9AM HOUR: Anthropic Unveils Public IPO Filing 10/26/26

0:00With the Discover Cashback Card, it's payback time when you earn cash back on everyday purchases. Activate and earn 5 % cash back at different categories each quarter on up to$1 ,500 in purchases. That's 5 % cash back at different places each quarter, like grocery stores, on gas, and at restaurants. It pays to Discover. Terms apply. See discover.com slash 5 for details. This episode is brought to you by Schwab Market Update, an original podcast from Charles Schwab. Schwab Market Update, an original podcast from Charles Schwab, provides an information-packed daily market preview on stock updates, U.S.

0:39and global economic news, monetary policy decisions, and key results and statistics that may impact your trading. Download the latest episode and subscribe at schwab.com slash market update podcast or find Schwab Market Update wherever you get your podcasts. Market insight and analysis. You're listening to the opening bell of CNBC's Squawk on the Street.

1:07Good Monday morning. Welcome to Squawk on the Street. I'm Carl Kington. Jim Cramer is back at Post 9 of the New York Stock Exchange along with me and David favor. Pivotal weeks on deck. PCE, Fed decision, hyperscaler earnings. And now Elon Musk's warning about the more vulnerable neoclouds. Core Weave approaching 51 pre-market. Brent's backing off a Friday's close at 117. That was about a six-month high. Our roadmap begins with Musk predicting a neocloud shakeout and confirming a$40 billion SpaceX debt deal for AI chips. Plus, Meta announcing an up to$95 billion stock offering. Now, some of it's happening near term.

1:43Much of it happening next year. We're going to tell you all about it. Of course, what's it for? Well, to fund its AI expansion. And we're also keeping an eye on Shares of Horizon, a company reporting an earnings beat for the third quarter. Also hiking its full year 2026 earnings guidance. Another busy week ahead for the markets, as we said, Jim. A lot sort of in the way of the bulls, whether that's the Fed or the midterms or this notion that AI spending is topping out of sorts. Right. Look, this is no doubt about it. This is a real thicket. This is trouble. The fact that oil was down for today gives a little impetus to the bulls.

2:22But the amount of money that needs to be raised at this particular moment is extraordinary. And I don't know whether we have the horses. I don't know, David, if there's enough to be able to handle what everybody needs. I think we're going to have to start to think about capital, access to it being a differentiator for the companies that are obviously in competition with each other in terms of the AI build out and their continued enormous appetite for compute. So just this morning, we just mentioned, we've got Meta going to issue$22.5 billion in common stock. That does come with a vote, by the way.

3:04Another$12.5 billion of a preferred. So they're going to raise$35 billion at Meta very quickly. And then they're also going to have an at-the-market offering of the new C-class of shares that don't have a vote that they're going to be issuing next year. That'll be at the market, so that won't necessarily be all at one time. but they're going to be able to sell as much as$60 billion of that depending on demand. So that's$95 billion. You've got$40 billion finally. It's a bit more structured, the deal from SpaceX in terms of a special purpose vehicle, an Apollo, the lessor, buying the chips, the GPUs.

3:42But basically, it's SpaceX credit. It's going to be on their balance sheet, I think, as a lease. But it's SpaceX credit. They go out on the market, and it's$40 billion that's going towards GPUs. And we're obviously waiting for others to come public or tell us about their plans to come public, such as an anthropic. And they're going to be hitting the market for money. So this and then you got those Musk comments, guys, which were fascinating over the weekend. Do you think self-serving or just kind of the way it is? Kind of both. I mean, that's that's Musk self-serving, but at the same time, also some truth to that.

4:17Right. Just talking about the fact that, and he says it here, you know, it's going to be all about getting access to capital and to debt markets and capital markets overall. And then he goes on to say, Carl, about the neoclouds that what he could see guppy level death and consolidation. He may not be talking about the biggest names. He may be talking about some of the smaller names. To your point, David, I think we are getting the S1 from Anthropic. Yeah, so there it is. We've got the S-1 being filed by Anthropic. Of course, one of the two key frontier labs that we've been following so closely that has been an enormous company when it comes to the private markets, but has plans to come public and probably will do so, let's call it, the next three to four weeks.

5:06We've got to see revs. We've got to see a lot of things in this S-1 now. By the way, I don't believe, and we're going to take a look at it as quickly as we can, that we're going to get a capital raise and how large it's going to actually be. In other words, are they coming for$100 billion? What's the number? And it's not clear, but I want to make sure we'll see because what really investors want to know is what's the latest quarter trending like? That last third quarter we got, as reported, at$16.8 billion in revenue. It was a deceleration of the revenue growth rate that has worried sort of markets generally.

5:39So, you know, will we get it or not? It's always amazing to me that people call and never know that we're on TV. Right. But, guys, you know, the risk factor is also going to be fascinating. And, again, we did get this draft, remember, reported from the Wall Street Journal last week as well. So a lot to look at from Anthropic in terms of the S-1 that we, again, have been waiting for for some time. And that obviously is indicative of the company's plan to become a public company, let's call it, in the next three weeks, three or four weeks, certainly we've all been hearing prior to Thanksgiving and prior to that week, Carl, was when they probably would want to get it done.

6:22What I think is interesting, Carl, last week, obviously a little out of pocket, but the one thing I was focused on was cybercrime. And we have this, and we're all talking once again about fundraise, and it's like we forgot something that made every bank chill, Every company chilled. All people chilled. And it's like, OK, on to the next, because there is such a plethora of news that it's a fire hose of news. I want people to know. I mean, here we are, like, looking at adoption of CLAW. What are we doing? It is just a moment in time where the things are happening way faster than any time I've ever seen.

7:04You're right about that. We definitely have a ticker symbol now and some lead underwriters. For more on the filing, let's get to Leslie Picker back at HQ. Morning, Leslie. Hey, Carl. Good morning. A historic day with Anthropic revealing its S1 filing. Obviously, this is expected to be a massive multi-trillion dollar IPO. In this first filing, we don't get an offering size, nor do we get a valuation. But we have hundreds of pages, about 460 pages to dig through here that can kind of give some sense into what we can expect from this company. as it goes public. As you mentioned, there are underwriters, Goldman Sachs, Morgan Stanley leading this deal, Morgan Stanley serving as stabilization agent for Anthropic.

7:50It also did that for SpaceX as well. As you mentioned, we have a ticker symbol. It is A-N-T. We knew they had plans to list on the NASDAQ, and that has been confirmed here in the S-1. Obviously, there are hundreds of pages to kind of dig through to get a sense of, you know, what this company is all about, including financials, risk factors, use of proceeds, which we obviously know, as David was talking about earlier, the use of proceeds to build out the AI infrastructure is tremendous for all of these companies. Anthropic, no exception here. So continuing to dig through, we will flag all of the nuggets as we find them.

8:32But a fascinating read so far, guys. Leslie, always. Did you get that? Sorry. Page 136. No, I haven't gotten there yet. A significant increase in revenue per dollar of compute. Maybe we're looking at, when we talk about deceleration, maybe we're looking at the wrong thing. Leslie, the revenue that they're getting per compute is extraordinary. Yeah, I'm still getting to that specific page, but you're right. And I was reading through page 62 as well there, Jim, where they kind of say they're in their early stage of revenue growth. This is part of the risk factors and that they have experienced rapid growth in recent periods.

9:10Their historical rates are not indicative of future performance, which is a typical risk factor that you would see. But it is tremendous as you look at the growth here, because on an absolute basis, and we've talked about this over time, you know, on a run rate of, say,$32 billion, if you take the first half of$16 billion for the rest of the year, They're only assuming that that growth rate, you know, is what transpires, which is not what the market is expecting. You know, that wouldn't necessarily be indicative of a two trillion IPO. But because the growth rate is so high, investors are underwriting a much higher valuation for this company.

9:46Well, they did 16 billion in the first half. They did 16.8 billion in the third quarter. So obviously it's right. Exactly. But, Leslie, as you know, the worry has been that actually the growth rate has been decelerating. I don't know, Jim, that we have or Leslie, that we have any numbers in here that show what things are looking like for the current quarter at this point. That's what it is. And that's, you know, that's going to be a key question. Leslie, these things get updated in the weeks ahead. Right. I mean, I think it'd be helpful to explain to people blanks are going to start to get filled in.

10:18Whether they are where things stand in terms of the financials, how much they're going to ultimately raise, what valuation they're going to target. All of that's going to get filled in with amended S-1 filings over the next few weeks, correct? So the rules are such that it's 15 days for this S-1 to stay public, to give people a chance to read through it. As we mentioned, it is a lengthy document. It is very complicated, and it's out there for people to read at their leisure, I recommend. And I would expect most everyone who is watching will be doing that over the coming weeks. After that 15 days has passed, then they are able to set terms for the offering.

10:54What that means is they'll have potentially a price range. SpaceX opted for a fixed price, but a typical IPO would choose a price range by which to assess demand for their shares. That you can kind of back out based on shares outstanding and some diluted shares to come up with a valuation target for this company, as well as an offering size for this company. So that roadshow process, you know, two weeks from now, if it were to launch on the most expedited timeline, that would take a week, a week and a half, basically, before they would set a final price for the IPO and then start trading the following day.

11:30Again, it is listed on the NASDAQ, presumably will trade in a variety of places under the ticker symbol ANT. Interesting chart we were just looking at, of course, and I think what Jim may have been referencing. They're obviously getting a lot more money, but they're also paying a lot more, too, which you've got to remember as well in terms of what they're paying for compute. Think of how much they're paying just to SpaceX, which we were just talking about. They're talking at SpaceX about getting$50 billion for every gigawatt of compute that they have. And so while the revenue they are able to generate is going up, the question also becomes, Carl, what's their expense in terms of how much they're paying for that compute?

12:11I mean, looking at risk factors, we've talked about how many elements would be layered into the risk factor. The first one is about the possibility of not getting enough compute. Right. That's the lead. Right. And I think that that is really the main focus of Elon Musk's remarkable conference call last week, where he's talking about this amazing expansion in the amount of compute he has. Now, there is a page, and I feel awful that it's just so piecemeal that we're doing, but it's better than nothing. Page 137 has this margin. Operating efficiency drives margin expansion. The margins are exploding.

12:46And when you think about compute, you start thinking, all right, how many people, how many companies are really going to be left? And I think that was also the point of Elon Musk. But this company is most definitely here to stay. And I think people have to recognize that there's so many companies that need to be paid that at last we've got someone who can pay them. Yes, that is the hope. Kate Rooney's had some time to pour through this as well. Kate, what are you seeing? First thing here is a very unique voting structure for the founders here. So there is a single special share that's going to be issued.

13:22It's going to be held through this entity by its seven co-founders of Anthropic. It will hold a 50.1 percent power, basically, of the voting rights here. So they also talk here about the long-term benefit trust retaining the right to appoint anyone to the board. So it's a very unique governance structure. It doesn't look like the percentage of ownership is filled out quite yet when we're looking at the S1. And that could be coming later in terms of how much each founder owns. We have the shares on their own, which don't mean much because you don't know what percentage of the company that is. I also want to highlight for you guys customer concentration.

14:00So this filing talks about dependence on just a handful of very big customers and cloud partners as well. It says your two customers accounted for about 12 percent of revenue last year, nearly a quarter combined. And then the first half of this year, they talk about just one customer basically accounting for about 10 percent of that disclosure. And they don't say who that is. They're sort of customer A, customer B. But, you know, you can back into that and assume that it is one of the big cloud partners here that they have that could include Google. It could be Amazon as well. And that's what we've got so far, guys.

14:38But I would say there's also we're going through the risk factor as well, which is quite long. So we'll keep you posted. That class you mentioned, though, of the strategic business partners, Kate, that's the class N. Is that what you're referring to? The, you know what, girl, I'm doing a lot of control. There are a slew of classes here. There's A, F, E, T. Are you talking The voting structure is a unique—it says a Class F, I believe. Hold on a sec. Well, that one—so two separate things. The voting structure is basically a single share that has super voting powers. And there are—this is unique.

15:16It's not just one founder. There are a handful, I believe it's six or seven co-founders here that are all sort of part of this one group that will have majority, 50.1 % voting power here. But we're still going through it, guys. And we'll get you if we have any other details on exactly who the customers are. And I think Picker's probably digging through this as well. No, we're all digging through. And again, you know, as we said, okay, we're not going to get some of the key things we care about here in the capital markets, namely. All right, how much are they raising and what are they going to seek as a valuation?

15:51We will get that, but we don't have it right now in front of us because that's not part of this. You know, we do know Goldman Sachs is lead left. It's, by the way, same underwriting kind of structure as for the SpaceX IPO, which I think is generally viewed as quite a positive launch for that company. One of the things, guys, also that's standing out here is some of the credit risk that they're talking about. They do point that out in the risk factors, which are quite long. But they talk here about the risk of just debt markets in general, which, of course, you guys have been talking about. But it's something to look out for as some of these AI labs really do lean more on debt.

16:28Rates are going up. And so that is a factor for Anthropik that they're calling out explicitly here in the S1. Yeah. Well, we've been saying, I mean, and it's been made very clear even this morning, access to capital is going to become a differentiator just generally. And so you want to hit the markets when you can. And they're going to be raising an awful lot of equity. Will it be more than SpaceX raised in its IPO? Will it be more than what Alphabet raised when it shocked everybody with its raise in the spring of what, some 80 plus billion dollars in equity? Will it be more than the 95 billion now that Meta is talking about raising 35 billion?

17:03And this week alone, we've got a couple of companies that people are worried about. What will Microsoft do? What will Amazon do? Right. And not to mention deals that are still out there, guys. Sixty billion dollar Broadcom deal that's still out there for compute. But Musk didn't get the$40 billion this morning, right? Yeah, he got$40 billion through Apollo as well. He's an NVIDIA customer, which is why that's all. For buying GPUs, because he is a hyperscaler. And who's he actually making them available to? The likes of Anthropic, which we know is potentially Compute Star. There is a larger, I do want to come back to the larger, right?

17:37We're talking about Anthropic, a company, by the way, for those who may not have focused on its history through the years, that came out of OpenAI. Remember, OpenAI, of course, begun as a nonprofit to make sure that AI was developed safely, founded by none other than Elon Musk, along with Sam Altman, the two of whom are now, let's call them... At odds. Yeah, at odds. Yeah, enemies. They don't like each other very much. Not even revenue. And Dario Moday and his sister and a couple others came out of OpenAI saying, you know what, you're not doing it safely enough for us, and launched this company not that many years ago.

18:12And their enterprise. And their focus on the enterprise really gave them a lead to a certain extent. And we've all talked so often about their ability to offer tools for people to code. And now so much coding at every company is now done by Claude, by AI. Don't rule out Cursor. Open AI has come back. Don't rule out Cursor. This competition continues. but they will be the first of the two labs that we've been focused on for so long, at least the last, let's call it almost four years since ChatGPT was first introduced, to come public in the next few weeks. But it doesn't come, guys, without this overall concern around safety.

18:52And that is certainly one part of, and we've wanted to focus on it. We've heard this so many times from Dario, of course, in various essays, not to mention their own researchers and the like, but it's worth taking a look at. Concerns have been raised within our organization and by other researchers, regulators, policymakers, industry participants that sufficiently advanced AI systems could, if not developed, deployed and operated with appropriate safeguards, oversight and security measures pose catastrophic or existential risks to humanity. Particularly if such systems were misused, replicated, weaponized or allowed to operate with a high degree of autonomy.

19:27And they go on to say they've observed their models are contributing to the pace of their own research. And these risks will increase as AI systems substantially automate AI research and development. For example, they accelerate AI progress beyond our ability to evaluate and safeguard new models. That's been the debate going on right now, of course. And this thing happened so fast, even maybe in the last few weeks, what I'm hearing from major companies, it's now smarter than we are. Yes, I think how they put it is unexpected, strategic or self-preserving behaviors on the part of their models. And then you get to this interesting question around governance, I think.

20:08If you really are serious about making sure this is safe for humanity and you are a public company, conceivably with a duty to your shareholders to do everything you can to become as profitable as you can, as quickly as you can, can you balance those things? And that's where this public benefit corporation that they're going public as becomes important. That's where this, was it a trust that they have? Yeah, they have a very big commitment. Trustees as well. You know, it will be fascinating, guys, to see how, and both OpenAI as well, which, of course, has a foundation still that is a nonprofit that does still have enormous influence there, including a 26%, 27 % stake in the company.

20:50How are they going to approach all of this? Can you really pace the frontier if you have a duty to your shareholders? And don't forget, there are lots of cheap models up against them at the end. And if you're fighting off open source models in China and the like? Or are you just in a race no matter what? I was going to say that this, a lot of people talk about the railroads. No, this is a gold rush. And you've got to get there early. You got to before you have to start panning. You want to start getting the easy gold. These guys, everyone's on the easy gold. But pretty soon it's going to be the hard gold.

21:26Leslie, you're watching something within this, right? Yeah, Carl, I was looking, you know, we were talking about the third quarter financials, which was discussed at an investor meeting more recently. But they don't appear to be in this S1. However, I have been chatting with people familiar with the matter who tell me that they should be included and updated in the days or weeks to come. So that was kind of going back to David's earlier question about whether they were in there and the possibility of the filing being amended to include them. Additionally, I was just looking through the lockup provisions here, which seem to mirror something similar that we saw with SpaceX with this stair-step approach for a lockup.

22:09And this is important because historically you've had very simplified lockups at 180 days. They kind of just allow anybody to sell and that could put some pressure on the stock. What we saw with SpaceX that was relatively successful, I think, by many measures, was more of a drip drip approach to the lockup over time, which also kind of coincided with the fast track entry into some indexes as well, which was seen as a more of a bullish signal for the stock. So it kind of you had the pressure matching with with more of the bullish signals. And we see the same thing here with Anthropic, where they say this is on page 321, for example, on or after 6 a.m.

22:51Eastern time on each of the dates that are 27 and 48 days, respectively, after the date of this prospectus, up to 6 percent of the aggregate number of lockup securities held by such lockup parties may be transferred, essentially. So examples like that, they're basically a timeline by which certain individuals will be able to sell. And there are certain dates on or after 6 a.m. kind of following the date of the prospectus. So kind of an interesting bit of information here is the lockups can be a concern and an overhang for recent IPOs. We've been focused on SpaceX's lockups, which haven't actually impacted that stock too much.

23:31Speaking of SpaceX, we are getting, and again here, we try and figure out what are we learning that's new or what has already been reported. They are giving us some detail in terms of their planned spending on compute at Anthropic. And I did want to share it specifically to SpaceX. We agreed to spend an aggregate of up to$120.5 billion over terms extending into 2029, consisting of approximately$3.1 billion in the aggregate per month once all capacity has been delivered, which they expect in December 2026. That is specific between Anthropic and SpaceX. And they do say that there are initial non-canceable periods that end in January of just next year.

24:16Remember, we've learned that there are short-term agreements, but they could go to 2029. And perhaps we did not know that starting in December, they'll be spending$3.1 billion a month with SpaceX alone paying for that company for compute. It all kind of links up, doesn't it? SpaceX goes out, they raise$40 billion. Apollo's helped in terms of leasing GPUs, which they're then going to turn around and put in a data center that they then lease the capacity to the likes of Anthropik. The data center is 50 billion dollars. A gigawatt is 50 billion dollars. And here's Elon Musk saying he's got 1.7 billion, but he intends to get 10 gigawatts of compute online in 2027.

24:58Yeah, he's doing, well, 10 by the end of next year. At 50 times 10 is 500. I got to hand it to him. If he can do it, then everyone's willing to buy it. We have other sort of details about their spend. and$44.9 billion spent with Microsoft between November of 2026, so basically starting now and May of 2033, regardless of actual usage. We had heard about cloud commitments as much as$31 billion, but now they're telling us it's$44.9 billion they'll spend with Microsoft. And this is, again, this is what the company's obviously been focused on. One of their most senior people, Tom Brown, been focused on that.

25:40Securing compute is one of the keys. Amazon,$110 billion between May of 2026 and April of 2036. We have heard about that previously. Some of that has been reported. Google,$114 billion between April of 26 and July of 2033. I mean, these numbers are enormous, as you might expect. And they are the commitments that Anthropic has made to secure the compute it needs, to keep bringing in the revenues. Someone's paying for something. Can you blame Meta for this mixed shelf filing as soon as possible this morning? No, that was good. I guess that was good timing, although I don't like the at-the-market.

26:19And yet, somehow, Oracle managed to do at-the-market without hurting people. Look, in Meta, you have to expect that the rally, if there is one, goes slowly because at every turn, there's stock available. You're talking about the$60 million in Class C common. Right, which everybody thinks is so darn easy to do. But I don't know. My charitable trust owns Meadow. We sold some. Do I want to buy it back? I want to see how heavy it's going to be. Well, I don't even think if we take a look at Meadow, and we're going to get started with trading here in just a few minutes, the stock didn't really look down at all, even with$35 billion,$22.5 billion of common being issued now, and then another$12.5 billion of a convertible.

27:01Let's say they do it in the hole at$6.20. Everybody in the world wants to be in it because it maybe comes back to$6.35. Which bankers handle any of that? It's Goldman and Morgan Stanley. The amount of fees in the last 48 hours are going to make it so you've got to buy Goldman and Morgan Stanley. Yeah, there you go. Do you have a problem with Musk's thought that if you don't have substantial revenue that's not data center, debt markets are going to lock you out? I have no problem with it. I think that he is a self-serving comment that also has the possibility of being run. And yet here's Nebius fairly down, and Corweave, of course, is the Achilles heel, if he's right.

27:49Corweave has to figure out a way to be able to get that interest rate down, and I do not know what it's going to do. Do you have any idea how Corweave can get out of this jam? How they get out of the jam? I mean, just say it's a slow company. They have a lot of contracted compute already that brings money in, right? They don't have that much unsecured debt. They have to find a guardian angel that we don't know about. But there tends to be a lot of guardian angels out there. Well, to that point, though, Broadcom, AMD, and NVIDIA are stepping up for$5 billion of this meta offering. Yeah, now$5 billion each or$5 billion?

28:23No,$5 billion total. I mean, they need to find, if I can just be a little humorous for one sec after a tough week, they need to find Fredonia. Yeah. Like those high school dances where you had to partner up, find someone quick to dance with? Find someone quick because they need someone on high. Don't you think, though, I mean, to be fair, when Musk put his thing out, he was talking about what he said, 200 neoclouds out there. I mean, there are a lot of smaller players out there that are doing a lot more on the come. That's true. Than are the big players such as a core weaver or nebius. Well, of course, what would happen is it just couldn't raise a lot of money from here.

29:02It gets crowded out. Right. But the companies that we see on the screen are reacting positively because there's going to be money coming to them. There's an entity, Anthropic, that then gives money to everybody, theoretically. And we've all been waiting for it. And this is really important. Those who thought it might not come until next year are just relieved. They're relieved because it means that there is money in the system and Anthropic lives. One Anthropic to live because it's the key company on the upside. For the sake of the market, you mean. Yes. I mean, look, we got through SpaceX and it was brilliant.

29:40It was. It's gone very well since then. The stock has performed extremely well. One would have to say it's weathered the lockups or at least that have come its way thus far. Is there that much money? It is going to embark on the largest deal of all time, namely buying Tesla. So that'll be interesting as a test for SpaceX shares as well. Is there enough money? Enough money for what? If an Amazon announces anything, you build it. There's enough money for them, Jim. But to Musk's point and to the point I've been making this morning, access to capital being a differentiator, it's not clear that there's enough money for everyone.

30:15There we go. Totally agree. Very excited. Well, Citadel's on the tape today saying the credit is going to be tested. There's a lot more supply on the way. We'll see what we get here at the opening bell in a few seconds. It wouldn't be just, you know, if SpaceX really needs all that compute, I think they're betting that they're going to be able to buy it from companies that can't finish the compute. Yes. That's what I think. Stranded compute is how Musk put it last night. Let's get a look at this opening bell. quite a first half hour of our show today at the big board on Wall Street at the Ball Court.

30:51An aluminum packaging company marking the appointment of Ron Lewis as CEO at the NASDAQ in Times Square. B &B Standard, a digital asset treasury company celebrating its recent name change from CEA Industries. And at CBO in Chicago, the CBO floor operations team here, Jim, as we kind of got into the low 7600s last week. Well, last week we first of all realized that the danger of cybersecurity, with some real people saying it's dangerous, hurt, situation. I also think, look, the war, oil went up so much that if you believed that the consumer had a shot and that worse would not have to talk tough, well, you lost that.

31:37You lost that because of the war. And I think that the backdrop of the war is going to make it so a lot of companies, people are going to say it's the last good quarter. There is a multiple contraction going on right now from both the data center, but also just from interest rates that, David, you know, people just don't want to pay up the way they would have. And I look at a Verizon, for instance, and, you know, Verizon is up against Starlink. And I looked at it, I did the work, and then I said, what am I doing the work for? It's up against Starlink. You really believe that? Yes, I do. They're putting together all that different spectrum and all the different towers.

32:17I just say, listen, why do I want to be in Verizon? Why do I want to be in AT &T? You're not alone because, I mean, basically the owners of all three of the majors, T-Mo, AT &T, and Verizon, have been selling the shares as the competitive threat from Starlink has continued to grow. Right. I guess, as we've reported many times, as Starlink starts to put together a real terrestrial network to offer wireless services. That said, it's going to take them a while. Right, but at the same time, you want to go and you want to see who's going to have the great deals for the duo. And then don't look at Reddit, because Reddit's got some people who have been saying the duo, that there are cracks after they use it.

32:57Now, I know that sometimes there's just these complainers. Apple hasn't said anything, but let's not forget, Apple stock has been the best. If it can put to rest these issues of instances of cracked screens and you can get it from T-Mobile, I think you may find, you know what, why don't you buy 5.0? Did you get yours on Friday? No, I'm waiting for the big deal to be announced. I think there'll be a big deal. It's cheaper. ATT is a big deal. Yeah, we were reporting on Friday about some of the deals out there already. Oh, they're big. They're big. and you can watch the Giants. I have a new team, right?

33:33And you can do some work and you can, I don't know, for all I know, gamble. Just gamble like a demon because when they're going to give you 10 and you're the Giants, you want to take that. Among the risk factors in the S1 out of Anthropic, it does center around regulatory and social pushback. I know Kate Rooney's been watching some of that, right, Kate? Hey, Carl, they say this explicitly in the S1. It is here on page 27 for those following along on political risk. I want to read you one of these quotes, guys. It says they talk about navigating and evolving in uncertain times. They say here there are divergent global regulatory landscapes.

34:14And they do talk about this stakeholders, different legislators and regulators being one risk. And then this is another quote here. The political environment and our engagement with policymakers and public positions on AI and safety could subject us to heightened security, political backlash or disproportionate regulatory action. So they're calling it out here. There's more as well. Other notable items in terms of the risk factor, you guys have been talking about it, substantial debt. They say here on page 32, their ability to meet the substantial and growing financial obligations of these requirements to secure compute.

34:50It depends on their ability to generate significant revenue growth. They also need to grow their customer base here, they say, and expand revenue rapidly to generate the cash flows necessary to satisfy these commitments. And then, guys, on the loss number, this has been a big open question. So we do know, and Leslie brought you this, Anthropic lost about$80 billion in the first half of 2026. We are finally getting a little bit of clarity on why and what that looks like. It is an accounting effect. So it's not like this is debt-related. It's not like this is a compute issue. It has to do with their soaring valuation.

35:26They did a lot of these as convertibles, and as the value of the company went up, that triggered massive accounting charges. So it does give you a little bit of color in terms of why they lost so much money. It does have to do with accounting. And that's the latest here, but we will keep looking through the risk factors. Also, no single customer represented more than 10 % of our revenue, Jim, for over nine months. Although top five is about a quarter of their revenue. We've been watching sort of that revenue concentration as well. They spelled that out a bit. Right, and they have very little international, so there's a lot of opportunity.

36:03David, Alex Karp is not listed so far as a risk factor. He's not? No. Well, he probably should be. His comments really resonated. Yeah. I mean, the chart that is going to be focused on by those who are favorable on the prospects for this IPO will be that one on what page am I on? That was amazing. 136%. We already showed it, right, the significant increase in revenue per dollar of compute. The one that has 5.2 times on it, guys. I think we've already built that. But does show significant increase in terms of revenues and in terms of compute infrastructure expenses. There it is. what they call that revenue per dollar of compute.

36:46So you can see that the right part, right? What color would you call that? I don't even know. Not the gray. I don't know. I don't think that's orange. Let's have a debate around this color because I think it's important. David, did you go to F45, which has the 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, the series? I haven't. But the point is that right column is not going up as quickly as the left column, right? Revenues are accelerating at a faster pace than what they're paying for. No, that's a page I love. That's a page that says, I want some. Can I have some? Well, it depends on where they price it. And you don't know that, obviously.

37:23If it's done by Morgan Stanley, I'll buy it. Yeah, well, it's Goldman lead left. I'll do it right. Stabilization, yeah. Goldman had a huge anthropic position. Remember when it was like$300 billion in March, April? Yeah. Yeah, I mean, by the way, there's a number of huge anthropic positions. And once we get the pricing, once we can figure out, is it going to be a trillion and a half? Is it going to be more? Then we'll sort of back into everybody's net worth. Remember, this has been around now for a number of years. We've got investors, Carl, who've been in there the entire time, been buying all the way up from an extraordinarily low valuation.

38:07Obviously, there's one guy in prison who really wishes that he had not ended up there. Because I think Sam Franklin Freed would have owned as much as 7 % at one point in his company. I think it's important that you mention that he wouldn't want to be in jail. That's just something I would put out there. Even with all that money, it's still a negative. Well, you never know. Maybe he gets pardoned and he's like, you know. He's been trying. Well, why is he not pardoned? I don't know. Doesn't a page of pardons blow in the window and sign? I don't know why. I don't know either. That's a good point.

38:40I know. But, Carl, can I just explain that in all my career, I don't get at 9 o 'clock while we're trying to figure out what's going on and who's been up at Instacare, Maple Bear, whatever it is, maybe gets a deal. And suddenly, you know, we're frantically looking through something as if this is usually done at home. Yes. After the close. Journal says they're targeting the week of November 16th. Yeah, that's what we've been thinking. I went down before Thanksgiving. You prefer to get this out of the way. Oh, yeah. We want this out of the way because there could be something behind it that we haven't thought of yet.

39:12That's why they came out today, because you've got to have at least 15 days. Yeah, look at this. Everybody loves it. Everybody's been working with the 17th is the day. We'll see. But, look, I think that the key thing was that Musk has historically been accurate. I don't know where he's going to get the 10 gigawatt. I mean, Geek Watcher, you need over and over again what we've been hearing is you need power, right? You need equipment. Well, and Jim, to that end, we are going to get both Vertiv, I think, and GEV this week. I know. I'm going to have Vertiv on, and I think that Vertiv is the linchpin.

39:48I do think, by the way, you have to watch these cybersecurity stocks. They're not done. They're not done. Do you think Vertiv, CAT, GEV, LAM Research, Teradime, what is that rollover telling you right now? Right now, it tells you that people say, I don't care. I'm not paying what people say that they used to pay because we had interest rates going up and we may not have enough money for everybody. And it's kind of absurdity that everything is going to be built. These reflect what Musk is talking about, the neoclouds that have the money. And these are all neoclouds in the company stocks. Right.

40:24Even though I know Vertov's got a lot of firm orders, I don't know if people say those continue. No, really? Really? I'm a believer in burden. Right. But we've all looked. I mean, it has been Dell. Dell is another example of a stock that's just been killed. How about Micron? As there was concern about the ability to build new data centers and fill them up. Did you kill Intel? Who killed Intel? I had nothing to do with that. You know, so I'm with three cyber guys this weekend. And what they say to me is, well, how about this? How about that? He goes, Eli Lilly,$2 trillion. I said, all right, I'll take that.

41:03It's 1.2 now, so that would be a nice move. This is someone who is actually deeply involved in cybersecurity. He says, the one you want to be in is Eli Lilly. I said, aren't you, like, involved with CoreWeave? He said, Eli Lilly. Incredible. It is incredible. It's nice to hear a name that has nothing to do with AI. Isn't it? Yeah, thank you for that. I mean, the forward PE of health care is higher than the forward PE of tech, mostly thanks to Lilly. Well, Lilly is, like Ken Langone said, going to be the biggest company in the world. All this capital raising, you made the point, has got the banks up.

41:36I don't know if that's the direct reason, but it can't hurt. Well, they're having a good quarter. All of them up over 1%, the big names. I'm talking JPM and Goldman and Morgan and Wells Fargo. The banks are having a good quarter. Just that the one thing I worry about is that will people use the sweep? is this the end of the idle cash these companies have? Explain what you mean. You own a CD, whatever. There's a little bit of money that they can use. Dry powder. Yeah, dry powder. There's a lot of people who carry things that aren't interest rate bearing because they're unaware. And maybe they will become aware.

42:15Why would they become aware? Well, because you can just say, listen, if I have any excess money, it's cash. Oh, you're talking about the agents. Sweet. Yeah. No, I think the agents can do it. I just don't want the agents to run away with my money. Right. Well, that is an issue. I mean, looking back to Anthropocene, the risk factors, models have shown evaluation awareness and self-preserving behavior. Wow. This is why I like Shark Ninja. Can't close the bay doors. I want companies that have nothing to do with this. David, I can't close the bay doors. Yes. Sorry about that. Not going to happen. I'm afraid I can't do that.

42:47What are you, Slim Pickens? That was a different movie. All right. He's Keir Dullian. That was a different movie. We did get EQT range resources, Jim, speaking of bank activity. I thought that was a great deal. Once again, range resources is western Pennsylvania and northeast Pennsylvania, where there's a lot of natural gas, and there's been a pipeline that does now make it very easy to build, say, in West Virginia, take all that Marcellus Nat gas. And maybe even if the election doesn't go right for data centers, Maybe you can just take West Virginia and just make it the data center capital of the world.

43:22Where? West Virginia. You don't like John Denver? Yes. That's a big song over in Europe in soccer and international football. I didn't know that. Yeah, it is. They all know the words. Do they sing it in the stadiums? Is that what they do? Yeah. Kind of like we do Sweet Caroline. Right. Amazing. Right. Maybe that's a good moment to do something. I don't know. We do have some calls on the tape today. Ross Storrs is a good one, Jim. UBS goes to buy. You know, there's an example. The previous King was TJX. They're saying that Conroy, it's not a one-hit wonder that he really has something going at Ross.

44:01This would be something I historically have liked Ross, but my channel, Truss, loves TJX. And this is like Ross coming in from nowhere, as if we've got a brand-new golfer who's just passing all the golfers. And I just am shocked that this is not a one-hit wonder. I thought it was a one-hit wonder. You did. Yes, I did. Why? Why? Because I thought TJX would catch up. TJX, have they been caught napping? I don't know. They don't seem to get caught napping very often at TJX. They did not have a good last quarter. And that's why I think this is a very pertinent piece. And then it's Evidence Lab, which I very much like.

44:40Yes, we know you love Evidence Labs. I do. Evidence Lab is for real. They spend a little more time, I think. Do they spend a little more time? Yeah. Well, it takes a while. Just stamp. There it is. There it is. The stamp. There it is. The evidence stamp. Just that alone takes more time. At Walmart and retail, Jim, in your absence, Walmart has held$120. Costco is holding$1 ,000. I think these, again, these are the companies that are holding line on inflation at a time when everyone else— Look, some people say Walmart is using that surge pricing that they keep saying they're not. But, you know, I don't know Walmart in all the years that I followed it.

45:15And I followed it since 1984. I can't remember when anyone ever accused them of dishonesty. That place does not. I don't want to say that anything can't be changed, but that place is pretty darn good when it comes to consumer. And Costco's been. I was on Costco's site this morning, and I send slower prices for the various things that I saw on the Costco site. Because I'm looking for the phone. Of course. I'm waiting for the phone deal. Of course. Guys, certainly an important news morning here. for us on Squawk on the Street, given everything we've been covering. Uh-oh, what do you got? Also an important morning as well for us in terms of just counting the years that we've been working with our amazing colleagues.

45:54And in this case, I am talking about one, William Chip Aiken, our producer, for 30 years, my colleague. Fantastic. And ours for so many years. Yeah, they're ringing the bell maybe also for him as well. The last 15 years, Chip has been in our ear. Some surprise there. Of course, he and I are also alums of the famous Tufts University as well. Same class. That's his wife, Karen, his son, Harris, who have just joined him in the control room. Surprising him, we hope. And we just want to thank Chip for an amazing career that is continuing, of course, we hope, for far longer. but has spanned some 30 years.

46:39That is quite an accomplishment to deal with people like us, especially. Day after day. Day after day. With somebody who completely ignores rap cues and doesn't even know what they are. No, I don't know what that is. But that's okay, because I'm new. I'm young, okay? I'm new and young, and I can't wait to find out how this stuff works. Again, Chip Bacon, wow, 30 years, senior live producer, and the man who's told us so many times, and we'll probably in a moment tell us. All right, guys. Right now. Actually, I don't think we've had an ad break this last hour. No, and we shouldn't because this is one of the most amazing news mornings ever.

47:15It is. It is. Congratulations, Chip. Making this show work every morning, it's hard. It's hard work. We're so grateful. Jim, we're going to work our way through a pretty big week here. Yeah, we had a big week. Fed decision. I'm curious to know how you're thinking about Google, Microsoft, Amazon, too. I think one of those is going to say we need more money. And I just don't know which one. Now, Google stock is down from when they did their deal. Amazon's, I think, can talk about the$50 billion semiconductor market. And I think it's possible Amazon's saying right now we can make a lot of money. Right now.

47:50Microsoft is the one that has been most conservative, does have co-pilot. We don't know how much open AI. I think the world of Amy Hood is the gatekeeper. Does she think the world of me? I like Amy Hood. Okay. That's good. Yeah. It's nice that you like her. I like the Duke speech. I am a, well, I'm at a loss for words about her response to me about my, why do you think that is? I don't know. Well, because there's been a loss of words. Like, no. We gave everybody the inside story of everything, didn't we? Chip, this, whatever. Watch bonds as well. As we said, Fed Decision on Wednesday. We'll get durables tomorrow, some conference boards, some new homes.

48:37Obviously, we've been watching that dip below 5 % last week. And Morse is going to say, I'm not done. I'm just beginning. Maybe. Still sub 5 on the 10 this morning. Don't go anywhere. Ever notice how every new AI tool asks you to explain your business to it all over again? Google Workspace with Gemini skips that. You get Google's AI built right into the premium versions of Gmail, Docs, Sheets, and more. No new tools. No extra logins. Just one subscription that gives you personalized AI right inside your actual workflow. Make work less work. Try Google Workspace with Gemini at goo.gle slash workspace with Gemini.

49:19At Fanatic Sportsbook, we're talking real fan cash on every bet. Spend it on jerseys, tickets, collectibles, and more. Rewards hit different here. New customers can bet and get up to$1 ,000 in fan cash over 10 days. Fanatics, the only sports book to reward every bet. Must be 21 plus. Gambling problem? Call 1-800-GAMBLER or 1-800-MY-RESET. CT, call 888-789-7777. MA, call 800-327-5050. NY, call 877-8HOPE-NY. MD, visit mdgamblinghelp.org. Offer not available in all states. Terms apply. See Fanatics app for details. This episode is brought to you by Schwab Market Update, an original podcast from Charles Schwab.

49:53Schwab Market Update, an original podcast from Charles Schwab, provides an information-packed daily market preview on stock updates, U.S. and global economic news, monetary policy decisions, and key results and statistics that may impact your trading. Download the latest episode and subscribe at schwab.com slash marketupdatepodcast or find Schwab Market Update wherever you get your podcasts. It's been a little bedeviling for commercial aerospace lately. Boeing in a 20-plus percent drawdown, down 14 for the year. It actually had its lowest print of the year on Friday. Right. I'm still with Kelly, but you know what?

50:33Obviously, when you hear about JetBlue and maybe the creditor problems and obviously oil going up, people do not. We have them? We'll get them tomorrow. Oh, damn. I mean, he... Look, Kelly's your true. Yeah, look forward to that. Well, GE Aerospace also has been in a bit of a down draft last several weeks. Larry's terrific. I would buy that stuff. But look forward to Orchberg tomorrow at 9 a.m. For now, Dow up 146. Stay with us. Ever notice how every new AI tool asks you to explain your business to it all over again? Google Workspace with Gemini skips that. You get Google's AI built right into the premium versions of Gmail, Docs, Sheets, and more.

51:15No new tools, no extra logins, just one subscription that gives you personalized AI right inside your actual workflow. Make work less work. Try Google Workspace with Gemini at goo.gle slash workspace with Gemini. Hey, Fidelity, can I get a second opinion on stocks in the Fidelity app? With Fidelity, it's easy to get an outside opinion from independent experts in a single score. And then? When you're ready, trade U.S. stocks and ETFs with no commissions. That's right. I am always right. Investing involves risk, including risk of loss. Zero dollar commission does not apply to customers designated by Fidelity as a professional equity trader.

51:53A limited number of ETFs are subject to a service fee of$100. See details at Fidelity.com slash commissions. Fidelity Brokerage Services, LLC, member NYSE SIPC. Something amazing is happening in data management. Even though storage needs are constantly changing, companies are only paying for what they need now. That's because CDW customized an ever pure evergreen one SLA driven storage as a service solution, delivering hybrid cloud storage that's easy to manage, efficient and continuously evergreen. Simplify storage administration with ever pure and CDW make amazing happen. Find out more at CDW dot com slash ever pure.

52:32Jim, you have a special edition of Matt tonight. Yeah, we've got some good guests. We have Mark Casper, Stocks on Fire, TMO. George Kirsch, continue your excellent discussion that you had, because I just can't emphasize this hacking thing enough. And John Giamato, he's turning around Black Prairie. But I think, look, most important, I lost...

52:56Sorry. It's okay. My nephew was my right-hand man in my left brain. started when he was in, just finished high school. He only asked Jeff Zucker to be the head writer. He wanted to be the writer, and the head writer didn't need other writers. And he passed away last week. He was 41. I'm going to have to talk about him tonight. I'm sorry. It's very hard. He was my guy. We are so sorry for you. Don't be sorry. I heard this week that he was such a big part of the show that scripts were often Masonized. No, not so. The only thing he used, Masonizing, and it was well before Chachipetit. It was well more.

53:40You didn't trust any of those. 21 years, he wrote. Not one single mistake, if you can imagine. Not one single mistake. He solved the European debt crisis. He solved the COVID. He solved the Great Recession. He was right on every single one of them. He was right about everything. He was brilliant. Okay, how about this? He was 41, and we are glad that we had him for that long. And I will talk more about him. He was highly emotional. He was my citrus kid. He was head writer. He saw Zucker, and Zucker said, you know, Harvard's hard. He goes, Zucker, go on there. And he said, maybe for you. And he never missed a deadline at Harvard, even though he was a great Latin and Greek scholar.

54:25He came back and wrote for us. He knew my voice. He was my right arm and my left brain. He was remarkable. He loved this network, and he loved Mad Money, and he lived for it. And great parents, obviously, Todd Mason, my sister Nan, Jenny, a girlfriend, and died of an enlarged heart. I went to see, you have to identify next to Kim with my sister, and all we could think about was how we all wish we had a better relationship with Cliff. He was ultra-high-functioning. I've never seen anything like it, but at the same time, he was very uncomfortable with people. I'm going to talk about that because I want people to know who he really was, which he was a genius, a brilliant, irascible man whom we all loved, But we couldn't figure out why other than the fact that when that copy came in, you sent him a 5 ,000-word memo, and you got 1 ,250 words of the greatest summary in the world 25 minutes later, and he was playing video games the whole damn time.

55:36Incredible. Sorry to get choked up. I loved him so much. Our thoughts continue to be with you and with your family. Thank you. And we need that. We need that. Appreciate it. I look forward to that remembrance tonight, Jim. Thank you. Mad Money at 6 o 'clock. Stay with us. You've been listening to the opening hour of CNBC's Squawk on the Street. All opinions expressed by Squawk on the Street participants are solely their opinions and do not reflect the opinions of CNBC or its parent company or affiliates, and may have been previously disseminated by them on television, radio, internet, or another medium.

56:08You should not treat any opinion expressed on this podcast as a specific inducement to make a particular investment or follow a particular strategy, but only as an expression of his opinion. Such opinions are based upon information Squawk on the Street participants consider reliable, but neither CNBC nor its affiliates and or subsidiaries warrant its completeness or accuracy, and it should not be relied upon as such. To view the full Squawk on the Street disclaimer, please visit CNBC.com forward slash Squawk on the Street disclaimer. This episode is brought to you by Schwab Market Update, an original podcast from Charles Schwab.

56:38Schwab Market Update, an original podcast from Charles Schwab, provides an information-packed daily market preview on stock updates, U.S. and global economic news, monetary policy decisions, and key results and statistics that may impact your trading. Download the latest episode and subscribe at schwab.com slash marketupdatepodcast, or find Schwab Market Update wherever you get your podcasts.