Big Tech's Capex Is Half of Wall Street's Profit Growth

12 points by speckx 2 hours ago on hackernews | 17 comments

jqpabc123 | 2 hours ago

in other words, the entire stock market is now being driven by "Enron" accounting where nothing is really as it appears.

ralph84 | 2 hours ago

Enron was massive accounting fraud. Extraordinary claims require extraordinary evidence. Where is your evidence of accounting fraud? "Lots of companies investing lots of money into AI and it might not work out" isn't fraud.

toomuchtodo | 2 hours ago

The AI Economy (Part 2) - What Enron Can Teach Investors About the AI Boom - https://michiganfinancial.com/blog/the-ai-economy-part-2-wha... - August 4th, 2026

> Enron is often remembered simply as one of the largest corporate frauds in American history. While this is true, there are lessons to be learned from the way investors acted during this time. At its peak, Enron was viewed as one of America's most innovative companies. Investors believed it had fundamentally transformed the energy business through sophisticated trading platforms, financial engineering, and technology-driven markets. Its stock price reflected extraordinary confidence in the company's future rather than its demonstrated earning power.

> Part of that confidence was fueled by Enron's use of mark-to-market accounting, which allowed the company to recognize estimated profits from long-term contracts immediately instead of waiting for those profits to actually be earned. Future expectations effectively became present-day earnings. Those projections were wildly "optimistic". When investors realized the expected profits were unlikely to materialize, confidence disappeared almost overnight. Between August 2000 and November 2001, Enron's stock collapsed from roughly $90 per share to just pennies, erasing billions of dollars of wealth and devastating employees and retirement investors alike.

Draw your attention to:

> Despite the advancements created through use of AI and the increasing numbers of applications, investors still face an important question: How much of today's valuation reflects profits that have actually been earned, and how much reflects profits investors hope will eventually exist?

and

> Artificial intelligence represents one of the most significant technological developments of our generation. Its long-term impact may ultimately justify much of today's enthusiasm in the market. History, however, reminds us that markets frequently price revolutionary technologies well before their economic potential is fully realized and fluctuations occur in even the most stable sectors.

> The lesson from Enron is not that today's AI leaders are engaging in similar behavior. It is that investors should remain cautious whenever valuations become increasingly dependent on future expectations rather than demonstrated earning power.

TLDR https://en.wikipedia.org/wiki/Reflexivity_(social_theory)

Anthropic's IPO prospectus illustrates this fairly well.

Anthropic's IPO Prospectus Is a Fucking Doozy - https://news.ycombinator.com/item?id=49914149 - September 2026

Anthropic's IPO prospectus shows AI vision, surging costs - https://news.ycombinator.com/item?id=49886005 - September 2026

> Matt Levine's summary of Sam Altman's (and most of AI industry's) business plan remains accurate: "We will create God and then ask it for money."

WarmWash | 2 hours ago

OK, but where is the AI fraud part?

MTM accounting isn't fraudulent. Investors knew Enron was doing it. The fraud part was when those projected profits fell apart, Enron would hide the losses.

AI having bonkers valuations isn't fraudulent, maybe delusional, but not fraud. Fraud is the act of outright lying to investors about things that are factually true (not hypothetically true).

"We will have trillions in profits" = not fraud, forecasts can be anything

"We have trillions in profits" = fraud, they don't and they are lying.

toomuchtodo | 2 hours ago

You're focusing on the wrong part in the example. Enron's valuation was driven by irrational exuberance and market sentiment. In the case of Enron, the numbers were a fraud. In the case of AI, the numbers show utter carnage but everyone is trading on hope like it's musical chairs. There simply will never be enough revenue to justify to date and forward looking capex. At some point, investors are going to figure out they've been sold a bag of beans.

> "We will have trillions in profits" = not fraud, forecasts can be anything

The forecasts are dishonest and not grounded in reality. I might as well tell you my magic startup is going to consume total global TAM of all human economic activity. Sure it is. If we're going to excuse fraud as forecasts when it's clear the forecasts are simply fever dreams, we agree to disagree.

AI needs $6T in annual revenue to justify data centre boom - https://news.ycombinator.com/item?id=49898952 - September 2026

From the above thread: "To put this into perspective, healthcare, real estate and banking each earn ~$1.5T in revenue annually. This is REVENUE, not profit which is obviously lower than the revenue in every industry. I fail to understand how AI can achieve $6T in annual revenue when the TAM for healthcare and banking is a quarter of that in the US."

You might as well argue you can break the speed of light and control gravity if facts no longer matter. Are those forecasts made in good faith?

Additional citation:

"AI is a multipole Enron" | David Gerard - https://www.youtube.com/watch?v=Iur27O732w8 ("Will Guyatt from The Tech Report talks to David Gerard from Pivot-to-AI about Oracle referring to a gas-pipeline delay at a mega data centre as an act of God, and why the business of building AI infrastructure increasingly looks like a house of cards.")

fuzzfactor | 19 minutes ago

In my other comment I mentioned "bottled wealth", and they never have issued engraved certificates for this even when it grows more massive than the underlying shares.

You just can't pinpoint who controls this non-ledger value at any one time.

But I would estimate it's owned by a player known as "sentiment."

ralph84 | 2 hours ago

> valuations become increasingly dependent on future expectations

Valuations are always dependent on future expectations. What else would they be based on? That's one of the main functions of a stock market: providing price discovery via competing buyers and sellers placing bets on the expected future economic prospects of companies. Of course fraud interferes with price discovery, but again, where is the evidence of fraud?

fuzzfactor | 41 minutes ago

Enron functioned remarkably well regardless of fraud, in the same financial/stock market and accounting "standards" as everyone else.

The rules are still evolving but have been a cat-and-mouse effort for more than a century.

>Valuations are always dependent on future expectations.

This is the point. The market is a very responsive organism. It is not easy for a group to reach the point where it can evolve faster than the market can adapt. For some though, this has always been the holy grail. Rarely does it work. When it does some people do get much richer much faster than otherwise, but by comparison the market is so huge it's still usually not a significant impact on the overall market.

Future expectations can have a tendency to become bigger than the "whole" market though. Even if not that massive it still represents more paper wealth that could be more quickly extracted than the same companies will be able to contribute in earnings over the same period.

You could say extracting this was built into the system to allow this type of "bottled wealth" to flow, or alternatively, a much wilder version has been reigned in to arrive at what we have now, in which Enron itself has been taken into consideration as much as it could be at the time.

There really shouldn't be much problem unless a highly leveraged entity gets so big so fast that it is no longer insignificant relative to the market niche it operates in, or even the whole market. If it's also "evolving" faster or more creatively than the market that's something too.

In that equation fraud does not need to exist, and is not necessarily a prominent term.

Nobody at Enron tried to defraud me even though in the end they failed to pay thousands in invoices. A single company built from a house of cards shows the textbook example of how outgrowing the ability of future expectations to deliver can cause bigger companies to collapse further and more abruptly than otherwise.

It would have been much worse if I had been a shareholder. But I was a valued contractor and everyone I had contact with was exceptional, sometimes the smartest person in the room.

toomuchtodo | 2 hours ago

It'll be interesting to see what happens to US equities, driven by the Mag 7, when the bottom falls out and this capex investment ends. There simply isn't enough capital for this to continue forever, but it can continue for a while longer.

Joel_Mckay | 2 hours ago

There are still good opportunities to invest in firms with legitimate revenue flows, but avoiding the consolidated risk exposure is difficult.

The problem is the unsecured debt loads, fictional market projections, and secondary impact on real firms supplying products, services, and financing.

A good posture, is figuring out which firms may profit from the fire sales. The hard part for Bears and Bulls alike is they are both usually wrong about when a correction event happens (if they are honest players.) =3

https://www.youtube.com/watch?v=rE75WvOtcu8

downrightmike | 2 hours ago

The problem is contagion, those really bad firms will absolutely wreck everyone else
> The English draft accurately matches the Korean source with correct numbers, headings, footnotes, and links. No Hangul remains. No corrections needed.

Good to hear!

CuriouslyC | 2 hours ago

LOLClaude.

yathern | 2 hours ago

My goodness. It's the first sentence of the article even. At least it's only the third post from the top of HN right now. I guess favorably we can assume a human wrote the Korean version and just translated it via ChatGPT.

demibabs | 2 hours ago

LOL

graypegg | 2 hours ago

The idea of the author assuring you that "no corrections are needed" is immensely funny haha. If the author is around this thread, it's all good to translate content dude, and thanks for doing so! But you definitely should translate the output back to your native language and give it a read as a sanity check.

Archive link for context in the future: https://archive.is/g6Znd

simpaticoder | 2 hours ago

Unless something big changes we're heading toward a period of unprecedented capital concentration, even more than we have now which is extreme. There will be a few winners out of the current frontrunners who own all non-physical human-like labor. Owning a few tens of shares in those companies will be generational wealth, so it makes sense more money is flowing into the stock market and into tech stocks in particular.