花旗分析师：AI债务风险被夸大，软件行业加速分化与整合

Well, I think there's a lot of things, but I would say maybe the number one thing, particularly this week, has been what I would describe as the bond market math. Right. So you see all these headlines, there's a coin flip chance that core wave is going to go bankrupt over over the next few years. You're looking at the CDs spreads and things like that. Yeah. Uh, it's not as dire. Is that right? It's not that all of a sudden, core wave is paying 15% interest rates on the debt that they have on the balance sheet.

Certainly the unsecured stuff the market is indicating the cost of financing is going up. Uh, but these customers that core we've has are some of the most profitable investment grade customers on the planet Microsoft uh, Regal, etc.. Right? So it's it's not like they're, you know, serving these customers that are that are going bankrupt. So the day before Thanksgiving, Heath Terry walks in the office and says, I need 40 pages Monday.

I mean, that's how they roll at Citigroup. Is there going to be a roll up of all this stuff? How do you foresee the haves, the winners and losers coming out of this five years from now. Well, I think it really comes down to scale, right? I mean, you look at your weave in some of the larger neo clouds, it certainly that we cover Oracle. That'll be topical next week with their investor day. Uh, these have real scale. They have, you know, gigawatts of capacity.

There's a lot of smaller players. You were talking about this, uh, IPO in South Korea earlier. Like, I think the there's not going to be hundreds of these, of these neo cloud. So I think this, you know, if we do enter a cycle where we're a cost of capital is a lot higher. Uh, that is a problem for, for some of the smaller players and ultimately benefits the large, uh, scaled ones that, that, uh, we cover today. That's kind of where I wanted to go to, looking at your coverage.

Uh, Tyler, a lot of buy ratings out there on your software, but also a lot of hold ratings as well. That kind of gets me to where? Just software in general. Yeah. How are you framing out for your clients AI as a competitor or a friend or enemy? How are you framing that discussion now? Because a lot of people initially just threw software out the window. Totally. And obviously I cover a broad range of companies, everything from the NIO clouds, hyperscalers down to to application software.

I think what we've been describing, uh, to our clients is you're seeing this bifurcation, particularly in the second half in software earlier this year, as you pointed out, if you pick the number one performing software stock, it's still underperformed the worst performing semiconductor SoC, right? I mean, snowflake was at $100 in March or April, and now it's, you know, three x that. So I think you're seeing this bifurcation where software is accelerating some company's growth rates.

We look at snowflake, we look at Palantir, we look at Microsoft. On the other hand, it's a real problem. We think for uh, a number of these other application companies, you look at Adobe, uh, some some smaller companies out there, uh, are, you know, are seeing a lot of this pressure from, you know, it budgets getting diverted, uh, to, I think another theme, though, and there's a fresh headline, uh, today, Salesforce Buying Praise, which is, uh, we cover both of those.

You're seeing a lot more consolidation to both buy, uh, strategic as well as, uh, you know, some certainly some private equity bids out there. So, um, I do think you have this bifurcation dynamic going on. I take a look at like again, a Salesforce, an awesome company has been a great stock for such a long period of time. Are they risk? Well, I think there's that's probably one of the most debated names in our coverage, right?

I mean, if you look at the stock clearly it's, uh, perform better than, than some of the software names out there. Uh, it was up, you know, 20% post last quarter, this big announcement, uh, with cloud force, right? Uh, but you're seeing them do a lot of M&A like praise this morning. You've seen half a dozen other, you know, relatively large, uh, deals. Uh, and so the question is, can they actually organically accelerate their growth rate, um, the right time.

Right. couture. This can't say enough about his work at Citigroup is well, of course he looks at Microsoft with a target. I guess it's 100 bucks up. That gets my attention as well. It's just so ginormous and carries so much historical baggage. What's Microsoft look like in five years or ten years? Nadella is maybe retiring, moving out, whatever. Yeah. Where is where is this behemoth in five years? Besides, Steve Ballmer has never sold his share.

Yeah, I think the is we look at Microsoft, I mean, they have still some of the most underappreciated pricing power in the market, right. Uh, if you think about everything, just think about their office business for for a moment. Right. Well, under $100 a month to get basically access to the modern productivity system, cybersecurity, AI through copilot. Uh, and you compare that to how other how other software companies are charging hundreds of dollars a month.

So we think there's a lot of room for them to monetize, particularly as they start to build out copilot agents. Uh, and a lot of the exciting things they're doing there. You know, I think what will be really interesting to see over the next few months is their first party models. Right? So in our most recent CIO's survey, uh, Microsoft's own in-house models that, uh, that they've developed, uh, according to the CIOs we interviewed, are right up there, uh, with the leading labs, uh, out there.

And so we think that will be a big part of the story. Um, big picture two of Microsoft's most important businesses, which is the office franchise as well as Azure. We see those accelerating, uh, over a multi-year period. And so I think this is a company where we're going to be seeing, you know, north of 20, maybe 30% revenue and earnings growth. Uh, you're next year. Did you say 20% revenue growth? Yeah, yeah, yeah. The lawyer said you could say that compliance is just sitting tight.

He's got a buy rating and a big price target on there. Tyler. When the city technology research team gets together for their weekly or monthly call or whatever it is. Does anybody ever have the gall to say how much of this I think doesn't really pan out? And what does it mean for everybody? I mean, is there anybody asking that question because everybody is all in on this, I think. But nobody can really tell you what it.

I don't know. It just feels like there's so much, you know, kind of weighing on this I story. How do you guys think about the big, big, big picture? Yeah, I mean, I think there's different degrees of how this plays out. It's not a 100 or 0 thing, right? I mean, I think clearly, uh, what we're seeing today. Right? You just look at the, the data points on the ground. We were talking about Neil Cloud's earlier. Demand is orders of magnitude larger than the supply.

Uh, you look at the hyperscalers backlogs. I mean, they're growing. They're growing triple digits. If you and I stood up in Neo Cloud business today, maybe not a good idea to go public, but there's no one else that has capacity for 2026, uh, available on the market. So this stuff is is is sold out. Um, it's really hard to see that type of scenario where this this doesn't pan out at all. Do you have a single Best Buy?

Our top idea right now is, uh, MongoDB. MongoDB? That's not. What are you looking at me like? John Tucker. You you you were in their IPO, right? Oh, yeah. Give me that. Give me their 15 second. Call them MongoDB eight. What is it and why is it your top? So it's a database company. You need databases to build agents to build applications. We know there's way more software being created today. So there are natural beneficiary of that.

And ultimately we think, uh, their business is going to accelerate. There's a lot of ups