The AI Demand Bubble with Ed Elson

**Ed Zitron** (0:02)
Call Zone Media Greetings and salutations, it's me, Ed Zitron, and it's Better Offline.
That's right, we are back with Prof G Markets' Ed Elson, and we're here to talk about the naughty little buggers in the Magnificent 7, primarily the big four, but really the big three, the Amazons, the Googles, and the Microsofts of the world. Ed, thank you for joining me again.

**Ed Elson** (0:36)
Thank you so much for having me, I'm very, very excited to be here.

**Ed Zitron** (0:39)
Yeah, so tell me, they just reported earnings, and as you well know from the conversation we just had, I got some points to make about it, but why don't you run me through how they did on their earnings?

**Ed Elson** (0:51)
Absolutely, so let's just run through what we learned. On Microsoft's end, we saw pretty good growth, $90 billion in revenue, up 18 percent. Amazon, similar story, $200 billion in revenue, up 20 percent. Meta's revenue increased 28 percent. I mean, on its surface, this was pretty good. Just look at the numbers, even the net income, which exploded for many of these companies. But of course, that's not really what matters here. What really matters is how is the AI trade actually working out? There are two main questions that we really want to understand when we look at these big tech companies. The first question is, how much revenue did you generate from your AI business specifically?
That's what we really want to know the answer to. And then the second question is, how much of that revenue came from two relatively unstable companies, financially, specifically OpenAI and Anthropic? Those are the two questions I want to know the answer to, I know you want to know the answer to. We didn't get those answers. They didn't really tell us.
And what they did tell us was not very helpful kind of papering over the reality, which makes you think maybe the reality is something that they don't really want us to see.

**Ed Zitron** (2:11)
So that's the funny thing as well.
Before we get to the meat of what you and I both know we're going to discuss, there was also these weird massive equity gains. So the net income part of earnings is now not super useful, because in all three cases of Amazon, Google and Microsoft, they all posted massive equity returns based on the increasing value of OpenAI and their shares of either or both OpenAI and Anthropic, right?

**Ed Elson** (2:39)
Yes. Yeah, a wrinkle in there that I'll address, but we'll start with Amazon and Google. The answer to that question is yes. Amazon's net income, so their earnings, exploded 245%. That's crazy growth, more than tripled. Google's exploded 298%.
Again, crazy growth. How did that happen? What's going so right?

**Ed Zitron** (3:01)
What could possibly-

**Ed Elson** (3:01)
You dig into it.
Exactly. The answer, in Amazon's case, is that 85% of their net income came from unrealized gains in their stakes in Anthropic and OpenAI. So essentially, they basically looked at that company. They said, okay, I guess it's worth this amount of money. I guess it's worth a trillion dollars. And they marked that up, and then that was reflected in their earnings. And then in Google's case, 87% of their earnings came from their unrealized gains in their stakes in Anthropic and SpaceX. So those earnings of those companies, you can't really take seriously because they're not reflecting the fundamental underlying health of the business. They're reflecting the optimism associated with the valuations of these AI startups and these frontier labs. So that's the first problem is the earnings themselves are not really reliable. In Amazon and Google's case, we'll get to Microsoft in a second.
The other problem is that that then funnels down into how we value companies. Specifically, one of the most important valuation metrics in investing is the price to earnings multiple. That is how we figure out how expensive or cheap a company really is. We look at their earnings, then we look at the price of the stock, and we compare if the two match up. Now, as of the end of last week, which is when I did this analysis, Amazon was trading at 19 times earnings. Google was trading at 17 times earnings. The S&P, the overall market, currently trades at around 25 times earnings. So if you were to look at those multiples on Amazon and Google, you'd say, oh my gosh, these companies are crazy cheap. It is time to buy. But of course, we know that actually, no, those multiples have been massively skewed and compromised again, by those unrealized gains, by those paper gains.