A Convergence Of Risks

First, an update from the last blog post...

We have entered a lethal Twilight Zone between stock market rally and stock market crash. In this dubious gray zone of incipient market disintegration lives the most insidious lies that we have seen since this AI super bubble began. And the lies are large for a reason - because there is now blood in the water, hence the bagholders at large need to be convinced that this bubble is not a bubble. Or, if it is a bubble, it will continue indefinitely. For "years and years" we are told, even though it has been going on for years already. Hence, I find these types of lies the most egregious because there are already ample warning signs to take down risk.

One of the most commonly circulated lies is that the companies involved in this AI bubble are highly profitable unlike the companies involved in the DotCom bubble. The people advancing this lie either have forgotten the Y2K bubble or they were in elementary school at the time. The two bubbles are actually very similar as I showed on Twitter. During the Y2K bubble, investors first flocked to the "DotCom" stocks because they had explosive growth rates in page views aka. "eyeballs", despite being highly unprofitable. Unfortunately, the business model of selling dollar bills for 50 cents began to come into question, so investors started dumping DotComs to buy the "high quality" Tech names. The problem was that the high quality stocks became massively overvalued as well because they were primarily selling to the unprofitable DotCom companies that were imploding left and right. So it is that when the bubble imploded, Microsoft collapsed -70%. Oracle -80%. Cisco -80%. No big Tech stock was spared. The Nasdaq 100 index of largest cap stocks lost -80% and took 15 years to recover to the old high.

In a deja vu of Y2K, what we are seeing over and over again now are mega cap vendors announcing massive deals with unprofitable partners, that are financed by the vendors themselves. We saw another one just today. Essentially, Microsoft is buying cloud hosting services from Microsoft:

The problem with AI is that there is as yet no timeline for when it will actually make money. And ironically any attempt to reduce depreciation expense will only serve to LENGTHEN the time it takes to become profitable. It's a catch-22 - stay on the efficient frontier of the latest hardware and face crippling depreciation costs OR depreciate more slowly and fall behind everyone else in the AI arms race. As I show below, the magical and as yet totally undefined AGI (Artificial General Intelligence) level aka. Human-level intelligence, is only the first step on the timeline towards monetization. First we will need apps and robots that harness AGI by replacing all of humanity in the workforce while causing GDP to collapse like a cheap tent.

Next on the list of warnings, last week we got news that Michael Burry of Big Short fame was inexplicably shutting down his hedge fund just a week after he had come out as short on Nvidia and Palantir and only days after he had asserted that Meta and Oracle are "cooking the AI books" by under-reporting depreciation. Therefore it was confusing as to why he shut down his hedge fund, which gave bulls a victory lap on the news. Subsequently it's come out through the Wall Street grapevine that he shut down his hedge fund so he can focus on his personal "family office" portfolio and thereby become fully focused on the AI Big Short 2.0 without the constraints of impatient investors. Far from victory for bulls, this is the opposite - hardened conviction by the world's most famous short seller:

"His chart serves as a warning that stocks may have already topped out even as OpenAI, Meta, and other AI giants prepare to invest trillions of dollars in microchips and data centers as they jostle for position in the nascent AI market."

Next on the list of risks and warnings, this week we got news from the world's largest bond investor, Jeff Gundlach, repeating Jamie Dimon's warning by saying that markets are rife with what he calls "garbage lending":

“The health of the equity market in the United States, it’s among the least healthy in my entire career,”

The veteran debt investor is concerned the $1.7 trillion private credit market is engaging in “garbage lending” that could tip global markets into their next meltdown."

Recall that "garbage lending" went into overdrive in 2025 and even caused two banks to nearly implode back in October. "NDFIs" are Non-Depository Financial Institutions that arose after 2008 to fill the void of low quality lending that had been filled by banks up until the financial crisis. NDFIs are essentially hedge funds that borrow from banks and then turn around and lend to subprime corporate borrowers while everyone pretends that the bank loans are AAA rated.

So put it all together and you have a Y2K-style DotCom stock market bubble at the same time as there is a GFC-style subprime credit bubble. And we are all just waiting for the Minsky Moment to arrive unexpectedly out of nowhere.

And end a six year bubble that might go on for "years", or already be ending.

No one seems to know.