Google Erases a Record $500 Billion as AI Eats Into Sales

0:02 Bloomberg Audio Studios, podcasts, radio, news. You're listening to the Bloomberg Intelligence Podcast. Catch us live weekdays at 10 a.m. Eastern on Apple CarPlay and Android Auto with the Bloomberg Business App. Listen on demand wherever you get your podcasts or watch us live on YouTube.

0:24 We're looking at Alphabet shares falling as much as 14%. Paul, that is the biggest move ever on an intraday basis.

0:32 It's just extraordinary.

0:33 And again, what I think a lot of people are thinking, this may be a fundamental re-rating of the stock as it relates to the AI threat.

0:42 Right.

0:42 And it all comes down to the Google advertising and search business coming in weaker than expected when it comes to revenue. Because now when people try to Google something, they'll still do that. But you were telling me how when you Google a question in there or you post something in there, you're not clicking on anything.

0:59 I'm not clicking through nearly as much as I used to.

1:02 And that's how Google makes its money, right?

1:04 Or people need to be clicking on those specific links and going into the businesses.

1:08 But instead, you look at the summary, the Gemini summary, and you move on.

1:11 And this is going to be... I mean, once we've hit this turn, you wonder how long... before more people do that and kind of opt out of clicking through everything entirely.

1:23 Let's bring in Matt Bloxham right now.

1:25 Matt is one of our senior media and tech analysts joining us from London to talk a little bit more about Alphabet as well as Microsoft. Matt, how are you thinking about what Alphabet told us? Do you get a sense that folks within that company as well are regarding this kind of as a turning point?

1:44 I think we've been in this kind of transition period for a couple of years with Google. As you've been talking about then, that there is this big change in how people are interacting with Google's products and potentially using different products. And so there are two things that they're grappling with here. One is that as people move from just regular Google search to AI mode, there's this difference in monetization that they haven't quite figured out yet. And we've seen that time and time again across Google and particularly with meta as new products have been introduced. Initially, the monetization isn't that strong and they have to figure it out. And brands and media partners have to figure

2:26 out how they're going to spend money against that and make it work. So I think to some degree, that's just going to take some time, but they will figure it out. And I guess what's reassuring, if you're a Google investor right now, is that Gemini has gone through a billion users. Engagement is going up. So hopefully it is a monetization opportunity. challenge that they can address. Obviously, the other big threat there is you've got the likes of Muse coming in from Meta, which is a rival to Gemini and another tool people might use rather than Google to search for things and to buy things. And so even if they figure out the monetization problem, you've got a market share problem. And so that, for me,

3:09 is the kind of twin challenges that they've got and the slowdown in ad sales growth. It's still growing, but there was quite a big slowdown in this quarter is starting to make people feel a bit more that they're not fully on top of that. And even though their cloud business is growing and grow arguably a little bit faster than expected, it's nowhere near enough to compensate for the kind of the profit slowdown at their core ads business.

3:37 And at the same time, they're investing $ 200 billion in capital spending this year, potentially $ 300 billion next year. So the sums don't quite add up yet.

3:45 So Matthew, kind of at its core, most basic level, the advertising business for Google is clicks times cost per click, what they can charge an advertiser for me clicking on an ad. And simply put, I am clicking materially less often.

4:03 So now the question is, okay, how do you solve for that issue?

4:06 Has the company identified any kind of strategy there?

4:11 Yeah, you know, I mean, the whole industry is moving away from search engine optimization to kind of engine optimization. And so both sides of the equation are trying to figure out the best way to kind of. Replicate that kind of monetization flow into a new environment. And, you know, we're not fully there yet. But certainly the industry is kind of on its way to kind of working that out. And also, you know, sponsored links has been a big, big part of it. So not only are they kind of getting revenue from click-through, they're getting revenue from people wanting to be at the top of that list. And again, that same concept applies in a a genitive engine optimization world, but the commercials of it

5:00 are still a bit fluid.

5:03 I feel like before AI became a thing, people were already kind of getting discouraged or annoyed with Google searches because a lot of the times you wouldn't be able to get what you wanted in the first screen. You'd have to page forward, page forward until you found what you needed. So perhaps AI kind of accelerated that. Let's talk a little bit, Matt, about Microsoft and what we saw from that company. It feels like Microsoft delivered everything investors wanted to see, and now it can take advantage of that by selling a boatload of debt.

5:34 Yeah.

5:35 So, you know, I guess, you know, the real focus with Microsoft has been their Azure business, which is their cloud business, a bit like Google Cloud. You know, that grew 48% year on year, which is a little bit better than the market expected. And, you know, it's kind of proven to be profitable and they didn't spend as much on CapEx as people thought they might in the quarter. Now, I think some of that is because a number of the projects that they were planning to spend on have got delayed. So I don't think it was necessarily a conscious decision by the company to spend less. It's some of the things they plan to spend on, just they can't spend on yet. But I think in the round, the combination of that slightly better than expected to zero growth,

6:18 the fact that their kind of productivity suite of offerings remains resilient, that we didn't get a kind of new data point on cowork, but I think generally the sense is that overall that kind of Microsoft 365 environment is still proving very robust. The overall fundamental balance looked good, you know, and we're talking about fine margins here, you know, Google effectively in aggregate, you know, missed by less than half percent on revenue and profit and Microsoft beat by a couple of percent. And yet you see a hugely different reaction in the share prices and that just kind of shows you just quite how much people are extrapolating from kind of small directional shifts.

7:00 Stay with us. More from Bloomberg Intelligence coming up after this.

7:07 You're listening to the Bloomberg Intelligence Podcast. Catch us live weekdays at 10 a.m. Eastern on Apple CarPlay and Android Auto with the Bloomberg Business App. Listen on demand wherever you get your podcasts or watch us live on YouTube.

7:22 A lot of news again today in the tech space. That DoorDash deal, Marvell Technologies is in talks to buy a company.

7:29 A lot going on.

7:30 But first, I need to talk to Mandeep Singh about Google. Mandeep Singh, Global Tech Research Head, Bloomberg Intelligence. Mandeep, we've been having this conversation all day. Just begs the question, Google down 11% here. Does last night's results, do they mark a fundamental change in the business community? trajectory of Google's revenue on the search business? Or is this not structural?

7:56 Because it feels like AI may in fact be impacting their business maybe more than the street thought.

8:03 It is certainly impacting the profitability.

8:06 So look, this company grew 25% top line. When you are over $ 400 billion, 25% growth is mighty impressive.

8:15 The problem is.

8:16 A lot of that growth is in the cloud segment. and nothing can match the margins of the traditional search business. So if search shows any signs of a slowdown, which is if I were to compare just the Google search growth with.

8:31 Meta's ad revenue growth, it was half of Meta.

8:34 So clearly they have a problem when it comes to growing the ad side of the business relative to Meta. And that's where I think the streets response shows a negative reaction. But other than that, there was nothing wrong with the print.

8:51 But you could say, other than that, How was the play, Mrs.

8:54 Lincoln?

8:55 I mean, that is Google from investors' perspective. It's search. It's the growth rate I'm associated with search. It's a profitability I associate with search. That's the story.

9:04 So my question is, does the street need to kind of re-rate the story and say... Maybe I should be focusing more on the cloud business or something.

9:12 It is.

9:12 But even the search side has been re-rated already.

9:16 A couple of times.

9:17 So if you go back, you know, 18 months, the stock was down. Similar sentiment. Everyone was negative on search.

9:24 Guess what?

9:24 They released another model, Gemini new model. And it was at the frontier along with OpenAI and Anthropic. And the stock got re-rated and Google touched.

9:34 An all-time high.

9:35 Now they have missed out on the last few months in terms of releasing a new model, which is why everyone is suddenly concerned. And they have had some attrition in terms of their top researchers leaving. So suddenly everyone is concerned, are they at the frontier or not? And now in the last two weeks, they have released two new models. So there is a sense of urgency around releasing new models just to make sure that they are alongside Anthropic and OpenAI when it comes to what the frontier is. And Meta has released a new model, which is why the sentiment has changed completely on Meta.

10:11 Why?

10:12 Because they released Muse and it was powered by their latest model, which is at the frontier. So- Things change very quickly when it comes to these LLM rates, and that's what we are seeing. All it takes is a good new model release from Alphabet, and it'll be re-rated again, and we'll see the stock go back up.

10:32 But as search as we know it, the way that we used to use it, not dead, but on the decline, I mean, will it ever go back to what it was?

10:41 Well, I look at how these agent tech searches are built. They're built on Google's platforms. Every one of the agent offerings ties to Gmail. So why is the agentic search better in some cases? Because it can be more personalized. Why is it more personalized? Because it's reading off of Google's surfaces in terms of, you know, the intent that a user may have. And that's where I think Google has everything to make a good agentic search. It's just their models are not at the leading edge right now when you compare to Muse or Anthropix.

11:18 Let's talk to you about some M & A deals. Marvell is apparently close to a mega deal for a laser maker called Lumentum. This is a potential all-stock transaction. How critical is this for Marvell? Who is it more important for?

11:33 Clearly, Marvell, I mean, they are paying a lot.

11:37 This deal will be definitely close to $ 100 billion if you think about the size.

11:42 So pretty big deal.

11:45 And, you know, when you're buying stocks at an all-time high, which is the case, you know, with potentially this deal, you always run the risk that, you know, the synergies or whatever they may frame at the outset may not come to fruition. There are execution issues. But with that being said, there is a clear trend in the semiconductor world with more systems type of approach. That's what NVIDIA is doing, end-to-end systems. Google TPUs, I mean, that was the bright spot. They're selling TPU systems. So small chip makers like Marvell don't have a choice but to go the acquisition route. And this is what they seem to be doing here.

12:27 Boy, Lamenton's up 150% this year, so they're buying it.

12:30 And there will be a premium on top of this 150% that the stock has already gone up.

12:36 All right.

12:36 The other one in your space here, DoorDash, Instacart. What's going on there?

12:42 I mean, all of these companies are looking to consolidate because they feel agentic AI would change the interface.

12:51 How people order their stuff. And the more they own the supply... which in this.

12:56 Case, DoorDash would own Instacart supply in terms of the marketplace, that would make them stronger. Even if you're ordering through Muse, guess what? DoorDash owns the backend supply. So they have a lot more cloud compared to being.

13:12 The smaller player. So scale and consolidation is the name of the game.

13:18 So Paul raised a good question, which is, are regulators going to step in and allow DoorDash and Instacart to merge?

13:24 I think there will be.

13:25 And our analyst, Jen Rhee, who covers regulation, feels very strongly that regulators will look at this and say this in the online space, even though there is competition from Walmart and traditional retailers, they will not look at the broader market. They will look at just the online space. And that's emerging number two and three. I don't think it'll be something that will get approved right away, even though the Trump administration is pro M & A.

13:56 So that's the reason.

13:57 So you think there will be some regulatory pushback?

13:59 Yes, there will be. She thinks there will be some regulatory pushback.

14:03 Stay with us.

14:04 More from Bloomberg Intelligence coming up after this.

14:10 You're listening to the Bloomberg Intelligence podcast. Catch us live weekdays at 10 a.m. Eastern on Apple CarPlay and Android Auto with the Bloomberg Business app. Listen on demand wherever you get your podcasts or watch us live on YouTube.

14:25 Wanted to switch gears. Caterpillar reported numbers as well. In this story, you know, big equipment, backhoes, you know, bulldozers, all that kind of stuff. That's kind of become kind of an AI play because of their power business that they also have.

14:40 I remember when we first talked about this earlier this year. And at the time, you and I were saying, what? How can Caterpillar be an AI play?

14:47 Now maybe the flip side to some of that today. Chris Chiolino, Bloomberg Intelligence Senior, U.S. Machinery Analyst, joins us here. Hey, Chris, talk to us about CAT and the earnings release.

14:56 What did you learn?

14:58 Yeah, you know, it actually was a decent quarter. Top and bottom line beats reiterated the guide. But obviously in this market, that's not good enough. You had, again, another strong quarter of orders. Backlog continued to set another quarterly record. But, you know, their report comes on the heels of some pretty underwhelming earnings from some of the AI names last night.

15:21 And Cat, to your point, has certainly been caught up in this AI trade given their power generation and data center exposure. And that growing drumbeat around the durability of AI spending and power generation demand, that doesn't seem like it's going to abate any time in the near term here, even with what is still a pretty constructive outlook as we look into 27.

15:44 Chris, has the investor base of Caterpillar changed because it's now more of an AI play than it used to be? I mean, it was kind of a boring industrial that was very much tied to the fortunes of the global economy, and now it's seen as a growthier name.

15:59 So that's one of the key debates, right? I mean, is this a secular growth story or is it you're still traditional heavy manufacturer, a machinery producer? You know, if you looked at the valuation earlier this year, Caterpillar was trading north of 30, 35 times. You know, historically, this thing will trade 16, 17 times mid-cycle. It was still relatively elevated going into the print as well, north of 20 times. So I think that's also partially reflected in the pullback today in terms of some of the expectations.

16:29 But it's been this kind of push and pull debate around, is this still a truly cyclical company? Or is it more of a secular play on the data center build out? I think there's there's aspects to both. And there probably does deserve to be, you know, more of a premium multiple embedded in the stock relative to what we have seen from traditional machinery cycles.

16:51 What is the company saying about kind of how they view these the data center build out is because are they seeing any delays in from their customers or any cancellations from their customers? What are they seeing on the ground?

17:05 You know what? They've been very tight lipped and somewhat unsurprising. You know, the pretty uneventful commentary from them on the call that, you know, the data center inquiries and conversations really haven't changed all that much relative to last quarter. I think the one encouraging thing to take away though from the print, the implied orders were up like 35% in the quarter, which is much better than we were expecting.

17:29 So it does seem that the orders continue to come in on some of their large engines and turbines. They're already taking orders out into 2029 and 2030. So there is extended visibility here. There's going to be a lot of capacity coming on though in terms of power generation equipment, not only from Caterpillar, but a number of their peers.

17:49 The concerns lie less about, I think, 2027 and more in 2028 and 2029 as we get toward the back half of the decade. And are we going to be overcapacitized?

17:59 When you look at Caterpillar and how it's become this name tied to AI, is there any other industrial company, machinery company that kind of follows the same path but perhaps offers more value?

18:11 You know, it's an interesting question. I think probably the closest comp in that aspect would be someone like a Cummins. Right now, Caterpillar and Cummins both have very strong market positions when it comes to backup power on the power generation side for data centers. Obviously, Cummins has a much greater exposure to the more cyclical heavy truck market here in North America. And Caterpillar leans more heavily on construction and mining. But, you know, they're both have been trading, you know, well above their historical valuations, given the secular growth opportunity that the data centers have presented themselves.

18:52 Stay with us. More from Bloomberg Intelligence coming up after this.

18:59 You're listening to the Bloomberg Intelligence Podcast. Catch us live weekdays at 10 a.m. Eastern on Apple CarPlay and Android Auto with the Bloomberg Business App. Listen on demand wherever you get your podcasts or watch us live on YouTube.

19:14 One of the most read stories on the Bloomberg terminal here.

19:17 Has to do with dating. Gen Z. I'm not sure who Gen Z is.

19:21 Anymore, but.

19:22 Young people?

19:22 I think you have some Gen Zs in your household.

19:24 Do I?

19:25 Yeah.

19:25 All right.

19:25 Very good.

19:26 Dating engagement improves as app use sentiment rise. Nicole D'Souza joins us here at Bloomberg Intelligence. Internet and software equity analyst.

19:34 What's going on?

19:35 Doing it from my phone.

19:37 So it's interesting.

19:39 The dating survey that we did this year, the biggest difference that we've seen is that Gen Z is actually starting to date. Last year when we conducted the survey, 44 Gen Z respondents said that they were single but not dating. And that raised some concerns about the future of addressable markets for dating app platforms. But this year, only 34% said they were single and not dating, and 35% said they were single and dating. So it looks like these patterns are starting to change. Gen Z is getting out of the house. They're going on dates. So definitely a more optimistic survey this year.

20:13 Okay, so many follow-up questions.

20:16 Gen Z also likes to go to the mall now and they're indulging in some, I guess, old school behavior. They kind of like analog things right now. So why have they started to become more open to dating apps whereas before they were kind of avoiding it?

20:31 So I think dating apps have released a lot of features that really appeal to Gen Z. So one of them is in real life events. So they set up events in real life where Gen Z can come, mingle, meet people, and kind of look for a future partner. And so these events were specifically created because Gen Z was not swiping. They were not interested in staying on platforms and talking to people online. They didn't like AI in their dating apps. They wanted to meet in real life.

20:58 I didn't know where to go.

21:02 I'm sorry, we're showing our age in every way.

21:03 I'm just going to ask you, you and your cohorts, what are you doing these days?

21:08 So right now, I would say the most popular dating app in terms of growing users is Hinge.

21:14 Hinge.

21:15 So Hinge has been growing users. It is seen as more of a serious relationship app. So for people who are looking maybe to settle down or engage in a long-term relationship.

21:26 What's the gimmick for Hinge versus the others?

21:29 Hinge, I mean, I think it used to be that it was your Facebook friends, so you had kind of friends of friends in common. Now I think it's more just for people who are looking for relationships.

21:38 Okay.

21:39 Versus Tinder, which has always kind of historically been viewed as like a short-term connections type of app. But they're starting to kind of change that brand.

21:47 So....

21:49 Is this an advertising-based monetary model? Is that the economic model.

21:53 For these dating apps?

21:54 It's subscription.

21:54 It's subscription?

21:55 Yes.

21:56 Okay.

21:56 So are these things, are they good businesses?

21:59 So it's interesting. I mean, if you look at Hinge, I mean, because I think people are willing to pay for an app that will introduce you to your lifelong partner, Hinge can charge a premium and it still grows users. Whereas Tinder has struggled because people are less willing to pay for an app that's a shorter-term connection.

22:17 But if you meet your long-term partner, then you don't need Hinge anymore, and then you'll cancel, and then Hinge loses a customer.

22:24 It's interesting because, yeah, this is a conversation we have a lot with investors, which is what is the motivation for dating apps to function well as in meeting your partner and then also keeping you on the app.

22:34 Correct.

22:35 I think how we think about it is that the addressable market is always growing, right? People enter relationships, but they do break up. And then younger users are getting to a point where they're interested in dating and meeting a partner. So the addressable market will continue to grow, even if people meet on the app and eventually delete it.

22:53 How do I invest in these companies if I ever wanted to do so?

22:56 So Match is public. Bumble, which was public, was actually just taken private. So it's no longer a publicly traded company. And Grindr is another one that is public.

23:07 And I remember reading that Grindr was going to roll out some more premium offerings because the premium customer is still spending money. Is that an approach that we expect to see more of these dating apps take?

23:20 So it's interesting. Grindr's new subscription is called Grindr Edge. it's very AI focused. So a lot of the products are utilizing AI to connect you with people that you would have real interest in talking with, to kind of, you know, tell you about the person that you're talking with, give you kind of specific talking points, maybe resurface old conversations for you to take another look at. And so because of these AI premium features, there are people who are willing to pay. It could be something, but that other dating app platform's eventually maybe take a look at.

23:55 But we haven't really seen it yet. It definitely seems to be kind of a grinder.

23:58 I see.

23:59 Premium at this point.

24:00 Is overall dating app usage still growing or is it pretty, you know, full out?

24:07 I don't know.

24:08 I would say users, when you look at kind of the total market, it is growing. I think it depends on which specific dating app. So Tinder, which is the largest dating app by users, has been losing subscribers for for several years now, whereas Hinge, which is smaller, has been growing because it kind of has a different reputation as a more serious dating app. If you look at Bumble, they kind of moved away from their core brand being a woman-focused app, so they started allowing men to message first.

24:39 That didn't really resonate with users, so they started to lose a lot of their subscribers. So it depends on.

24:48 This is the Bloomberg Intelligence Podcast, available on Apple, Spotify, and anywhere else you get your podcasts. Listen live each weekday, 10 a.m. to noon Eastern on Bloomberg.com, the iHeartRadio app, TuneIn, and the Bloomberg Business app. You can also watch us live every weekday on YouTube and always on the Bloomberg Terminal.