Everyone Is Asking If the AI Bubble Will Burst. That’s the Wrong Question.
Listen to the audio version of this article (generated by AI).
The average American is in serious trouble.
Inflation is rising. Wages are falling. Real wage growth is running negative. Yet consumers are still spending — by raiding their savings — which is dragging the savings rate toward record lows. If those savings run dry before inflation is fixed, consumers will literally run out of money to spend.
All the while, more AI-driven layoff announcements are hitting the tape every week, threatening the very labor market providing Americans’ paychecks.
This situation is unsustainable.
If you’re feeling squeezed right now, you’re not imagining it. The data backs you up completely.
But history also shows that the periods of greatest economic stress for Main Street have often coincided with the most aggressive bull runs on Wall Street. (Including, potentially, the two largest IPOs ever attempted… which could arrive sooner than the market is pricing in.)
The next two years may be no different.
There’s a pattern here… and it’s not the one most people expect.
The Inflation Problem Is Worse Than the Headlines Are Telling You
Let’s start with the inflation data, which is quite alarming.
The personal consumption expenditures (PCE) inflation rate — the Federal Reserve’s preferred gauge — hit 3.8% in April 2026. That’s the highest reading since May 2023, the highest since August 2008 if we exclude the COVID-era distortion, and nearly double the average PCE inflation rate (~2%) going all the way back to 1992. The Cleveland Fed’s Nowcast model is already projecting that number rises further in May, potentially reaching 4%.
Core PCE inflation, which excludes food and energy, hit 3.3% in April. That’s its highest since November 2023… since April 1992, excluding COVID… and roughly 65% above its long-run average of ~2%. The Cleveland Fed Nowcast sees that number climbing to ~3.4% in May.
So inflation is hot — very hot — and getting hotter.
The Wage Problem Makes the Inflation Problem Even Worse
The inflation picture is bad. The income picture is worse.
Nominal personal income growth clocked in at just 2.5% in April, down sharply from 5.5% a year ago. That means wages are decelerating while inflation is accelerating. When you do the math — income growth of 2.5% minus PCE inflation of 3.8% — you get a real wage growth rate of negative 1.3%. Consumers are falling behind by over a point a month, in real terms.
Based on current estimates, that number likely deteriorates further in May, potentially hitting negative 1.5% or worse.
One year ago, the consumer picture was very different. Nominal income growth was running at 5%, inflation was 2.6%, and real wage growth was a healthy positive 2.4%.
Today, real income growth is deeply negative — and as we’re about to see, consumers are spending at the exact same pace anyway.
The Spending Paradox: Americans Are Broke and Spending More Than Ever
Nominal personal spending growth rose in April from 5.7% to 5.9% — the strongest reading since January 2025.
To be clear: that is the same spending rate we saw back when real income growth was positive 2.4%. Consumers haven’t pulled back at all. If anything, they’re spending more. But the fundamental support for that spending has completely collapsed underneath them.
So if wages aren’t funding this spending binge, what is?
Savings.
Savings Cliff: The Only Real Historical Analog Is 2008
The U.S. personal savings rate crashed to 2.6% in April 2026 — one of its lowest levels in modern history. The only comparable readings are the roughly 2% savings rate we saw in the year leading up to the 2008 financial crisis and the brief dip to ~2% in 2022 — which was less impactful because consumers had just exited 2021 with savings rates north of 20%, flush with COVID-era stimulus cash.
Today’s situation has no such cushion. Just a year ago, the personal savings rate was 5.5% — more than double where it sits today. In other words, U.S. consumers have burned through more than half their financial buffer in 12 months, during a period when real wages are going negative and getting worse every single month.
It seems the only honest historical analog for what’s happening right now is the run-up to the 2008 financial crisis.
Three Ways This Ends — and Why Only One Is Realistic
This is not a sustainable dynamic. Something has to give… and soon. The way I see it, there are three possible outcomes:
Option 1: The inflation problem gets fixed. Oil falls, PCE cools, real wages go positive again. Consumer crisis averted.
Option 2: Wages surge unexpectedly. Labor regains leverage, income growth reaccelerates, consumers get back ahead of inflation.
Option 3: Consumers hit a brick wall. Savings run dry. Spending collapses.
But unfortunately, options 1 and 2 are increasingly unlikely in the near term.
Even with the U.S.-Iran peace deal reportedly in progress, oil is still sitting in the $90–$100 range — which, all else equal, should keep PCE inflation north of 4% for the foreseeable future. That means the inflation problem isn’t getting fixed anytime soon.
And on the wage front? Good luck negotiating a raise when your boss just announced he’s laying off 20% of the company because of AI. Wix just became the latest CEO to join that chorus. When employers have AI as a credible replacement threat, workers lose all bargaining leverage. Wages aren’t going up from here.
That leaves us with Option 3.
The Prediction: A Consumer Brick Wall, a Political Revolt, and a March 2000 Moment for AI
Here’s how I think this plays out over the next few years.
Over the next 12 months, the consumer situation goes from bad to worse. Sometime in 2027, they hit the wall. Spending collapses. Consumer confidence craters. And since consumer spending still drives roughly 70% of U.S. GDP, the broader economy starts to crack. Companies respond to slowing revenue by accelerating AI-driven layoffs to protect margins. More people lose jobs. The savings rate hits zero. Financial stress becomes widespread.
And then the politics get ugly.
When people lose jobs, can’t pay bills, and watch a handful of billionaires get exponentially richer from the same technology displacing them, the backlash is inevitable. AI will become the face of economic inequality in a way that no amount of industry PR will be able to neutralize.
A populist, bipartisan anti-AI movement takes shape in late 2027 and gains serious momentum heading into the 2028 elections. Politicians on both sides of the aisle adopt anti-AI stances to acquire or keep power. And by late 2028 or early 2029, those politicians follow through with legislation — whether that’s a tax on AI profits, a forced cap on hyperscaler capex, mandatory review processes for new AI models, or a mountain of red tape that collectively slows the AI build-out.
That’s the March 2000 moment for the AI bull market — the point at which the Nasdaq peaked, the dot-com bubble burst, and a generation of investors who had ignored every warning sign watched years of gains evaporate in months. It didn’t matter that the internet was real. What mattered was that the valuations had run so far ahead of reality that any catalyst — regulatory, economic, or political — was enough to break the spell.
The Twist
But there’s a bullish twist here that makes this a trade, not just a warning.
Wall Street doesn’t stop rallying when the ominous signs appear. It rallies harder. The best part of every major bull market has always been the final chapter, when fundamentals become irrelevant, greed overwhelms caution, and prices defy all logic. I expect that to be especially true this time — because I think a significant portion of the so-called smart money already sees the expiration date coming.
Look at the behavior: The tech elite are aligning themselves with political power to protect their opportunity. SpaceX, OpenAI, and Anthropic are all rushing IPOs at historic valuations while the window is open. The hyperscalers are spending at a pace that only makes sense if they already know what’s coming.
The motto of the next two years? Make as much as you can, as fast as you can.
And that means the bid in AI stocks will be relentless. The FOMO will dwarf 1999. The money will keep flowing. And AI stocks could soar more than anyone expects over the next 12 to 24 months — before the inevitable reckoning.
The Bottom Line: The Final Leg Runs. Position for Both Sides of It.
Ordinary Americans are getting squeezed, the safety net is fraying, and the technology driving it all is about to make a small group of investors extraordinarily wealthy one last time before the backlash arrives.
You can have opinions about whether that’s right. You can also have a portfolio that reflects the reality of it.
The final leg runs. The reckoning follows.
Position yourself for both.
What does positioning yourself for the final leg actually look like?
Start with the IPOs.
I’ve already mentioned that OpenAI and Anthropic are rushing to go public while the window is open. What I haven’t shared yet is that buying either company on IPO day is probably the wrong trade — even if the timing is perfect.
The biggest gains from landmark technology IPOs have almost never gone to the investors who bought on day one. They’ve gone to the investors who owned the ecosystem around those companies before Wall Street showed up to reprice it.
That’s exactly the opportunity I’ve been preparing for. I call it the Pre-IPO Backdoor — a small group of publicly traded companies that supply, power, and benefit from both OpenAI and Anthropic, and that I believe will get significantly repriced the moment those S-1 filings land.
The window to get in ahead of that repricing is open right now. I don’t know how much longer it stays that way.
Here’s everything I’ve found — and the stocks I think you need to own before the IPOs arrive.
