In 1865, a young English economist named William Stanley Jevons published a book that made him famous for being wrong.
Britain’s engineers had just devised a way to make steam engines dramatically more efficient. As a result, the country would burn less coal.
Jevons, however, argued, “It is wholly a confusion of ideas to suppose that the economical use of fuel is equivalent to a diminished consumption. The very contrary is the truth.”
He was right.
Cheaper steam power didn’t shrink coal demand so much as it detonated it. Efficiency made the resource so useful that consumption exploded. Economists have called it the Jevons paradox ever since.
Why am I telling you this while oil is pushing toward $90, and Walmart (WMT) just painted a dour picture of the American consumer?
Because 161 years later, the same confusion of ideas is playing out in AI. The investors who see through it are the ones positioned for the next leg higher.
Let’s get into it.
First, the Ugly Tape
President Trump declared “Economic D-Day” against Iran, threatening secondary sanctions on anyone doing business with Tehran.
Oil popped. Treasury yields spiked, fully unwinding yesterday’s “Bessent put” relief in a single session.
And Walmart delivered a soft quarter, as U.S. comparable sales grew just 2.6% versus roughly 3.8% expected, with average ticket up only 1.1% as gasoline prices squeeze household budgets.
That’s a pressured consumer. And it’s why the early August recovery rally in AI stocks has somewhat flatlined.
But here’s what the recent selloff does not tell you — and this is the part most investors are getting wrong.
Bears have been passing around a chart showing AI token prices collapsing — round-tripping from a May peak of 2.07 all the way back to 1.02, essentially flat from where they started last December. If falling prices were the whole story, you’d conclude AI demand is drying up.
Now look at what the price line hides.
Over those same nine months, the underlying cost to rent H100 compute rose 27%, from $2.00 to $2.53 per hour — capacity didn’t get cheaper but scarcer. And token consumption went vertical: OpenRouter volume surged from roughly 5 trillion tokens per week last November to an estimated 31 to 33 trillion this summer. Google disclosed at I/O that it now processes more than 3.2 quadrillion tokens per month — up 7x year over year.
Flat retail prices on rising wholesale costs and exploding volume isn’t weak demand. It’s deflation by engineering — labs passing efficiency gains through to customers, who respond by using vastly more.
That’s Jevons, playing out in real time. And OpenRouter is just a 1-2% sliver of a global token market Goldman Sachs projects will hit 120 quadrillion tokens per month by 2030.
Walmart Shoppers Don’t Buy GPUs
Walmart’s report tells us plenty about gasoline-sensitive households trading down. It tells us nothing about hyperscaler capex, data-center backlogs, or optical networking demand. Microsoft (MSFT), Amazon (AMZN), Alphabet (GOOGL), and Meta (META) are not adjusting multi-hundred-billion-dollar AI budgets based on same-store sales trends.
Today’s AI weakness is a sympathy selloff — sentiment, not fundamentals. And sympathy selloffs are where opportunity lives.
The portfolio news backs that up. Micron Technology Inc. (MU) is committing $10 billion to a new research institution spanning next-generation chip manufacturing and memory architectures. CoreWeave Inc. (CRWV) signed a multi-year cloud deal with Hudson River Trading built on Vera Rubin and HGX B200 systems. Corning Inc. (GLW) expanded its fiber supply agreement with Zayo through 2030 as AI network buildouts accelerate. Enovix Corp. (ENVX) is doubling drone-battery capacity in South Korea on a pipeline that grew 41% in a quarter. And Palo Alto Networks Inc. (PANW) inked a $1 billion alliance with NTT DATA spanning AI governance and security.
We like all, alongside Alphabet, Meta, and Microsoft, which we still like even as Australia’s new news-payment levy grabs headlines.
Owning the Right Names Matters More Than Ever Today
Expect volatility to track the news cycle out of Tehran and Washington until a Middle East off-ramp materializes… and we think one is more likely than not in the coming weeks. But the catalyst that reasserts fundamentals over noise arrives next week, when Nvidia reports earnings.
In 1865, efficiency didn’t kill coal. In 2026, cheap tokens aren’t killing AI. They’re feeding it.
What the recent selloff really proves is knowing that AI fundamentals remain intact isn’t enough on its own. Days like this are also a reminder that owning the right AI companies is only half the battle. How much you own — and what you own alongside them — matters just as much.
That’s something Louis, Eric, and I have been working on behind the scenes for weeks, and it’s exactly why we rebuilt the AI Revolution Portfolio from scratch.
Between the three of us, we’ve issued more than 200 buy recommendations over the past year. Some have turned into extraordinary winners — my Lumentum (LITE) call is up 645% since last August, and Louis’ Nvidia (NVDA) position is up 375% since 2023. But when Walmart wobbles and oil pops and the tape drags AI stocks down in sympathy, the real question isn’t “which stock do I own?” It is “does my portfolio actually hold together when the tape gets ugly?”
That’s the question no single recommendation can answer. It’s why Louis has stepped into a new role, and why the AI Revolution Portfolio is now built around roughly 20 hand-picked positions, each with a specific allocation percentage attached… not just a list of names to figure out on your own.
Since inception, the portfolio is up 106.7%. Since its last rebalance alone, it’s gained 58%, more than double the Nasdaq’s 25% over the same stretch, even through stretches of exactly this kind of noise.
The newly rebuilt portfolio is live now, along with our AI Revolution Position-Size Calculator, Eric’s report naming the AI stocks to sell before the next shakeout, my report on the one stock I believe has true 100X potential in this cycle, and a recorded Board Meeting where the three of us walk through every position and why it earned its place.
Today’s selloff is a sympathy trade, not a fundamentals story. The AI Revolution Portfolio is built for exactly this moment — separating the noise from the signal, and making sure you’re sized correctly in the names that actually matter.
Click here to access the AI Revolution Portfolio.
The post The 161-Year-Old Warning Hiding in the Tech Selloff appeared first on InvestorPlace.
