The News Today On NVIDIA Stock, Micron Stock, Anthropic, Tech Sell-Off - NVDA Update

I made the point that if you looked at the AI trade and and how how profitable it was until the last month or whatever, you could expect some type of I don't know if it's a 50% pullback. Are we there yet to a 50% pullback on the overall trade, would you say? Because it went it was out of control. Yeah. As far as as are are we near the beginning or the end of this anx? >> I I think we're at the bottom. You know, there's a well a phrase a lot of hedge fund managers tell me which is bottoms are dangerous because you don't know how much more downside you have at the bottom. But I think we are at the bottom because number one um the the market for AI is still very tight. It's not an over capacity market and the constraint we're seeing is preventing more spending or political opposition preventing more spending. that is slowing the growth, but it's not creating a top. I think the second sign is the crypto exploit. You know, the $6.3 billion loss from Binance, it's barely caused a decline in crypto prices because if crypto is a leading indicator for risk, you've already seen people sort of washed out. It's barely hurting crypto prices. >> But the the NASDAQ 100 is down 9 and a half%. That that makes sense. The S&P is only down by 3 and a3%. You think that's the bottom? It's not going lower. >> Yeah, I uh in some ways I think that this correction that we thought would happen, you know, uh is already played out in the NASDAQ and the equal weight is now climbing back to new highs. So I think we had a wash out in the equal weight earlier this year, an 8% draw down. Now the NASDAQ is on its way to a 10% draw down. So you've had most of the market correct actually >> that Leslie Picker has been looking through the amended filing as well. Morning Leslie. >> Hey Carl. Good morning. Um yeah, as you mentioned they kind of uh lay this out basically in the nine months ended September 30th. So in order to get that Q3 number and see what the most recent numbers are, you have to do a little bit of math here. So kind of eyeballing it. Uh so forgive me because we're working with some breaking news here. that it appears that the revenue came in for Q3 around 17 billion. Um they say that's up 1,127% year-over-year. Uh which kind of matches the growth rates we've seen on a year-over-year basis. Um in terms of adjusted operating income, pretty sizable margin there. Um that is uh you know something that we've been looking at. Uh sorry I lost my page there. There we go. uh about $2.5 billion dollars, decent margin. Um it looks like those margins are increasing from the first half of the year. Um an adjusted net income of about $2.9 billion. Um another thing that I was kind of looking for with regard to this amended S1 um is the amount of commitments that they have with regard to future spend. That's important because uh the cash load of about $131 billion is very sizable, but they do have commitments to spend on capex in the future. And so um the updated amount for that is about $900 billion for compute and infrastructure. Of course that is um you know over the course of anywhere from 1 to 20 years. Some of that is cancelellable, some of that is non-cancellable, but again, a sizable amount uh that they are committing to spend on capbacks, which kind of helps put into context what we're dealing with with regard to the cash uh for for this company, guys. >> Um the third quarter number we already kind of knew, right? 16.8 billion. So we we don't know October number, correct? >> No. And I I wouldn't expect that we would know October numbers before the S1 unless it's something that is discussed in kind of one-on-one sales meetings between the bankers and the prospective investors who are trying to get a sense of what the future might look like or um you know what they've seen. But it's unlikely that we get essentially audited numbers for October before this company goes public unless it does postpone or kind of go on a more stretched timeline for it for its S1. That would give them time. Oh, so there's not an expectation we ever get that number filled in is what you're saying. I thought that was I thought we might have expected that given October's over. They won't fill it in. >> Yeah. No, October is over. I haven't heard that we would expect an October number in the filing per se, but it could be something that's discussed with investors. >> Okay. Okay. Um, right. So, you're going through like we are trying to figure out what what changed. So that $900 billion number or 895 billion in compute commitments is is it higher than the number we got when we saw the original S1? >> It is. >> Okay. >> Yeah. The original S1 had that number around $820 billion. Some of that is leases. Some of that is capital expenditures that they're expecting. Some of that is cancelellable. Some of it's non-cancellable. Some of it would be spread out over the course of a year. Some of it up to 20 years. So, it's, you know, we're still kind of going through uh hopefully we get some sort of timetable that that explains what the commitments look like on a year-by-year basis, but um that's important just in terms of their future expenditures because right now they're, you know, >> operating income looks pretty good relative to to revenue. Uh but the question is kind of what that looks like over time as those expenses start to really get recognized um and start to hit both the balance sheet and the income statement. >> Yeah, I can't stress enough guys, you know, again that October number maybe we don't get it. I I don't know um that how important that will be and that's part of the narrative of overall demand >> and fueling so much that's going on. So >> obviously not getting it this morning. It it's still not clear whether you know we will or won't before the IPO. Leslie obviously saying we may never see it until until they go public. >> Look, I would have felt far more comfortable. We got a lot of these stocks going up. I kind of needed that number, but I also feel like Michael Earlston uh when when he speaks to it momentum, come on. It's not today. >> I want to point you guys to page 138 in this amended filing here. adjusted operating income did increase for those three months we're talking about for the quarter going from two and a half billion uh to compare that to 2.2 billion. But the important thing here is margins. So they talk about operating efficiency that drove margin expansion. This chart got a ton of interest and buzz I would say. You know last week when the S1 came out looking at their operating efficiency it looked like that adjusted operating margin fell. It was 19% down to 15% in this current quarter. And then it also looks like here there is a compute and infrastructure expense in the quarter that did rise. It was about 5.5 billion. It's now closer to 10 billion at least in the current quarter that we're talking about. I guess this was the third quarter I should say. And so higher compute, higher infrastructure expenses. And we talk about what rates are doing to the cost of the buildout. some of the supply chain here um anthropic paying more for that or also as part of this has just tried to get access to more compute. So margins still you know double digits and they look profitable on a adjusted basis but it got a little tighter in the quarter. So I think that's something that investors will be watching. We'll uh keep digging through this guys >> and I'm going to pull over the king daddy of AI and that's Nvidia. I spent my Sunday, Saturday and Sunday, uh, training for my December high rocks like usual, but I spent a bunch of hours in the library. I love going to the library. You could really lock in and get in that amazing flow state. And I am like, ah, you know, let's go to Yo Finance. Let's check out the valuation on Nvidia after Friday's route. And I I couldn't believe it. I could have I kept hitting refresh on the page to make sure I had it right. Nvidia's PE ratio 12.9* 4 earnings. Like I just mentioned the broader market S&P 500 19 times. Oracle I JUST MENTIONED IS 15.3 TIMES. YOU have Nvidia trading on a 4p of 12.9 times. But it gets even better or should I say almost uh more unbelievable even though these are the numbers as they are. Nvidia stock trades on a pouchy. 27 price toearnings growth ratio better known as the PEG ratio. Now, when I was studying for my CFA level one, the bottom line was if you were trading at a PEG ratio of one, that kind of tended to be fair value. We have Nvidia, the growth juggernaut in the AI boom now trading at a PEG ratio under 3/10. Are you kidding? That is absolutely mind-blowing. You can see Nvidia stock is becoming under pressure. There are a couple concerns here with Nvidia. First, of course, all the free cash flow concerns the hyperscalers and if they are forced to pull back because investors are pissed off and they are forced to pull back on capex next year in 2027 that could impact the order flow at Nvidia. So, there's that. That's why you're seeing Nvidia stock trading at valuations I've never seen Nvidia trade on before. The other concern is that Nvidia is at the center. They are at the center of this AI boom with the circular financing, giving money out to this company, investing in this company, giving more money out to this company, investing in this other AI company. If the AI trade slows down, if that slows down, it could be bad news for Nvidia. Okay, it could be bad news for Nvidia. I'm not saying they're going to have to raise cash or anything, but it could be bad news for Nvidia. It might lead them to purchasing less stock under their big stock buyback program that they announced a couple weeks ago. So, there's that. But again, I am I've been very surprised by this Nvidia valuation. Now, they have a chance to correct this valuation when the company reports in a few weeks. You want to hear Jensen W come out there strong. You want to see a good strong earnings beat. who want to see solid guidance and you also want to get a sense of how much stock Nvidia has been buying or will look to buy with the stock off its eyes. >> So when you see the others go by the wayside who can't raise capital at all. Okay, >> that obviously informs valuation of all of them. So we see what Musk said about Groank about its unsustainably high cost. So I think the AI trade is uh is definitely on pause and this is very typical at the peak of a cycle. I think we saw the peak. >> He's channeling Ben Writes very much of Melius today who has an interesting note. AI, this is quote, AI trades facing its toughest moment. Funding capex is everything. And he's cut price targets on a number of names. Nvidia, Marll, AMD, Dell, Broadcom. Okay, he cuts the targets there. And he says, quote, "We have at least 5 trillion dollar plus companies, soon to be seven, obviously anthropic and and open AI that have to issue hundreds of billions in securities, stocks, bonds, etc. The amount of paper so large it has become the primary risk to the AI trade more than power, chip supply, memory, and even political bottlenecks. Multiples could remain under pressure as the buildout's going to feature more back stops throughout 2028. How about that? That this is the game changer that Weiss identifies. That's what's fundamentally changed. I agree with Steve that financing is the challenge right now in the markets and financing is the reason that we're going back and forth with a negative tone here. I mean these numbers are pretty outstanding when we think about SpaceX a week and a half ago $40 billion of debt. Microsoft a week ago 25 billion. Last week I think it was last week 37.5 billion from Meta. Now we've got Oracle coming out with 20 billion. These are real numbers that when you add them up the money has to come from somewhere. Now, Scott, before we were talking about financing, you were talking about deleveraging and positioning. That's actually a positive. We've been doing that for weeks. But what I take away is the biggest positive of all of this, which will swamp the funding, but I'll come back to that in a second, is the capex plans. Last week, what we saw from Alphabet and Amazon was that they're not backing off on the capex. We saw a slight decrease from Microsoft, basically based on supply chain. It was very slight. I'm not going to worry about that. The capex plans are still very much intact. And for those three companies that I mentioned, the capital markets are open. Yes, pricing is a little bit more difficult and even for Oracle, the capital markets are open, albeit with a little bit more difficulty in the cost of capital. But the point being is that the biggest spenders out there still have access. I think the key tell here and what will keep the market going back and forth until this market clearing event happens is the Anthropic IPO. One quick note on that. I don't think Anthropic or its bankers would have brought that unless their soft road show. And Scott, I think you were out there in San Francisco when they did that soft road show was pretty wellreceived. And we got reports and so did you that it was very wellreceived. >> Obviously demand for it is going to be great. >> You say obvious. I'm not sure it's I think there's pretty obvious that demand is going to be great. The question is it two trillion? Is it one and a half trillion? whatever it whatever it is demand is is going to be great. I think we can make an educated statement agreed >> and and suggest that. But I I wonder if it's time to watch what the companies do and not what they say. Of course, everybody's going to stick to their spending plans. Anthropic, as we learned today from the amended S1, has spending commitments based 20 years out. That's not worth the paper that it's written on, is it? We'll see. Got good news for We'll see what's going to happen over the course of time. >> So, let's see what really comes to fruition in terms of the hyperscaler. >> They're raising money >> plans. They're raising money not just to sit on it and invest it in T bills. They're raising money with a purpose. Maybe the direction of yields matters most. >> Well, >> that's why Melius, by the way, cuts these targets. Nvidia from 420 to 315. That's a pretty sizable >> Well, Nvidia is financing AR. I mean, Nvidia, you know, there's no way they can live up to all those finances. >> I know, but these aren't like chump cuts here, right? Marll goes from 445 to 300. AMD from 660 to 600. >> Dell from 735 to 590. Broadcom from 610 to 450. >> 330 would be on Nvidia one heck of a price rise from here. Maybe the 420 was just a little bit too far out there. As an analyst, maybe you're looking at that and saying, "Yeah, >> I gotta bring this into something that's still aggressive, but not not unreasonable." >> But it matches sort of Weiss's sort of temperament around the whole trade. You've taken the other side by and large of that >> in terms of financing will continue in terms of I am optimistic. I am look I don't two things I don't think last week Amazon would have come out and I forget the exact number I want to say 195 billion uh for next year I don't think Alphabet would have come out with those numbers uh unless they believe them and and again you know besides that you got SpaceX rating raising 40 billion like these these companies are raising money to do something with it and if you're Nvidia you are a beneficiary with Broadcom same with AMD Bren Yep. So, first of all, let's talk about Mel as an analyst. I've been in this business since the early 90s. Analysts follow price, not the other way around. And these prices have come down. And so, the analysts are following suit as the prices were going to go up. They're like, "Okay, well, yeah, another 20%, another 20% higher." So, they're they're just reading the room of the market. I think as as long as as the hyperscaler goes specifically with Google, Amazon and Microsoft, you know, Meta Meta and Oracle I think are are different from that. Microsoft is the only one that's reading the room. And so I think that this debt issuance does have a a rubber band it's pushing on because while technically theoretically the debt markets are open for these companies, I'm not sure just the math equation of how much the debt market can absorb at a time. And so these are just historic numbers that these pristine balance sheet companies are issuing. But the corporate debt market can only absorb so much at a time. And so rightfully so, yields are going higher. I think for these companies wanting to issue debt. I think I'm right in the middle of Stephen Steve and Jim. I think that we're in this settling point. The midterms are a clearing event. They always have been. They're a clearing event. I will say from like 1942 to last midterms, 21 out of 20 times, 21 out of 21 times, the S&P is positive and the average return is around 16%. So maybe this is the first time it's negative, but if I'm going to Vegas, say I bet on red or black, and that red is that that statistic, I'm going to put it all on red. So I would caution people to get too bearish. The last thing I'll say is last month was a clear signal. The rotation out of tech is in in earnest. Nine of the 11 sectors were up, four were up double digits. You would think you would get follow through this month because there has been technical damage to the tech names. So I think like look at the companies individually sell calls don't don't run away from tech because I do I still think these companies will deliver at large but you do need to see the NVIDIA the Dells the the DRAMM companies start moving higher because they're still getting me all the money and that to me is what's most concerning is where the money's going. Those stocks are not are also not doing well. Micron arguably is the cheapest stock that ever has existed ever in the history of in the history of stock markets got hammered went from 1100 to 8 860 rallied to 980 back to 860 can't get a can't what I mean you just say forget about this and own it this is a gift here in your mind >> it's an absolute it's Santa Claus coming early you're talking about a company that's got 130% annual revenue growth 110% annual earnings growth trading at a 5p E. Now, if you think this whole but we're in the middle of a bubble of AI, then don't own it. But if you finally want to own it, chief, here's your gift from Santa. >> All right. >> Um, when do you think the market will wrap its head around, you know, this kind of idea of normalization? >> Yeah, I think the question you're asking is how low do multiples need to go to reflect what's beginning to be deceleration and earnings growth rate in the growth rate in data center capex spending, etc. Um, I think about a company like Nvidia. Nvidia trades at about 14 or 15 times next 12 months earnings. Now that has been cut, you know, almost in half over the last 18 months. Nvidia's, you know, stock is trading as though um data center capex is going to peak in 2027 probably. And uh my fundamental outlook is better than that. I think it'll I think growth will continue in data center construction spending, you know, beyond the end of 2027. But yeah, investors are rapidly pricing in decelerating growth and and perhaps an eventual peak. How much is enough? I think that's a question that we we really can't answer for certain. Um I think what we should start to see are companies allocating capital to share repurchases in a signal and expression of their belief that the stock is attractively valued. We're actually seeing Nvidia do that. Nvidia said that they will spend half of their free cash flow on share purchases and dividends going forward, but most of the other big tech companies aren't in a position to do that these days because they're spending all of their free cash flow building data centers. So, we're actually seeing equity issuance, not repurchases from the likes of Meta, Oracle this morning, Google a few months ago, and uh probably others to come. And so, Jed, if if share repurchases are one example of how we can distinguish between companies that might be making more productive AI investments, what other metrics are you looking at or what other signals are you watching to determine if these massive investments these companies are making, many of them bringing themselves a free cash flow negative in doing so, will actually be productive investments in the long run? or if there's potentially overbuilding capacity. >> Yeah, great question. I think the the key metrics for us are number one uh demonstrating accelerating revenue growth. Um we're seeing that for example at Amazon's AWS unit where revenue growth accelerated for the sixth consecutive quarter and management provided guidance that they expect that acceleration toi continue into 2027. That's an important sign of the ROI on data center capex. Another one are rental prices for the GPUs themselves. Uh those rental prices have been strong. They've been increasing throughout the year. That's impressive. That shows you demand exceeds supply. Even as we're getting new Nvidia chips, you know, over the last couple months, rental prices have been very high. So that's that's another another important one. And then I think a third one, maybe less quantifiable, but but equally important is the pace of new model development from the the big labs, Anthropic and OpenAI and Google. But then also um product you know turning those models into products and MetaMuse is the one that comes to my mind first there um you know cursor and the coding assistants are another but if if if the tech industry is is churning out model improvements if they're churning out new exciting products that people are using every day that is going to drive GPU usage it's going to drive the cloud company's revenue growth and the lab's revenue growth and it's going to drive new software innovation and growth there. So those are some of the key leading indicators for us. >> Uh in Nvidia is a strategic partner in both of you. Um have you talked to him? How Michael has he received this news this merger? >> Yeah, Matt talked to Jensen really early on in the process and and explained what we were thinking about and I think the reaction from Nvidia has been super positive. In fact, we're going to come off this call and have another discussion with them really trying to explain the merits of this combination. And you know, as you correctly said in your comments, Nvidia's invested in Marll, Nvidia's invested in momentum. We think we can actually bring a lot more product to bear to Nvidia a lot more quickly in this combination. The NASDAQ 100 fell again today and it's now nearly 11% off its early October highs. Big question is is the worst tech behind us or is it safe to start shopping for deals at this point or look out below? And that's why tonight I want to return to a rubric that I put forward two weeks ago reviewing our progress against the four tangible hurdles that I gave you that I'd like to see cleared before I can feel better about tech. The first thing that I said was that we wanted to see an improvement in the technicals. That's the charts. They got to get better. And that simply hasn't happened at all. If you're looking at the charts of the broader indices, the NASDAQ composite, NASDAQ 100, even key tech indices, subsector indices like the Philadelphia semiconductor index, the socks, you see the same thing, an ugly double top. And you know what? A double top almost always leads to lower prices. Worse, we're starting to see even more negative signs in charts of some individual stocks. Specifically, we've seen a number of important stocks where the 50-day moving average has crossed below the 200 day moving average and an event that's dramatically known as the death cross. Now, I have never told you to sell a stock just because you've seen this chart, but I do want to bring it out that we're seeing it in a Broadcom, G Vernova, Alphabet, and and this is very telling, okay, very telling because there are a lot of people who live and die by the charts. Also, when it comes to technicals for the market as a whole, the last thing that I'll note is that the market is not in the aggregate oversold even as some tech stocks have been slaughtered. The indicator that I use, the S&P short range oscillator from Alpha Market Edge, showed a reading of 1.1% at Friday's close. That's basically neutral. Actually, more a little more overbought than oversold. This is because this hasn't been a true marketwide sell-off. Instead, it's been more of a rotation with the pain more or less isolated to tech stocks as we saw once again today. As I told you at the top of the show, a lot of other stocks are acting sensibly in relation to the bond market. Next, the second thing I said we needed to see was the hyperscaler earnings reports. want to see how they're doing. Simply I said I wanted to see no degradation the hyperscalers huge tech spending plans. Now we have indeed gotten most hyperscalers earnings reports at this point but for the most part they've gone the wrong way. That is if you're bullish with tech. The good news is that most of the large hyperscalers are not dialing back to capital spending agendas. That gives us some hope that the AI infrastructure theme is not in fact dead. In fact, but as we explained on Friday night, the bad news is that the only hyperscalers that did dial back its capex plans, Microsoft saw the best reaction from the market for doing so. Didn't go up, of course, but it went down the least. And I'll take that in this lousy market. Just today, we saw shares of Oracle get crushed. Oh my god, this was horrible. Another obliteration down 9% uh around 9% if the company launched a new 20 billion fundraising effort. 10 billion in equity and 10 billion in bonds. Tell pay for its data center spending, which seems endless. Oracle indicated last week that this could be coming sometime in the next three and a half years, but the market clearly didn't expect it to come in the next 3 days. THEY KNOW NOTHING. >> CLEARLY, THE market has taken a new stance that's that's skeptical of major capital spending plans, especially if a company has to issue stock or new debt to pay for them. And that is something that we'll need to overcome somehow if we want to text tech stocks to rally. Oracle was really horrible today. It really just caught a lot of people by surprise. Stock was up big on Friday. The third thing that I was looking for was simply a resolution to the midterm elections that we should get by Wednesday. While the consensus seems to be that a major success for the Democrats is a negative dell for tech and the data center trade. I'm not sure it's as simple as that. First, the chance of Democrats making gains in Congress has already likely been factored in at this point because the polls and prediction markets have been indicating this outcome for several weeks now. Frankly, I think it's just getting past this election. I think that's going to be the positive, a positive development even for tech stocks because politicians who are opposed to data center will at least start quieting doubt about the issue. Could we see one more leg lower with a big blue wave outcome? Maybe. and and we could also see more pain if the mid midterms are followed by an escalation the war in Iran as many suspect although it wasn't clear from the news today. I think we're near the end of the midterm election overhang. Uh I say by Thursday by Thursday should look clear. Finally I say that we need to see a successful anthropic IPO and while that's also unresolved there's been some nice progress in this front. This is the best one of the four. Last Monday the company released the S1 filing which contained the IPO perspectives. Then today, Anthropic released the updated S1 with even more detail, including third quarter financial results. Those new numbers showed that revenue growth is still very strong, but slower. Anthropic's quarter- quarter revenue growth seems to have peaked in the second quarter at 144% and it fell to 47% in the third quarter. On the other hand, the company's profitability, at least in adjusted basis, has been improving throughout this year. We did not, as I was hoping for, get any details on October numbers. There'll be plenty of time to cover more specifics from the anthropic perspectives. Don't worry, we're going to cover it wall to wall. For now, what you need to know is that the deal is on track. If all goes as planned, road shows should begin next week, and the IPO should price and begin trading in the third week of November. And really, we don't care how well or poorly anthropic shares do after they price. We just want to see that there's enough demand for the IPO to happen. In less consequential news, we learned that this weekend that anthropic rival OpenAI is lowering its ambitions for a new private fundraising round. After previously seeking to raise $30 billion at a 1.4 trillion valuation, it's now looking to find 20 billion and it's cut its target valuation to 1.25 trillion. Still pretty great. Some people thought it's a bad sign when it comes to investors demand for AI labs at this point, but Anthropic is the much more important test for this market. Here's the bottom line. Couple weeks ago, I put forward four things I'd like to see happen before I could feel more positive about tech stocks. None have gone our way thus far. So, it's not surprising that the sector has remained under pressure. But if you're a tech optimist, the good news is we're approaching some potential turning points and the anthropic IPO keeps marching along. That's good. We need that deal to happen if we want tech to get back on track. Without it, expect many more days like the last few where the most important tech stocks just cannot hold their gains. All right, I hope you're all doing well today and staying calm in this market. Monday was a mixed day in a market and a decently red day for many tech stocks due to multiple news stories and events. Reports that OpenAI has slashed its planned funding, rumors online about a lot of people trying to sell compute at below market prices, Anthropics amended S1 filing, Oracle raising more capital, and market participants unsatisfied with the terms and pricing of Marll and Luminum's newlyannounced merger. All of those news stories weighed on tech stocks Monday ahead of Tuesday's election and a potential anthropic IPO on horizon. Let me address each of those. First up, the FT published a story Friday night saying OpenAI had reduced its fundraising target from $30 billion to approximately $20 billion while lowering its targeted valuation from $1.4 trillion to $1.25 trillion. That story set the stage heading into Monday. And then over the weekend, there was a lot of talk online from multiple people from various startups saying they were getting unsolicited offers to rent compute well below market prices. That spooked some market participants as they began worrying that we may be nearing a potential glut in supply. funnily enough, I noticed many founders stepped in and replied to those posts, saying that they would gladly take very large amounts of that capacity if it was actually available at the rumored prices. I thought Nick Kami had an interesting take on the matter, pointing out that not everyone who says they have available compute actually has available compute. Put simply, it's one thing for someone to claim that they have compute available to be rented at attractive prices. But actually having compute that is operational with all the necessary land power shell networking and other components, that is a very different situation. I've been hearing a lot recently about some opportunistic individuals supposedly offering compute at below market prices, but in reality, they're looking to secure offtake commitments that they can then bring before potential lenders in the hopes of raising capital that would allow them to get started in the business of renting out compute. So, put simply, if someone is offering compute well below market prices, I would be very careful about assuming that they actually have operational compute available for rent. Additionally, we need to be clear in saying that there will be some companies who purchase or rent a lot of capacity, but then realize that they don't need all of that capacity for their specific workloads. In that situation, those companies may try to offload some of their capacity. But that does not mean that the world is not compute constrained. It just means that that specific company took on more capacity than what they actually need for their own specific workloads. The rumors about cheap compute over the weekend put pressure on many AI infrastructure stocks Monday. Also on Monday, Anthropic filed an amended S1 with the SEC, revealing new details that were not present in the company's original S1 filing. I'm not going to cover everything, but I'll mention some of the important points that likely had an impact on AI stocks. Monday, Anthropic revealed Q3 revenue of roughly $16.95 billion. A decent quarter-over-quarter increase in compute and infrastructure expense, going from roughly $5.55 billion in Q2 to $9.86 billion in Q3. Q3 operating loss was $567 million. and Q3 net loss was roughly $23.3 billion. Once again, Anthropic's large net loss relative to the company's revenue got a lot of attention from the usual bears. For the 9 months ended September 30th, 2026, net loss was roughly $14 billion. As was the case when we got the original S1, the vast majority of that loss is from the change in fair value of convertible notes, derivative liabilities, and safes. That amount was roughly $94.3 billion for the first nine months of the year. And it is responsible for the vast majority of Anthropic's $ 104 billion net loss. Again, that is mainly due to Anthropic's valuation increasing and the subsequent change in fair value of convertible notes and other instruments. That change in fair value is recorded as an accounting expense. So yes, as crazy as it may sound, the vast majority of Anthropic's net loss is due to the company's valuation rising and the subsequent increase in the value of convertible notes and other financial instruments. That is very different from the company burning through $14 billion in the typical operation of its business. Now, something that spooked some market participants is that there do appear to be some short-term quarter over-carter negatives in these Q3 results. So, for example, while revenue grew more than 46% quarter-over-arter, operating expenses grew at a faster rate, rising more than 65% quarter-over-arter. Another point that spooked some market participants was that adjusted operating margin was 15% in Q3 versus Q2's 19%. Now, both of those points I just mentioned about operating expenses growing faster than revenue and adjusted operating margin being lower quarter- quarter. It's important to consider those details within the proper context. As you can see from this chart, over the longer term, Anthropic has made notable progress going from an adjusted operating margin of -21% in Q1 2025 to positive 15% in Q3 2026. And while quarterly revenue grew at a slower rate than operating expense in Q3, the long-term trend shows a different story. If you compare the first nine months of 2026 to the first nine months of 2025, revenue was more than 13x, while operating loss narrowed from $5.2 billion in the first nine months of 2025 to 1.26 billion in the first 9 months of 2026. When considering the sequential decline in adjusted operating margin, it's also important to keep in mind that Anthropic spent $5 billion on unusually expensive short-term compute in Q3. That put a damper on adjusted operating margin. Another short-term negative that spooked market participants was that revenue per dollar of compute in Q3 was $1.72 versus $28 in Q2. Again, this is another short-term negative quarter overquarter that is overall a long-term positive. Anthropic generated $1.72 for every dollar of compute and infrastructure expense recognized in Q3. Let's not miss the forest for the trees here. We should expect that there will be lumpiness in this company's results given the nature of this industry. For example, there's been a lot of talk online about revenue reacelerating in October on the success of Opus 5.5. As you can see in this graphic from Open Router, Anthropics token volume surged in October. I know comparing token volume to revenue is like comparing apples to oranges, but still there's clearly an acceleration in usage that took place in October, and that's not reflected in the Q3 results we got on Monday. Given the lumpy nature of this business, we're going to see fluctuations in various metrics in any given quarter. The lumpiness that is reflected in the company's Q3 results appears to have spooked some market participants. But it's important to keep the long-term trend in view and not get too caught up in the sequential results of any given quarter. I also want to mention Anthropic's cash position and spending commitments. As of September 30th, aggregate committed spend was approximately $895.4 billion. That also spooked the market slightly because it appears to be an increase from the amount disclosed in the original S1. Importantly, roughly $447.5 billion of that amount. So about half of the total amount is expected to be spent before the end of 2030. Substantially all of the remainder. So the other half of the total amount is expected to be incurred over the course of 15 years from 2031 to 2045. It's not like Anthropic has to spend all of that money today. Not even close. Also, at the end of September, Anthropic had $95.6 billion of cash and cash equivalents, $36.3 billion of short-term investments, and 14.8 billion of availability under a credit facility. They also have tens of billions of dollars worth of funding support from their strategic partners. And that's not even including the capital that they're preparing to raise. As I've been saying for a while now, I expect the frontier model companies will be able to fund their compute purchases for at least the next two to three years. So, I can understand that some market participants got spooked by the apparent lumpiness reflected in the Q3 results. Still, the long-term improvement in the company's financials is very impressive. I would not read the sequential deceleration in Q3 revenue growth or the decline in quarterly adjusted operating margin as a break of the overall longerterm trend. This is a lumpy business. Let's not miss the forest for the trees here. Something else that weighed on many AI stocks Monday was the announced merger between Marll and Lamentum. From market participants perspective, there are two parts to this news. On one hand is the technological and strategic business perspective. And on the other hand, there is the terms of the merger such as pricing and timing. The price action we saw in momentum shares is largely due to the second part, the terms of the merger rather than the technological or strategic significance of the deal. It's important to make the distinction between the two. Strategically speaking, this merger makes a lot of sense for both companies. It will allow them to better compete with companies like Broadcom, for example. And it should also be positive for Nvidia because the combined company should be able to provide more product to Nvidia. But now, let's cover some of the important terms of the deal because this is what the market took issue with. First, this is an all stock transaction, not a cash transaction that disappointed many Luminum shareholders. Now, since it is an all stock deal, some of the numbers I'm about to share with you have changed following Monday's price action in the company stocks. So, keep that in mind. As per the press release that was published Monday morning, the deal had an implied enterprise value of $115 billion for Momentum, representing approximately $1,132 per share of Luminum common stock and a premium of approximately 23.9% based on each company's unaffected closing share price as of October 28th. Again, that has changed after Monday's price action. Also, according to the press release, postc closing, current Marll shareholders will own approximately 67% of the combined company on a fully diluted basis, while Luminum shareholders will own approximately 33%. Under the terms of the agreement, Luminum shareholders will receive a fixed exchange ratio of 4.5 shares of Marll common stock for each share of Luminum common stock they own. And this next point is very important for shareholders. The transaction is expected to close in the first half of calendar 2028 subject to approval by Marll and Lmentum stockholders, regulatory approvals, and other customary closing conditions. Again, I think this deal makes a lot of sense from a technological and strategic business perspective. However, the terms of the deal are what disappointed market participants. Many Luminum shareholders voiced their opinions online that Lummenum is selling itself at what they believe to be a discount given Luminum CEO's recent comments about optical components being sold out through early 2029. Many people are wondering why Luminum would sell the company at such a small premium given the strong multi-year visibility. So, you can imagine why this news weighed on Lummen stock and by extension other AI stocks. It made people question if something is wrong beneath the surface. The value of the deal seemed odd given the CEO's recent positive comments about strong multi-year visibility. Many market participants were disappointed by the value of the deal. Now, I should mention that the press release says that the deal is expected to be accretive to Marll non-GAAP earnings per share in year 1 post close. That said, from a stock perspective, given the fact that the deal doesn't close until 2028, it likely creates a cloud or overhang over the stocks of both Lamentum and Marll until the deal closes in 2028. For Lmentum specifically, it puts a lid on the stocks stand alone upside because now even if LMU reports very strong results, investors are still going to value the company from the perspective of the 4.5 Marll shares that they're to receive if and when the deal closes in 2028. Also, if there's any regulatory uncertainty over the next 14 to 20 months before the deal closes, that could also be a negative for the stock's short-term price action. You're probably also going to see a lot of arbitrage between the two stocks leading to additional volatility at Monday's closing price. And using the 4.5 share ratio announced in the press release, there's about 12.7% of upside between one share of Lentum stock and 4.5 shares of Marll stock. That's not much of a premium. So the deal makes sense technologically and from a strategic business perspective. But many shareholders, especially Luminum shareholders, are not happy with the pricing and the timing of the deal. I mention all of this because this news and the subsequent price action in shares of Marll and Lmentum put additional pressure on many AI stocks Monday. Something else weighing on tech stocks Monday was news that Oracle filed to sell $10 billion worth of stock and another $10 billion worth of bonds to fund its buildout. All of those news stories about OpenAI slashing its fundraising target, rumors about cheap compute, Anthropics amended S1, the terms of the Marvel merger, and news of Oracle issuing more shares and debt. All of that weighed on AI stocks during a time of heightened volatility right before an important election as market participants are worrying about an escalation in the Middle East shortly after the election. So, that's what happened to tech stocks on Monday. I'll also briefly mention that Monday evening on the Apple earnings call, leadership spoke about the memory shortage, saying, quote, "What we're seeing with memory pricing right now is unprecedented." and quote for the December quarter, we expect even higher memory costs. As I've said many times, while there are similarities between this current moment and the memory cycles of the past, there are also major fundamental differences. I expect this moment to last longer than the typical memory cycles of the past. As I said late last week, with an election this week and a potential anthropic IPO on the horizon, we may experience some choppiness in the market over the next 2 to 3 weeks. Sentiment shifts and turns frequently, but eventually it always comes back to fundamentals. Focus on first principles. Focus on the fundamentals and do not allow short-term volatility to cause you to lose sight of the long-term fundamentals. As I often say, remember to stay calm in this market. Remember to maintain a long-term perspective and do not make any hasty or irrational decisions. Looking ahead, we have Nvidia earnings on November 17th. Now, in case you're new to the channel, I want to make sure that you have at least a basic understanding of the underlying long-term thesis. So, let's cover that. Now, I don't know what's going to happen in the short term, but from a long-term perspective, I am very confident that Nvidia will be worth much more in future years than it is today. When Jensen was on the Lex Freedman podcast not that long ago, he was very seriously raising the possibility of Nvidia becoming a $3 trillion revenue company in the near future. If that happens in the coming years, then it is very plausible that Nvidia could one day be worth tens of trillions of dollars in market cap. That might sound crazy, but that's what Jensen is implying when he raises the possibility of Nvidia becoming a $3 trillion revenue company. I guess the question at that point is what multiple the street will be willing to give Nvidia. I don't know the answer to that question, but I truly do think that Nvidia will be worth much more in future years than it is today based purely on the fundamental growth of the business. Based on everything I'm seeing, the world is still computed and I expect that to continue at least through calendar 2028. In a computed environment, developers will use whatever viable compute they can get their hands on. Today, there are no GPUs that are sitting dark due to a lack of demand. Like there was fiber sitting dark due to a lack of demand at the height of the dotcom bubble. Back then, companies were laying fiber in the hopes that use cases and demand would eventually show up. Today we are seeing the complete opposite. As I've said many times, when market participants compare this AI revolution to the dotcom bubble, they ignore the fact that the internet is already here this time. This means that mass adoption of the technology and new use case development at scale are immediately possible. We don't have to wait years for it to show up. It's already here. The world is compute constrained, which means there is not enough supply to satisfy demand. New capacity is utilized as soon as it comes online. The hyperscalers are monetizing capacity as soon as it comes online. Each of the hyperscalers spoke about being supply constrained on their most recent earnings calls. Additionally, many of the clouds are building out into contracted demand. They're not blindly building in the hopes that demand will eventually show up. No, they're building out because they have signed contracts and in some cases significant prepayments from their paying customers. This AI revolution is fundamentally different from the do-com bubble. And 2026 will be a pivotal year for the AI industry thanks to the rapid adoption of agentic AI and the proliferation of agentic systems in the world's leading enterprises. The leading AI labs revenues are surging right now. Agentic coding and the implementation of Agentic systems in large enterprises are new use cases that are increasing inference demand significantly. That subsequently is increasing compute demand. The rapid adoption of Agentic AI is why we're seeing an inflection in inference demand. It's why we're seeing the leading AI labs revenues surge. I wish both anthropic and open AI were public so the public could see the ramp in their revenues. I think the leading labs surging revenues may be the initial proof point that grabs market participants attention and causes them to realize that there will be a clear ROI on AI infrastructure. I think the leading labs surging revenues will also help assure investors of the longevity of Nvidia's growth since these labs revenues are directly tied to compute. If they had more compute, they would have greater revenues. It really is that simple. Demand is not the problem. The problem is a lack of supply to meet the demand. As I've said previously, I expect the world to be compute constrained at least through calendar 2028, possibly longer. And so regardless of what happens in the short term, it's important for long-term investors to remain focused on the fundamentals, maintain a long-term perspective, and remember that we are only in the early stages of aic systems being adopted at scale. This will increase compute demand significantly. And after that, the next surge in compute demand will likely be fueled by physical AI. We're no longer talking about digital agents performing digital tasks. With physical AI, we're talking about physical AI agents performing physical tasks in the real world. NVIDIA CFO has called physical AI, quote, a multi-t trillion dollar opportunity. and the next leg of growth for Nvidia. This industry will fundamentally transform society and Nvidia has positioned themselves to benefit massively. NVIDIA sells the hardware for the data centers where the models are trained. They offer omniverse where the models are taught and tested. And Nvidia also sells the hardware that allows ondevice real-time inference through NVIDIA AGX allowing robots to have intelligent interactions with the real world even when they are not connected to a data center. Notice that Nvidia is taking a holistic platform approach to physical AI and they're embedding themselves as the underlying foundation supporting all of it. Over 3 million developers are already building on the Nvidia robotic stack and this is not getting enough attention. As for production ramps, Blackwell Ultra has ramped and remains in high demand. Vera Rubin is rolling out to customers. Nvidia Gro 3 LPX is in full production. Later on, we're expecting the launch of Reuben Ultra in 2027 and Fineman after that in 2028. We have a clear data center product roadmap stretching into 2028 and Jensen believes that AI infrastructure spending will reach three to 4 trillion annually by the end of the decade. That means Jensen is expecting growing AI demand and an expanding total addressable market underpinning all of this. I don't think we are anywhere near any type of bubble bursting type of event. With all of this in mind, I seriously think that Nvidia still has plenty of runway ahead of it. And I think this company will be worth substantially more in future years than it is today. At least that's my view of the situation. Quick note before I wrap up. All of the compilations on this channel are edited by Finn Vid with original structure and commentary. Occasionally the same edits appear elsewhere on YouTube. If you're looking for the original version, it's always here on this channel. Thanks for watching, Finnvid. I appreciate your support. Remember to stay calm in this market. Remember to maintain a long-term perspective and do not make any hasty or irrational decisions. With all of that being said, I hope you all have a great rest of the day and I'm curious to hear your thoughts about Nvidia in the comments below. Please leave a like on this video so more people will see it. And while you're down there, please consider subscribing. It's free and you can always change your mind. Thanks for watching and hopefully I'll see you in the next