Why the AI Bubble Hasn’t Popped — ft. Josh Brown | Prof G Markets

Okay.

No, no. If you don't find it funny, then that's just what we do.

Great to see you. I apologize for dressing like the great and powerful Oz. I didn't pick up on it until I saw myself on a video screen. That's the energy I came into the year with.

So sick. Oh, yeah.

So great. You know what? I've been in Florida for 2 weeks. I have a tan.

The weather sucks here predictably, but like it's not even bothering me. I just had an amazing couple of weeks for the holidays and I feel good. Ready for '26. Ready to rumble, ready for 2026.

I think the key thing was not getting too negative in November when Oracle was blowing up and it seemed like this was going to be like this comeuppance for all the data center spend and all of a sudden like the floor was about to fall out from under all of us. And like if you fell prey to that narrative, which obviously a lot of people... so there's two categories. There are people that fell prey to it because they are just nervous by nature or they tend to fall for all of those types of bouts of negativity. But then there's another camp of people that really did want it to fall apart. They wanted it to crash.

And you know, some of that is like political, some of that is people who missed out on the AI trade for the last 3 years, but like whatever. You just have this huge chorus of people warning you like this is the next year 2000 and it's dot com, you know, 2.0. But I think the key to this year was not falling prey to it. And if you were able to stay the course, like we're talking right now, the Dow is going to have its first close, I think today, above 49,000. And the NASDAQ looks incredible, you know, through the end of last year and now into the first week of this year. And I think it's really, it's tough sometimes when everyone is screaming bubble, bubble, bubble. It's really tough to like, you know, stay the course and stick to your guns. But that really was the key to the fourth quarter.

Well, no. We still may end up having a bubble, but I look at price. I look at valuation. I look at all the things that classically you're supposed to look at. And the stock market just was not going along with the story that this is all coming to an end. If you looked at the way semiconductor stocks were acting, they just were not giving into that narrative. And if you like, I'm one of the people that comes on the show with you and you know my big thing that I try to get across when I can, what I'm trying to convey is that prices are more important than opinions and prices represent the sum total of people who actually manage money and are voting with their money. So again, I think there's a lot of wish-casting. A lot of people want to see OpenAI in the private market somehow like burst into flames. A lot of people want to see Larry Ellison and Oracle have a comeuppance. Like a lot of that wish-casting is people saying they think that's going to happen, but what they're really saying is I want this to happen.

And if you were looking away from Oracle CDS prices, which I don't think are particularly important, but if you were looking at actual share prices for Nvidia and Broadcom and Corning, which makes the fiber optics for all the data centers, and you were looking at, you know, you definitely saw stocks come down, but you did not see stocks plunging, stocks crashing. That's not how the real money was betting. And here we are just a couple of weeks out of that and most of those stocks have recovered. The SMH looks amazing. I think 86% of the names in the SMH semiconductor ETF are above their 50-day moving average. That didn't take long. So, I think it's really important when you hear people pounding on, you know, with a wooden spoon on a pan talking about bubble, ask yourself why are they doing it? Are they in the content business? Okay, they're doing it for attention. They're doing it for clicks. I get that. There's nothing wrong with that. Everyone's got to make a living. Are they money managers who are overweight small cap value, underweight tech? Okay, I get that, too. They're wish-casting. They want these stocks to blow up so they can call their clients and say, 'You see, I was right. I'm not a schmuck. I told you that. I told you so.' So there's a lot of that going on. Obviously people on financial television, it's high ratings when stocks like Oracle blow up. And then the next one and the next, they want that domino effect because this is when people are paying the maximum amount of attention to financial media. So there are a lot of people who have a vested interest in seeing this thing go badly and maybe it will, but I think it becomes really important to say, all right, who's talking about this that's actually invested? Who's talking about this that literally has money on the line, reputation on the line, that has skin in the game? I want to listen to those people. Not that they can't be wrong, but my god, I need a counterbalance to all the Chicken Littles and all the people saying the end is nigh because without that counterbalance, you're just listening to people who are wish-casting.

Yes.

My point is when in doubt, follow price.

Pay less attention to what people say and more attention to what they do.

Totally true.

I think you're exactly right. It's a very unique situation where one of the most important chess pieces is not on the board. It's a highly, highly unique situation. I've likened it in other venues. I've explained it as Keyser Söze in The Usual Suspects.

It's the main character in the movie. It's the thing that animates the actions of everyone else in the movie. And he doesn't show up until the last minute of the film. In old Hollywood, they used to call it the MacGuffin. So the Maltese Falcon. What is the Maltese Falcon? It's a stupid statue, but it's the MacGuffin. It's the thing that sets everything else in the movie in motion. OpenAI's share price is not tradable. Now, of course, there are people who are invested in it. But even if on a Monday they're bullish, on a Tuesday they're bearish, they don't have the liquidity to change their mind and act on that price in the way that we have in the public market. To which I would respond to you, I'd say you're exactly right. Thank God for public markets, number one. Because yes, there is a lot of concern about OpenAI and we saw it viscerally in the prices of the companies that were deemed to be quote OpenAI-dependent, companies that OpenAI is spending the most money on their services, and Oracle sort of became like an avatar for that. But let me ask you a question. Anthropic, one of the biggest players in AI, also private, spends about $5 billion annually that we know of on the AWS cloud to support its business. Anthropic is actually ahead of OpenAI in enterprise. They're selling to thousands of the largest organizations and corporations around the world. We don't have a share price for that one either. You think Anthropic is in as big of a drawdown from its high as the public share price of Oracle is? Probably not. Amazon actually consolidates some of the profit Anthropic is doing as a very large shareholder of the company. We think that business is gangbusters. So yes, you have to bucket these things, but then you also have to realize narratives in AI are shifting overnight. Seven months ago, the talk around Alphabet is that they're finished. They're done. They're going to get murdered by OpenAI and other LLMs. People are going to start their searches on Perplexity and Claude and ChatGPT and completely bypass the Google experience. And then Google goes up 65% on the year, becomes the best performing of the Mag 7, and all of a sudden it's Google's world and the rest of us live in it. That shift didn't take three years. It took like three months. So the narratives and the banging on pots and pans has to give way to the primacy of price. What are people actually betting on with real money versus who is wish-casting and just hoping the guy they don't like goes down for the count, right? Oh, I don't trust Sam or I hate Elon. It's so important to focus on price more than ever, not to try to sit here and parse 50 different people's opinions. That game is almost impossible. The real game is, okay, fine, but what are the buyers doing? What are the sellers doing?

So, we're money managers, not sell-side analysts and not chief strategists. So, we don't have price targets. We don't have our own internal earnings expectation. What we do for clients is we rationalize the risks they're taking with data. We try to give them the historical context that we think is more important than the prediction of the month or of the day these days because we're in the January where everyone's making their year-end predictions. But this is what I would share with you. The only questions that matter for this year. Number one, will the fundamentals continue to justify an above-average price-earnings multiple? That's it. That's the question. You either answer that yes, no, maybe, or let's say yes, no, I don't know. So, of course, like obviously I don't know, but I think the answer is yes. Why do I think the answer is yes? Well, the first earnings season we'll have is Q4. We'll start getting these reports in February. And here are the consensus estimates. You tell me what you think when I'm done. Wall Street thinks the S&P will deliver an 8.6% earnings growth quarter versus Q4 of 2024. Okay. They think the technology sector, which is obviously the most important sector for earnings growth and also for market capitalization, will do 25.8% earnings growth. That's an eye-popping number. If they do 24% or 27% or anywhere in that vicinity, wouldn't you agree that that would support a 21 multiple on the overall S&P given how important those companies are to market cap? That's what I would say. You're not getting that across the whole S&P. You're getting that in tech and a lot of that's coming from semiconductors. Okay, fine. But the number is the number. We don't put asterisks on any other year for the S&P 500. Why would we all of a sudden put an asterisk? Now, here are the full-year 2026 consensus estimates for sectors, the big sectors that matter. Wall Street thinks industrials will do 13.1% earnings growth. They think technology will do 29.7, 30% earnings growth for '26 again if they even get close. But here's what's really fascinating. Financials 10.4% for 2026. Consumer discretionary, huge sector. Amazon's in that sector. Tesla for some reason is in that sector, 11.7%. Communication services, that's Alphabet, Meta, Netflix, etc., 7.2%. So like if healthcare 8.5, utilities 8.7, that's become a de facto AI trade. So, the consensus thinks the S&P 500 will grow earnings once again for all of calendar 2026 by 14.6%. So, with a typical beat rate, it'll probably be more like 15, 16%. If you think the market delivers that, 21 times earnings is justifiable. And so it doesn't mean we'll get another, you know, 18, 20% return on the S&P, but at the very least, we should have a lot of winners in the market. 16 stocks of the S&P doubled last year. These aren't obscure, bizarre stocks no one's ever heard of. Some very big stocks doubled last year. All sorts of stocks went up 30, 40, 50, 60%. And I want you to remember the stock market wasn't cheap a year ago at this time. So the earnings growth has to deliver. So the real question that we need to answer is will it?

I think it will. And there's a chance that that's wrong. You could have some sort of an exogenous shock that changes everything. So, you have to just be like okay with that. That's the risk that we're all taking because just for fun, go read the January 2020 year-ahead outlooks from Wall Street. How many of them were talking about coronavirus? Zero. Hundreds of outlooks. Zero. And that obviously changed everything. Stock market ended up going up by the way. But I'm just making the point like the thing that could alter the course of 15% earnings growth this year is not really the thing that any of us could put our finger on. Now, a lot of people would say, 'Oh, it's going to be the AI data centers. They're going to slow down. There's going to be another Oracle, blah blah blah.' Okay, possible. But right now that's just not, I wouldn't say that's sort of like what should be anyone's base case that all of a sudden all this money that these companies have spent, they're going to say, you know what, forget it. It's just unlikely to me. So that's where I am on '26.