a supply shortage. Basically, they own the shortage. If you're building something with long-term supply constraints and there's not enough supply coming onto the market to match whatever demand increase, well, you could be pretty damn certain that that company has a lot of tailwinds in its favor. So, what's the biggest shortage overall? Bigger picture of the AI trade data centers. That's the shortage right now. That's the structural shortage dating out years. Supply is extremely constrained in the segment while demand is growing exponentially and that gap is set to get significantly worse over the coming years. Now, it's very important to understand though, this is not a short-term supply imbalance. A lot of bears like to say, "Yeah, all of these data centers are going to be built and then data center value is going to collapse and everybody's going to be left holding the bag." However, the unfortunate thing is that you're mixing up short-term and long-term trajectories. Short-term imbalances and long-term imbalances in supply. Just for a good example, right now I'm on vacation. As you know, in August, that's when everybody travels. So, the prices of hotels are unfortunately more expensive. That's a seasonal blip that happens all the time. But now if you flip to regular housing that people actually live in long term, that's a completely different issue. That's a long-term supply shortage. Housing supply over the long run has not kept up with demand. So housing in almost every market in the US of A has gotten more and more expensive. So that's the difference between a short-term supply and demand imbalance versus a long-term supply and demand imbalance. A lot of people are right now treating this AI trade and the specific players within the AI trade that are actually solving the shortage as short-term winners. And again, sure, the price of these stocks is cyclical. However, the long-term imbalance is a completely different story, and it's not going to be solved anywhere near soon, any more than say the housing shortage is going to be solved anytime soon. If you analyze just the rental market in the US, if you want to ask about whether a rental market is tight, you ask a question, how many units are sitting empty? In a normal healthy city, maybe six or seven out of every 100 apartments are vacant at any given time because people move, leases end, landlords repaint. That's normal. Some markets are going to be worse. Take New York City. Marcus and Milichap projects NYC will have the lowest apartment vacancy of any major US market this year about 2.4%. And that would be its 11th consecutive year under 3%. Why? Well, because the supply is way more limited than the demand. And that's accounting for the outflow that New York State and New York City have seen over the previous years. Now, remember that vacancy rate is 2.4%. According to CBRE's year-end 2025 North America data vacancy in the biggest US data center markets is 1.4%. That means for every 1,000 units of data center space, about 14 are available. 14. And in Northern Virginia, the largest data center market on Earth, CBRE's Q1 2026 global report has vacancy at an all-time low of 0.3%. Three out of a thousand. For all practical purposes, there's nothing available. If you show up today with a checkbook and you're like, I want to rent this space, they're going to laugh in your face. They're going to say, haha. So, data centers are more scarce than apartments in Manhattan. Northern Virginia is roughly eight times tighter than the tightest housing market in America. Now, with housing, it's important to understand that in some of the most tight rental markets, well, demand is flat or actually shrinking. Out migration has been a big problem in states like New York, California, you know the story. With data centers, though, demand is compounding and at the same time, states and localities are pushing back on new construction. Some counties have passed outright moratoriums. The supply side to build new supply of data centers is becoming more hostile than the supply side of building new housing, which in and of itself has been pretty hostile over many years. Well, communities are a lot more hostile to data centers. Okay, but Charlie, wait a second. There are lots of areas where they can build. Why don't they just use those areas and build much much faster? Well, even if you have all the approvals, it still takes a long time to build out an actual data center. Traditional construction timelines for a small data center used to be around 12 to 18 months, but those timelines don't apply anymore. The shift towards more powerful AI campuses has pushed schedules into multi-year territory, and interconnection alone can add 24, 36, or 48 plus months. Most of the buildings in the construction pipeline are already are already pre-leasased. There's years of shortages already locked in. So, even if these companies decide, we don't want to do any more buildout. We're just going to finish what we've already signed locked in contracts for. Well, most of the stuff is still sold out to 2029, 2030. So, there you have the floor. However, if you're not going to be a doom and gloomer and you assume that the overall trajectory is what the average analyst expects, which means massive buildout for not just the next four, five, six years, but for many decades to come. Well, you could see that the supply and the demand very different, very long-term structural disparities. So, now it's time to go on to the four companies I've identified that I believe are right in the middle of helping solve that shortage and should see outsized upside over the long run. Number four, Cororeweave, ticker symbol CRWV. So, Cororeweave is the largest pureplay AI cloud company in the market. Infrastructure built from the ground up specifically for AI workloads rather than retrofitted from general purpose cloud. Think about the difference between a commercial kitchen and a home kitchen. both cook food. But if you're trying to serve a thousand dinners a night, the home kitchen doesn't just underperform. It physically cannot do the job. Liquid cooling, highdensity racks, the networking fabric, that's the commercial kitchen. That's what the commercial kitchen has, and that's what it needs. Most of the other players are basically just like the home kitchen. Sure, they could produce a few meals or something, but they don't have the full stack available. Now, Cororeweave has a relationship advantage that's very hard to replicate and almost impossible to see on a balance sheet alone. They are effectively a reference customer for Nvidia. When a new architecture ships, they're among the first to have it racked and running at scale. That matters a lot more than it sounds. In a market where the constraint is time, being six months earlier to deploy a new generation of hardware is a genuine competitive edge. And that edge compounds because being early is what earns you the allocation to be early again next time. However, if you're looking at what value you're getting per dollar right now based on the current numbers, Corweave is trading around 91 bucks a share, roughly $49 billion market cap. But if you look at the backlog, the backlog is $99.4 billion. So the entire company is valued at about 50 cents for every dollar of revenue it is already signed not projected signed on forward sales guidance is 12 to$13 billion this year. That's roughly four times forward revenue and management says annualized revenue should exceed 30 billion by the end of 2027 which means you're paying about 1.6 times the exit run rate they're guiding to 18 months from now. For comparison, many smaller Neocloud plays traded around 10 times forward revenue. And the stock also happens to be very well off its 52- week highs. Number three, Apply Digital, APLD. Now, Apply Digital builds AI factories and leases them to hyperscalers on 15-year take or pay contracts. It's basically a landlord. Now, full disclosure, I've been making videos on this stock since very tiny fractions of the current price, well, as much higher prices. And quite frankly, I don't think that it's ever factored in the long-term value. To use the ever tired gold rush analogy, think of two people making money in a gold rush. One buys the equipment, digs, and hopes for gold. The other owns the land and charges rent regardless of what comes out of the ground. Apply Digital in this situation would be the second one. Now, what are you actually getting per dollar though? Well, Applied Digital's market cap right now is about $8.5 billion. At the same time, it's signed, contracted, take backlog $36 billion. $36 billion compared to an $ 8.5 billion market cap on a company that's long been proven, a company that just recently traded at much higher prices and a company that has very solid execution date now years. Right now, you're paying about 23 cents for every dollar of already contracted base term revenue. If you want to add the debt and call it enterprise value, you're still under about 40 cents on the dollar. Now, if you look at their margin structure, expected site NOI runs 85 to 86%. So $36 billion the $36 billion of contracted revenue translates to something on the order of $30 billion of lifetime net operating income on a $8.5 billion company. So if you want to just look at net operating income NOI, $8.5 billion market cap, $30 billion NOI. This company's been beat down because of market dynamics and market flows having nothing to do with the company itself. This company's executing. This company's at a great price right now. Number two, CleanSpark, ticker symbol CLSK. CleanSpark is a Bitcoin miner sitting on about 1.8 gigawatts of power, land, and data centers that just signed its first AI lease, its first proof of concept. And the market though, it's still valuing it like a Bitcoin miner, and nobody likes Bitcoin right now. So, it's down a lot because of that. But if you actually look at the long-term potential of this company transforming into an AI data center play, well, that can convert to a ton of upside. If I was going to make an analogy, imagine somebody owns a parking lot in a neighborhood that just got reszoned for highrises. They're still collecting parking fees. The income statement still says parking lot, but the land says something completely different. The market is pricing in parking fees and the limited upside with that instead of the massive apartment complex that's going to go on there. That's basically the flip of going from a Bitcoin miner to an AI data center compute play. And more importantly, they spent years buying power and land in rural Georgia when absolutely nobody wanted it. Not because they foresaw the AI boom. Again, they were mining Bitcoin and cheap power is how you mine Bitcoin profitably. But the consequence of this is that they assembled a 1.8 8 gawatt portfolio at preAI prices in a state that has since become one of the most contested data center markets in the country. So you actually can't do what they did anymore. You can't do that anymore in Georgia. Now if you look at the proof of concept, CleanSpark signed its first high performance computing data center lease, roughly $6.6 billion in contracted revenue over an initial 20-year term. It also includes two 5-year extension options. That contract climbs to 11.6 billion with those extensions. Now there's another part to this. Alongside the lease, the same tenant signed a letter of intent granting exclusivity over Clean Spark's entire Texas portfolio, which is about 718 acres up to 885 megawws, potentially several times larger than the Georgia deal with the same counterparty. So, what are you actually getting here per dollar? What is the value here? Well, CleanSpark's market cap is around $3.3 billion. The lease they just signed is worth $6.6 billion over 20 years or say 11.6 billion with both extensions. So the entire company trades at about 50 cents per dollar of contracted lease revenue and about 28 cents if the extensions ultimately get exercised. That's all from one lease on one campus out of a 1.8 gawatt portfolio. Next number one, Nvidita semiconductor NVTs. Now Nvidita makes the power semiconductors gallium nitride and silicon carbide that let electricity get from the grid into an AI chip at the voltages the next generation of racks requires. This is a key part of solving for the overall shortage of data centers in terms of the power bottleneck. Think about the power lines running across the country. They carry electricity at tens of thousands of volts and then a transformer near your house drops it down to 240 volts you actually use. Why bother with all that? Well, because high voltage is what lets you use thin wires. If you sent power at 240 volts the entire way from the plant, the wire would need to be about as thick as a tree trunk. Not just expensive, but that would be impossible. Data centers have now hit that exact same wall, except inside the buildings themselves. For 20 years, racks moved power around at about 54 volts, which was completely fine when a rack used a few kilowatts. The new AI racks need a megawatt. And at 54 volts, that would sadly take roughly 450 lbs of copper stuffed inside a single cabinet. It just doesn't fit. So, the industry is doing exactly what the power grid figured out a century ago. crank the voltage up to 800 watts and go back to thin wires. Now, Nvitas is the clear player to make that work. Nvidia is the company driving the Switch and Nvidas is inside the reference design for it already. Now, once you're written into a blueprint, swapping you out means retesting an entirely new system. And nobody does that to save a few dollars on a rack that costs millions of dollars. And you can already see this turn into numbers, even though they're small. While Nvidas' high power revenue grew more than 50% year-over-year, guidance for the current quarter is 13.5 million against roughly 11.1 million that Wall Street had expected. Gross margins are rising on their own just from selling more high power product and less cheap consumer gear. Bookings are at a record and they're carrying 557 million in cash with no debt. But overall, this company is down well from its 52- week highs. It has a ton of proof of concept. The numbers are improving rapidly and I think the next couple of years are going to be great for Nvidas in terms of catalyst that I want you to pay attention to. Number one, third quarter earnings early November. Their management has guided AI infrastructure to cross onethird of total revenue by the fourth quarter. So this is the report, the proof and not promises report where the pivot either shows up in the actual numbers or it doesn't. Catalyst number two, Nvidia's GTC March 2027. So GTC 2026 is where Nvidita unveiled its 800volt power delivery board and it kicked off a run that had the stock up some 346% year-to- date by early June. And by the way, if you look back at our timing, we actually called this run out as well. But anyways, I think this is going to be another opportunity for the stock to realize a lot of gains. Number three, deeper placement in Nvidia's MGX ecosystem. When Nvidia showcased the Nvidita's 800 volt to 6VT board inside its MGX AI factory ecosystem, the stock gained some 18% in one session with an intraday peak above 22% adding close to a billion dollars of market value in that day. So any kind of ecosystem placement news that's similar to this can cause big rally rallitos in the price. Of course, it's hard to put a date on those. You don't know when those are going to come out. 2027 production ramp, that's another catalyst. products are sampling now with major ramps flagged for 2027 which means the first quarter where 800 volt content converts from design wins into actually shipped revenue which people are going to love to see markets are going to love to see that very importantly customers beyond Nvidia the recent licensing deal giving magnet chip access to their GNSIC gen 4 and Gen 5 technology shows the intellectual property can produce revenue without Nvidita's manufacturing every part itself which is something good to see and any additional hyperscaler or grid infrastructure wins can cause quite the rally rallito for the stock and would be a sufficient catalyst. Anyways, those are the four stocks that I believe are set to benefit from the massive data center shortage. And now it is time for our sponsored segment on XCF Global, ticker symbol SAFX on the NASDAQ. So, what is XCF Global, Charlie? Well, XCF is a Houston-based producer of renewable diesel and sustainable aviation fuel, SAF for short. Its flagship asset is a refinery outside Reno, Nevada called New Rise Renewables Reno, permitted for a name plate capacity of 38 million gallons a year. In July of this year, that plant began producing renewable fuels, starting with renewable diesel with the plan being to shift the configuration towards sustainable aviation fuel as the ramp progresses. Beyond Reno, the company is working a pipeline of expansion opportunities in Nevada, North Carolina, and Florida, plus a proposed platform in Australia. And it has a three-way merger pending that would fold in a carbon credit business and a green methanol developer. Now, what's the market gap? Well, the gap here is a supply gap, and it's a large one. Aviation is one of the hardest sectors in the economy to decarbonize. Cannot run a longhaul widebody on a battery. The only near-term option that works with existing aircraft and existing airport infrastructure is a drop in liquid fuel made from waste fats used cooking oil and vegetable oils, which is exactly what sustainable aviation fuel is. It goes into the same tanks through the same pipelines into the same engines. The demand side is being set by regulation rather than preference. Europe's refuel EU aviation rules require a minimum share of SAFF in jet fuel supplied at EU airports starting at 2% to 2025 stepping to 6% by 2030 and climbing from there through mid-century. Airlines have made their own net zero commitments. On top of that, corporate travel programs are buying SEF certificates to hit reporting targets. The supply side has not kept up. SAF is still a low singledigit fraction of global jet fuel consumption. Building a facility that can actually make it is a permitting engineering and capital problem that takes years and a number of announced projects have slipped or been sheld. That is the core of XCF's pitch. The same hydrorocessing configuration that makes SAF also makes renewable diesel. And renewable diesel has its own demand pull from state lowcarbon fuel programs. US diesel futures are up more than 85% since the start of 2026 according to the company's most recent shareholder letter. and jet fuel pricing has been elevated alongside it on the back of the crude and refining disruptions we have covered all year. So the product being made right now is landing onto a strong pricing backdrop. Now what is the business model Charlie charito? Well XCF's model is assetbased. It owns and operates the plant buys feed stock converts it and sells the fuel plus the environmental credits that come attached to it. New Rise Reena went through a conversion project this year with Axens, a global licenser of refining and biofuels technology as its technology partner. Catalyst changes and equipment upgrades were the bulk of the work. According to the company shareholder letter, the plant is now running at process temperatures roughly 40° lower than before the upgrade, which XCF believes may translate into lower energy use and a lower carbon footprint per gallon as volumes come up. Production started with renewable diesel in July under a commercial framework that covers feed stock supply, logistics, and commercialization. Throughput is being brought up in a measured way as systems are validated under load, which is standard practice for a plant of this type. XCF signed a memorandum of understanding with BGN internationally, global energy and commodities group covering a tolling framework for SAF and renewable NAPA and joint development of distribution, marketing, and offtake across Europe, the Middle East, and other markets. On June 30th, the company signed a joint commercialization and development agreement with continual renewable ventures and New Rise Australia to advance a proposed refinery in Australia, a hea based facility with expected capacity of about 175 million liters a year of SAF and renewable diesel. Under that structure, XCF could provide technology and project support and can earn up to 10% equity interest in two milestone braced based tranches. While the partner retains majority ownership, it's a capital light way to attach the company's name and knowhow to a second geography. For the next domestic unit, New Rise Reno2 XCF has engaged Bank of America to help structure potential debt financing, including financing that may qualify for export credit agency programs. In terms of the leadership, Chris Cooper serves as CEO and was appointed chair of the board in June of 2026, consolidating leadership as the company moved from a construction story into an operating one. The commercial partnerships announced this year, Axens on the technology side, BGN on the offtake, Bank of America on project financing, and the CRV agreement in Australia have all been signed under his tenure. Now, it's important to understand that this is a micro cap company with extreme risks. Most small cap companies fail, and there's always a very long-term risk of dilution and heavy dilution in this space. And that is very much an ongoing risk with this company. Dilution is all but guaranteed. Refineries are enormously capital intensive. and this company has been carrying a heavy debt load against a small cash position while the plant was not generating revenue. These are all things to consider when you're doing your own research and coming to your own conclusion on this. Anyways, to cap it off, XCF Global is a very early stage, highly leveraged renewable fuels company that just crossed the line from building a refinery to running one. The case for the company is straightforward. They have a permitted 38 million gallon capacity that is now producing a fuel market with regulatory demand behind it and strong pricing in front of it. a potential technology partner in Axens, commodities partner in BGN, and some potential regulatory tailwinds in terms of the push to more renewable energy. Anyways, take a look at XCF Global's investor relations page down below. And as always, do your own due diligence. This is a sponsored segment and it is not financial advice. Have a great rest of your day. We'll see you in the next video.
Commentaires 0
Connectez-vous pour rejoindre la discussion.
Se connecterAucun commentaire pour l'instant. Soyez le premier à partager votre avis !