The AI House of Cards
When "Number Go Up" Becomes a Dangerous Game
Have you ever watched someone build an impossibly tall tower of playing cards, holding your breath as each new card is placed, knowing the whole thing could collapse at any moment? That's essentially what's happening in the artificial intelligence investment world right now—and the stakes are far higher than a card game.
Tech journalist Ed Zitron recently published a detailed investigation into what he calls "The Enshittifinancial Crisis"—a clever mashup of terms that essentially describes how the AI industry has created a financial house of cards that could topple and take a lot of regular investors down with it. I've always been quite skeptical of this AI bubble. This wonderful piece of investigative journalism has confirmed my hunches and it is far, far worse than a cynic like myself could have even imagined.
Ed's article is quite complex and very detailed. I'm going to try to break down what's happening in simpler terms, because this matters whether you own tech stocks directly, have them in your 401(k), or own the SPY ETF like most investors do.
On a personal note, I sense we may be approaching a season of economic turbulence—one that could bring real hardship to retirement accounts and household budgets, particularly if the AI bubble deflates anything like the dot-com crash did. Most believers I know have taken a hands-off and irresponsibly passive approach to managing their finances, perhaps assuming things will simply work out. I understand that impulse—life is busy, and financial markets feel complex and intimidating. But Scripture calls us to be faithful stewards, and stewardship requires attention and intentionality. Jesus spoke more about money than most other topics in Scripture, yet many of us spend our time scrolling through social media and debating pointless topics with people we will never meet. My hope in sharing this isn't to stoke fear, but to sound a gentle alarm: if you've been passive about understanding where your money is invested, now is the time to engage. Proverbs 27:12 reminds:
"the prudent sees danger and hides himself, but the simple go on and suffer for it."
You're reading this now—that's not an accident. Let it be the nudge toward wisdom and preparation that protects your family and positions you to be generous when others are struggling.
The Problem in a Nutshell
Here's the bottom line: virtually every AI company is losing money. Not just a little money—we're talking billions of dollars. OpenAI (the company behind ChatGPT), Anthropic, and dozens of smaller AI startups are all burning through cash faster than they can bring it in. The math simply doesn't work, and yet the investment dollars keep pouring in.
How bad is it? Consider this: OpenAI has committed to spending $22.4 billion with one data center company alone, $38 billion with Amazon, and $250 billion with Microsoft. Meanwhile, the company made only about $4.5 billion in revenue through September 2025. If your household budget looked like that, you'd be in serious trouble ... scratch that – you'd be bankrupt.
The "Letter of Intent" Magic Trick
One of the most troubling patterns Ed Zitron uncovered is how these companies announce massive deals that turn out to be... well, not quite real. When NVIDIA announced a "$100 billion partnership" with OpenAI, stocks jumped. The media treated it as a done deal. But buried in the fine print? It was just a "letter of intent"—essentially a statement that says "we might do business together someday, maybe."
When the actual earnings reports came out, there was no money changing hands. The deal didn't actually exist in any binding form. But retail investors who bought in based on the hype? They were left holding the bag.
This pattern repeated across multiple companies. AMD announced a big AI deal with OpenAI—stock surged 34%. A month later, their financial filings showed essentially zero revenue from the partnership. The stock has since dropped significantly. Anyone who bought in on the news got burned.
Where Is All This Money Actually Going?
Here's where things get really murky. Major tech companies—Microsoft, Amazon, Google, and Meta—have collectively spent over $776 billion in capital expenditures over three years, supposedly on AI infrastructure. But when you try to trace where that money actually went, the answers don't add up.
These companies don't tell investors how many AI chips they've bought, where they are, whether they're installed and running, or how much revenue they're generating. They've essentially asked investors to trust that all this spending will pay off eventually. Wall Street analysts, who should be asking tough questions, have largely gone along with the story because... well, stock prices keep going up.
It's a bit like a contractor who keeps asking for money to build your house but won't let you visit the construction site or give you receipts for materials. At some point, you'd start asking questions, right?
The Data Center Debt Bomb
Perhaps the most concerning piece of this puzzle is the explosion of debt being used to build AI data centers—the massive facilities that house the computers running AI systems.
In 2025 alone, there were at least $178.5 billion in data center loans. These facilities take one to three years to build and start generating any revenue. Many are being built based on contracts from companies that can't actually afford to pay for what they've ordered.
Here's a specific example: CoreWeave, one of the largest AI computing companies, has negotiated payment terms with OpenAI that allow OpenAI to wait up to a full year before paying its bills. Meanwhile, CoreWeave has $25 billion in debt and is losing money. If OpenAI decides it can't pay, CoreWeave could be in serious trouble—and so could the banks that loaned them money.
And those same banks—JP Morgan, Morgan Stanley, Goldman Sachs, and others—are involved in data center deals all across the industry. When one domino falls, it could knock over others.
Why This Matters for Regular Investors
You might be thinking, "I don't invest in AI startups, so why should I care?" Here's why:
Your retirement funds probably include these stocks. Major index funds contain companies like Microsoft, Google, Amazon, and Meta—all of which have bet heavily on AI.
Venture capital struggles affect the broader economy. When VCs can't get returns on their investments, they stop funding new companies, which means fewer jobs and less innovation.
Bank exposure could ripple outward. If data center loans start going bad, the banks holding that debt could face problems that affect lending throughout the economy. The Great Recession was a domino effect of bad loans and this could end up much the same or worse given the circular deals in the AI space.
The "Number Go Up" Trap
Zitron makes a compelling point about how Wall Street got into this mess: for years, tech stocks kept going up, so analysts stopped asking hard questions. If you're an analyst who says "this might be overvalued" while everyone else says "buy," you look foolish—until you don't.
This is exactly what happened before the dot-com crash in 2000. Companies were signing deals they couldn't fulfill, using accounting tricks to make numbers look better, and analysts were cheerleading the whole way. When the music stopped, regular investors lost trillions. It's incredible how similar this is to the dot com bust and both Wall Street and Main Street have their heads in the sand. Charles Mackay was spot on in his 1841 book, Extraordinary Popular Delusions and the Madness of Crowds:
"Men, it has been well said, think in herds; it will be seen that they go mad in herds, while they only recover their senses slowly, and one by one.
A Sober Assessment
None of this means AI technology is worthless or that every tech company will collapse. Microsoft, Google, Amazon, and Meta all have profitable core businesses that continue to make money regardless of AI's success.
But it does mean that the current valuations—the prices people are paying for these stocks based on AI growth expectations—may be dramatically inflated. If and when reality catches up with the hype, there could be a significant correction. And correction is probably an understatement.
The wise investor takes this information not as a reason to panic, but as a reason to be thoughtful. Diversification matters. Understanding what you actually own matters. And being skeptical of headlines announcing massive deals that turn out to be "letters of intent" really matters.
As the old saying goes: if something sounds too good to be true, it usually is. The AI boom has generated incredible excitement and very real technological advances. But the financial structures built around it look increasingly fragile. And when houses of cards fall, it's rarely the builders who get hurt—it's the people who bought in without looking too closely.
