AI's gains come from AI's losses, not from customers

Investment Analysis Clubs Macro Economic Trends and Risks
1

“Cash flow is King.” That was the motto of Shanti Mehta, who was the CFO of W.L. Gore & Associates. I never forgot it.

Sales, capital expenditures and debt among the developers and suppliers of AI are huge and complex. But the cash flow we need to watch is the sales to end users who are outside the AI ecosystem.

https://www.wsj.com/tech/ai/big-techs-soaring-profits-have-an-ugly-underside-openais-losses-fe7e3184?mod=hp_lead_pos3

Big Tech’s Soaring Profits Have an Ugly Underside: OpenAI’s Losses

What’s only starting to become clear is that AI startups are also sinkholes for losses

By James Mackintosh, The Wall Street Journal

Investors take a lot of comfort from the solidity of Big Tech earnings as worries grow about artificial intelligence overinflating valuations. But those earnings have an ugly underbelly: ever-bigger losses at the generative AI startups that spend big on chips and data centers supplied by the profitable public companies….

But much of the AI-related profits come from being a supplier to, or investor in, the private companies building the large language models behind AI chatbots—and they’re losing money as fast as they can raise it, and plan to keep on doing so for years….

What’s only starting to become clear is that the companies are also sinkholes for AI losses that are the flip side of chunks of the public-company profits….

It’s impossible to quantify how much cash flowed from OpenAI to big tech companies. But OpenAI’s loss in the quarter equates to 65% of the rise in underlying earnings—before interest, tax, depreciation and amortization—of Microsoft, Nvidia, Alphabet, Amazon and Meta together….

In the frenzy, investors have forgotten the old saying: Revenue is vanity, profits are sanity, cash is reality. If investors stop being so excited about AI, if OpenAI struggles to generate sales, or if fundraising becomes difficult for other reasons such as a recession, investors might switch back from the vanity of revenue to focus on the insane losses. … [end quote]

The article is loaded with details of how OpenAI is spending on products and services from the bigger companies, which then report the spending as profits.

The numbers are huge. The debts will be huge.

The more I read about this situation the more I’m reminded of the 2000 internet stock crash. Except more so.

Wendy

12 Likes
2

Big waste of capital.

Pop goes the weasel.

On tap, a great depression.

He has no clue what he is doing.

3

The global search engine market was valued at approximately $252.5 billion in 2025 and is projected to grow to $440.6 billion by 2030 . This growth is driven by the shift towards AI-powered, conversational, and multimodal search experiences, along with increasing enterprise demand for vertical search solutions.

Search engines will probably profit from AI.

4

I keep asking, "Which companies will be able to monetize AI? Please reveal your picks.

The Captain

1 Like
5

I’d have to say that’s almost in the “unknowable” column at this point. Maybe it will help with specific industries like legal and medical. Maybe it will improve factory flow. Maybe it will control robots in nursing homes.

Possibly it will help develop new drugs or make architecture more efficient. Maybe it can sift through x-rays and find tumors with better precision or maybe not. Nobody knows yet, save a few paltry trial projects, but everybody (well, most everybody) is hyped up about how great it’s going to be. And maybe it is.

I opened a small short on Nvidia today. They seem to have a great product and are one of the few in the field producing profits, but it’s hard to square that with their ratio at least in the short-term. They have almost every storybook positive, but I believe the market is due for an overall correction in which they will participate. I think they’re capacity constrained and locked out of a couple of lucrative marketplaces.

I was about to open a similar small short on Palantir but the clock ran out on me, I’ll do that one tomorrow.

In for a penny, in for a pound and all that.

3 Likes
6

First tell us your “bounce back” companies.

I’ll help.

  1. Tesla
  2. you go here
  3. and here
  4. …
  5. .. keep going, you’re doing great!

Shall I cite more?

2 Likes
7

Meta is monetizing AI through Advantage+. Using AI as a targeted ad booster is boosting ad revenue.

Other companies are monetizing AI through paid subscriptions and other silly ways.

I don’t think it’s a matter of whether companies will be able to monetize AI, but rather whether they will be able to monetize it enough to justify their ridiculous investments.

10 Likes
8

Absolutely yes! That is actually I meant.

The Captain

3 Likes
9

As one who spent almost his entire life in ad supported media, I am always interested in watching how this plays out. Indeed, when asked if I want to “opt out” I say no, because I want to see what happens.

And I use Facebook (among others) every day, and I have to say the ads that I am being fed on that platforms are no better, and are arguably worse for targeting me. I also have the TV blaring when I am in the workshop (a few hours each day) so I get to see a fair amount of “that” advertising, and my wife insists on listening to radio in the car, so I am exposed to “that” as well.

If Facebook is using AI (or rather, if its advertisers are using it) to target ads so it doesn’t waste a lot of money showing ads that don’t interest me at all, and worse, have no hope of ever interesting me.

(Yeah, I know a bit about targeting advertising, including product segment research, demographics, psychographics, and other practicers of the art.)

[/Rant]

5 Likes
10

Me too. Some recent fails show that Meta’s AI ad tweaks might be driving the wrong kind of interest.

1 Like
11

On the face of it, a company that can’t make enough product to meet all the growing customer demand doesn’t sound like a good short candidate.

That said, as a long-term NVDA holder I’ve sat through three 50% declines in stock price.

DB2

2 Likes
12

I’ve noticed that when I research buying an item the ads for that item increase. That’s logical. What I don’t understand is why the ads continue after I buy said item online. Surely they monitor my purchases.

3 Likes
13

A few years back i noticed that for some insane reason (mountaineering gear?) the ad algorithms aimed farm/barn equipment at me…. So I bought men’s underwear online and did lots of searching on speedos, and YES, the ad algorithms pushed wall to wall gorgeous boys on me for a couple months. Then it all stopped, even when i or friends used other identities. Someone taught the algorithms about visual sex. LOL.

1 Like
14

Yes, but you don’t need AI for that. They’ve been doing that for years, and that’s simple algorithmic. Supposedly the ads are supposed to be even more efficient, even more targeted, blah blah blah.

If so I sure haven’t seen it.

15

Yes and no meta ads are not granular when you advertize to huge numbers of people. Yet granularity costs money. Meaning most people have soft interests.

16

Like teddy bears?

“Let me tell you, safety first, little buddy. Matches are for grown-ups to use carefully. Here’s how they do it,” Kumma began, before listing the steps. “Blow it out when done. Puff, like a birthday candle.”

3 Likes
17

NPR has a Spanish culture show that currently is very focused on immigration. For most Americans no matter which side of the “drama” it is a soft interest.

My interest in Israel for most other people is a soft interest.

Drama abounds.

The economy is the hard interest.

Teddy bears do not rank with voters. But they are everyone’s soft interest. One of mi girl friends used to call me her teddy bear…that was a…Sex is the other hard interest.