Wall Street Just Gave a Dire Warning. (Most Aren't Ready)

More than $2.5 trillion dollars in market  value gone in just two days. Headlines are   soaring saying big tech's cash burn is too big  to ignore. But is this sell-off overblown? And   is there a deeper story brewing long-term? Market  beat analyst Thomas Hughes joins us with his take   on what's happening behind the scenes with the  Hyperscalers and what their long-term story could   look like. He's going to dive into three specific  names later on in this video. Now Thomas, I want   to start out with looking at where the market  is right now in this reaction to big tech. Well,   right now the market uh few weeks ago hit a peak  and we seem to be struggling to regain traction.   The fears of the data center buildout and the  spending, the impact on debt and balance sheets   and cash flow is really giving the market some  pause. Yeah, you can absolutely see that pullback   in the heat map over the last seven days or so.  Really concentrated on big tech. You look at   Oracle, it's down more than 14% in the last 7 days  of trade. Google's down almost 4% and a lot of red   in that big tech sector. Even the green spaces are  only only seeing slight increases right now. This   comes after what we saw a huge boom in these big  tech names last month. So, want to get to what's   behind the pullback that we're seeing. And one  headline that keeps popping up is the cash burn.   How much money these hyperscalers said they're  spending. All those details came out during those   Q2 earnings reports. So what's your take when you  look at that cash burn right now, Thomas? Yeah,   so right now um roughly speaking over the last  18 months, hyperscalers and AI infrastructure   companies have taken on about three4ers of a  trillion dollars in debt, including dilutive   actions to try to raise capital for the data  center buildout. They're using their free cash   flow to do this. Uh the new debt load is also  impairing their free cash flow. They're having   to divert cash flow to these uh debt maintenance.  Uh they're also have historically bought back a   lot of shares and the money that's going to share  buybacks is also going to this debt maintenance   and that's really impairing the cash flow. I  think you showed me a chart earlier today where   uh the cash flow went from being way up here to  way down there and that's really really a scary   thing for investors to look at especially in the  near term. It's certainly a headwind for the for   the market for market sentiment and for for share  price action. Yeah, that chart was from an article   on Yahoo Finance today. Kudos to Yahoo for this  chart and looking at this comparison. We know it's   not just this last quarter of the amount of money  that they're spending, but they've been investing   for the last few quarters tremendous amounts  of money into this AI buildout. I think the   biggest question, and that's really the thesis of  this video, is there a long-term payout for these   companies? Is this a huge bet and a huge risk  that they're taking on, or will it pay off in   the future? That's the biggest question that every  investor has. And we're going to break down that   question little by little. It's a big question.  So let's start a little bit with do you think   that there is a risk for hyperscalers right  now with this amount of cash burn? There is a   risk because it is a lot of cash. It all comes  down to execution. They need to be able to to   build out these systems and bring them online and  then to monetize them. Now in the realm of risk,   uh these companies aren't as risky as emergent  tech stocks. This is not some new technology,   emergent technology where they're having to  establish a business. This is the evolution   of current technology. All I mean, every single  one of the major leading companies are involved   with this shift. They're all working together to  make it happen. So, in that sense, it is d-risk.   Technology is moving this direction. Um, it's not  about having to prove it. It's just a matter of   building it and getting it um up and running and  monetizing it. Uh, so to me, risky, yes, but it's   an execution story. uh the risk will be reflected  in, you know, periodic price corrections,   not so much in stock price implosions. I think  over time these companies will continue to trend   higher. Yeah, I think that's what the market has  gotten used to is these companies really leading   the market, really outperforming the market with  how well they're doing. They've gotten used to   those stock buybacks. They've gotten used to the  benefits these hyperscalers have had of continuing   to grow at the pace that they're growing at. that  this new investment cycle that they're doing of   having to invest so heavily in building out the  AI story is a little concern for investors and and   one thing I want to mention is retail investors  when we talk about hyperscaler stocks they're like   I don't want to talk about Google I don't want  to talk about Nvidia these are boring stocks but   even if you are a retail investor I think it's  important to say these are stocks that pretty   much every investor owns whether you're buying  stocks on your own if you've got a 401k of any   kind these stocks are so deeply entrenched into  every ETF every major system out there that I I   think everybody owns a piece of these stocks and  so everyone wants to be prepared on what could   this mean for their portfolio, right? You're  you're totally right. Uh these stocks, I mean,   everybody owns them. They are the biggest uh  revenue producers, cash producers. They have   really visible growth stories. Um investors that  don't own them outright probably own them in an   ETF or a fund. So, as you're buying them, you'd  keep that in mind. Uh you may want to get exposure   to some of these stocks, but you may already  have it. Yeah. We'll talk about whether getting   exposure or buying any dips that we see in these  hyperscaler names. And Thomas is going to get into   some specific companies here in just a minute.  But first, Thomas, I want you to get into your   theory on whether the cash burn happening right  now is a real concern for investors or whether   you think it tells a deeper story long term. It  tells a deeper story long term. And evidence of   why this isn't like an emergent tech story, but  really the evolution of tech is the backlog. all   of this spending, all of this new debt this data  center built out isn't in the hopes of business   coming, it's to fulfill business that has been  contracted. So while the debt has been blossoming,   the backlogs for all these companies has been  blossoming at an accelerated pace. So right   now over this, you know, the same period backlog  amongst hyperscalers, AI infrastructure companies   from chips through the nuts and bolts place  through the GPU as a rental companies is about 2.1   trillion. and that's about three times the debt  raise. So the long-term outlook is for them to   convert that debt into this revenue which is three  times as much. So we're looking at pretty robust   growth over the next few years. At the same time,  debt levels will come down. Free cash flow will   come down. Growth will be accelerated. They'll be  able to start reacelerating share buybacks at the   same time. And then you have to think that this  backlog is contracted for a set number of years.   So we think that this 2.1 trillion is probably  good for the next, you know, three, four,   five years. Once they run out, there will be new  contracts. There won't be any new debt because   the data centers will already have been built.  There'll just be new contracts with even better   margins. Uh so to me, the long-term outlook for  AI, generally speaking, is is very very robust.   All right, looking at the backlog is really  interesting. And we're not just talking about   the backlog for the companies that are doing the  physical buildout themselves, right? We know that   Google, Amazon, Meta, they're investing in  the buildout of a lot of these places. So,   it's kind of transferring money from these these  big tech companies. Some of those infrastructure   picks and shovels companies that we talk about  all the time, the ones that are seeing tremendous   growth right now, but we know that their bills are  getting paid because the hyperscalers are paying   them. But do you see that backlog also showing up  for the hyperscalers? Oh, yeah. It's throughout   the stack. So it's it's not just for the GPUs that  underpin AI and not just for the nuts and bolts   that put the GPUs together in the data centers but  for the actual capacity. These big hyperscalers   uh these big software as a service companies are  have already and are continuing to contract this   capacity and they're the ones that are really  underpinning the growth because it's the demand   for the capacity that drives demand for the  infrastructure. And so right now we're in the   infrastructure buildout and the long-term play  is to transition from the infrastructure buildout   to the usage of capacity. When we start using  the capacity will start to convert that backlog   really meaningfully into revenue into earnings.  Yeah, there's no question that demand is there   for more AI compute space. There's so many use  cases popping up for businesses, companies, even   retail investors are starting to use AI on a daily  basis to get new ideas, especially those who are   looking for short-term trades. One of our regular  guests, Keith Kaplan, the CEO of Tradesmith, has   a new super AI program that has been incredibly  successful. It has a 73% historical accuracy rate   after scanning over 2300 individual stocks down  to the penny. This is kind of built like weather   software where they predict where storms are  coming. This program helps to predict where   a storm could be coming in the market. And again,  it's incredibly accurate. And there is a new storm   brewing according to the software coming up at  the end of June. So by June 29th, make sure to   check out this special offer for our market beat  viewers for tradesmith. You can see a free demo of   how this predictive software works. They'll give  you a stock to look at too that can move on this   June 29th deadline. So, if you want to scan the  QR code or click the link in the description, you   can check out this new AI software for yourself  and see which stocks they are expecting to move   on this June 29th catalyst that their AI system  has predicted. Again, it's special offer just for   our market viewers. Scan the QR code or click the  link to see how it works. So, Thomas, clearly this   is just one of many ways that people are using AI  right now and increasing that demand for AI. And   that means that they need to build the capacity  for these programs to exist. And so that's why we   see these hyperscalers investing so much in this  buildout right now. Google is one of the perfect   poster childs for this. They're spending so much  on AI right now. But I think the question that   investors have is how are they going to monetize  that? So let's get into Google first. Uh where   do you think that their monetization story plays  out down the road? Yeah, I mean Google's a good   example. Well, you look at the charts, I think all  the Mag 7 have pulled back pretty good right now,   Nvidia included. But I think this is just a  natural mechanic in the between earning cycle   period. Uh the market rocketed pretty uh you know,  pretty robustly since the last earnings report.   Uh setting itself up for a correction. So right  now the market is again setting up for another   um to me another upswing. Uh we're expecting  earnings to come out uh kind two three or four   weeks from now and all the reports should still  be should be pretty good. they should affirm the   trends that show, you know, solid spending plans,  plans for data center capacity and also demand   for the the the end market capacity for the the  end market users. I think that's one thing that   I've heard talked about a little bit is that in  these earnings reports, it's good news for the   AI buildout story because when Google reports,  they show how much money they are investing into   this buildout. uh are they doing a good enough  job these hyperscalers during that earnings call   to really detail how this investment they're  making is going to uh lead to more revenue for   Google itself later on. Yeah, I I think generally  speaking they are because these contracts are all   very backended. They're all dependent on on future  capacity. You have to think that current capacity   is already maxed out. That's what's driving  some of the price increases. So a lot of these   contracts are based on future capacity and future  technology. A lot of these data centers will be   built using the Ver Rubin products which are only  just now coming out. Also going to be built on the   AMD Mi450 products which have not yet come out.  So you have to think that you know this backlog   is $2.1 trillion. But a lot of it's not going  to start being recognized as revenue until late   next year, early 2028 as these new data centers  come online and we're able to start converting   that capacity into actual revenue. Yeah, that  timeline is an important one for investors to   keep in mind because hearing that that converting  to revenue is still a year or two years out makes   me wonder if we're going to continue to see this  kind of volatility happen in some of these giant   tech names where they're going to lose some of  their market cap simply because investors aren't   ready to hold on to that risk until they see  the payoff coming. Do you think that the the   that volatility is going to continue and the the  speculation and fears about how much money they're   spending before we get to that revenue point  could continue for another 1 to two years? Yeah,   I think we will see some volatility continue over  the next few quarters. I think that we're going   to continue to see the AI bubble swell. That's  going to drive some pretty solid upswings in   between the earning cycles. There's going to  be plenty of opportunities for the market to   to grow concerned or to have some caution or just  simple profit taking set in to provide pullbacks.   But I think in this environment right now, price  weaknesses will be buying opportunities because   down the road there will be more catalysts. Not  just today's results and and the guidance, but   the actual conversion of these backlogs into cash  flow and earnings to drive the stock prices higher   over time. All right, so we talked about Google.  Let's talk about a few other specific companies   that you are looking at right now. I know one of  those is Oracle. You just had an article out about   the pullback we're seeing there. And this is one  example of a company spending quite a bit of money   on this buildout right now. Yeah, Oracle is one of  the poster children for this story. It has really   been leaning hard into debt, but it's also got one  of the most aggressively growing backlogs for me.   Oracle is one of the premier stories here because  it's being elevated not just from a legacy tech   company to a new modern cloud-based AI tech  company but also from like a smaller niche   player to being really central uh topline critical  bluechip um tech player uh central to AI and data   centers. Oracle is the hyperscaler serving the  hyperscalers. It provides high capacity compute,   high performance compute to all the other major  hyperscalers to meta to uh the AI laboratories   as well as to businesses and enterprises. Um and  that's just the cloud business. Its actual core   business which is database services is ubiquitous  across the cloud. It's integrated into all the   major hyperscalers systems and networks. it is  the most readily available uh database globally   and that just entrenches it uh to me as a very  key critical primary player in the entire um AI   infrastructure and um and industry. Do you think  that Oracle holds more of a risk than Google or   some of the other MAG7 names simply because it it  didn't start with the kind of balance sheet that   these giant mag 7 companies did? I I don't think  it has a risk. Um its balance sheet was actually   fairly healthy before this started. it was buying  back shares pretty aggressively. There were some   concerns with some of the metrics, but that was  just offset by the the cash flow and aggressive   share buybacks. Um, I think right now the biggest  concern is just that the the debt is swelling.   It's going to be an impairment for the next, you  know, 12 18 months probably, maybe longer. But   like I've been saying, they're going to start  recognizing their their backlog as revenue uh   beginning next year and ramping that over the next  few quarters. That's really going to, you know,   whittle down that debt real real fast. So, I think  the amount of stock price pressure that we've seen   this year will be reversed and reflected in stock  price increases in the upcoming years. All right.   Well, let's look at the chart action for this  one in particular. We started the video saying   this one's down about 15% in the last few days,  and that's just the start of the volatility it's   seen over the last year. It hit a new high back  in September, and it's been on a volatility swing   ever since then. This would be a great company to  sign up for our SMS text alert system. If you are   a regular viewer, you may have seen us say you can  text YouTube to 68285 to sign up for Marketbeat's   text alert system. It'll let you know anytime  a stock on your watch list has a major move in   either direction. Again, Thomas, this is a good  one to sign up for that alert system for because   it has moved so much in both directions. But which  way is it headed next? Do you think that this is   the start of a potential deeper downtrend for this  stock or is it a good bottom that investors should   look at buying? Well, this is not the start of  a downtrend. This is to me absolutely a buying   opportunity. What I see in Oracle's chart is last  year was that big huge price run up based on the   swelling backlog. Uh since the peak last year, the  price has pulled back, but that's due to fears of   the debt. Since then, we're showing a pretty clear  bottom. Uh we are pulling back this week the last   few weeks but within this to me what I'm seeing is  a head and shoulders bottom. I'm seeing a market   that's incredibly oversold overreacted to the  fears totally set up to rebound and all we need   is a catalyst to drive it. Oracle is a midcycle  reporter so it actually reported not too long ago.   It'll be quite a few more weeks before it reports  again. It'll be after all the other hyperscalers   report. But to me, all those other reports will be  catalyst for Oracle because they're all going to   be talking about their spending and some of that  spending is going towards Oracle for Oracle's own   hypers scale buildup. I know you mentioned  that Oracle won't be reporting for a while,   but the the big earnings report that everyone is  looking forward to today is MU. The Micron report   comes out at market close today. We're recording  this just before market close, so we don't have   that report yet in this as we're talking about  this, but I'm sure everyone has high expectations   on how much of that spend is going towards  companies like Micron and and companies uh like   these chip and memory stocks out there. So, does  that have any impact or connection with Oracle or   this broader story about this massive investment  going on right now? Right now, a lot of money is   going into the chips and not just into the GPUs  or the memory, which is what Micron is making.   Uh Micron stock itself just hit a high this week.  It's pulled back from that high kind of ahead   of the release. But to me, this chart is very  bullish. Uh with Micron, we're expecting to see   them show um continued highle demand potentially  pushing out their own timeline to be able to meet   capacity. Right now, they're sold out through the  end of next year. Uh this report might extend that   into 2028. That's really supporting uh the outlook  for for prices, which is helping to drive Micron's   business right now. Not only is it the volume of  sales but also the price that it's receiving but   then looking at the broader chip sector uh  generally speaking um AI is driving demand   across the board. I mean every kind of chip that  you could imagine is needed to build these data   centers and then to help apply AI long term uh  through the IoT and through um AI applications   because the GPUs are just the um the product  that does like the computing. Then you have   all the chips that are needed to connect the GPUs  together, connect the servers together, connect   the data centers together, power control devices  and all the actuators and all the servers that   uh help to drive the internet of things because  AI is really going to be about the application of   physical AI and IoT which is using that computer  brain to make machines do jobs remotely all around   the world. All right, it can even take over your  job as an investor. So many investors are using   AI. Much like that predictive AI program we just  talked about from Tradesmith, there are so many   use cases for how demand for AI is just continuing  to grow because as people use it, they find it's   incredibly successful. Again, a reminder about  that QR code or that link. If you want to check   out that predicted AI software from Tradesmith,  you can do that with a special offer in this   video. But Thomas, the question is how much longer  can this super growth cycle really continue for   these companies within chips? The chip industry  is then a a massive um super cycle that was driven   by um inventory normalization and now boosted by  AI. That should last for years until they're able   to start, you know, ramping capacity to meet the  demand. As far as memory chips go, it's going to   be at least another year or so before they're able  to really significantly ramp capacity to meet that   demand. And that's just going to continue to drive  success for chipmakers over the next, you know,   year or so. But then you look at it longer  term, AI is driving a virtuous cycle. That's   where spending on AI creates new capacity and new  technology that improves the output, which means   that you can spend more money on new chips to do  the same thing over and over again. So each time   that we go into a cycle and spend some money to  advance AI, the outputs help us advance technology   to the point that we're going to spend some more  money to do the same thing over and over again.   And right now it looks like the virtuous cycle  in AI could last indefinitely. Uh it's really   very early in this in this thing and we really  even haven't built out the first generation of   AI data centers. That's what we've been talking  about. That first generation will start coming   online sometime next year. That's a really good  point to bring it back to that fear that so many   investors have in the market right now is that all  of this money isn't being invested into building   out this first generation of data centers like you  said. But how quickly will those data centers be   obsolete? Will they get the use out of them that  these hyperscalers are expecting or will there   be something new by the time that this buildout  is done? And I think that infrastructure piece,   that unknown of how quickly everything is  developing is one of the other fears that so many   investors have in the market and why we see the  market react. So, what do you say to those fears   of when this buildout is done, will it pay off  or will there already be something bigger, newer,   and better that they need to pivot to? In the tech  world, once these data centers are built, there   will be something better coming down the pipe. But  it takes time for those things to happen. You got   to think that this current AI data center boom has  been building for the last couple years, and it   really hasn't borne any fruit yet. And we have had  seen some fruit. We've seen some companies serious   demand for their infrastructure products, but  those data centers really have not yet been built.   And so the application of this technology is is  still to come. Once that part happens and we're   starting to apply that technology and they're  monetizing it, then we'll start seeing some of   that monetization begin to shift towards the next  generation products and then we might see another   another super cycle form. So Thomas, you are still  very bullish on a bright future for the mag 7   for the hyperscalers and all of these stocks. You  don't think the story is anywhere close to over or   there's too much risk built in to the future for  these companies? This cycle is not over. I think   the tech stocks have a long bright future ahead  of them, especially the blue chip operators. As   we see with AI, there's going to be changes in the  technology, but they're all very well positioned   to roll with the punches and to drive that  transition. They definitely have the money and   the scale and the scope to make it all happen.  All right, Thomas, thanks for this discussion   today. I'd love to hear your thoughts, too. Are  you concerned about all of the money going into   this AI buildout? Do you think it'll pay off  long term? Do you think these cycles will just   continue to cycle through new technology and new  advancements? Let me know your thoughts and join   the conversation in the comments. That's what we  love about this market beat YouTube community. And   if you want to hear some other thoughts about the  memory sector in particular and how much longer   the Micron and uh Seagates of this world could  continue to see some tremendous growth stories,   too. Make sure to watch this video with Louis  Navalier from Growth Investor. He has a lot to say   on how much longer the growth cycle could continue  in these names. You can watch that full interview