The AI Hype Is Eating Its Own Tail—And Software Stocks Know It

The market’s having a laugh at software stocks’ expense right now.

While Nvidia’s floating toward record highs on the back of AI euphoria, battered software names are getting told to proceed with caution. That’s not a coincidence. That’s not even really about software anymore. It’s about who gets rich from AI and who gets left holding the bag when the dust settles.

Let’s map what’s actually happening beneath the noise.

Two people enjoying a casual meal with various lunch bowls and refreshing drinks at a light-filled café. Photo by thecactusena ‎ / Pexels

The Divergence Nobody’s Talking About

You’ve got Nvidia trading near all-time highs. You’ve got Qualcomm up 7% on news it’s partnering with OpenAI and MediaTek to develop smartphone AI chips. You’ve got a bank CEO literally running his earnings call through an AI clone—a preview of what Customers Bank thinks the future of digital workforce looks like. Meanwhile, software stocks are in what the market’s politely calling a “washout.”

This isn’t a market sector being rational. This is a market sector making a choice.

The choice is simple: infrastructure and applications built for AI are printing money. Software built by humans for the last 15 years? That’s becoming inventory nobody wants. It’s the difference between owning the railroad versus owning a buggy whip factory in 1895.

When Microsoft and OpenAI shake up their partnership—specifically capping revenue share payments—that’s a signal that the power dynamic is shifting faster than anyone expected. OpenAI’s not negotiating like a junior partner anymore. It’s behaving like the thing everyone actually needs. Microsoft still owns the cloud infrastructure, sure, but OpenAI’s increasingly looking like the one steering the ship.

Detailed close-up of a newspaper displaying global financial market statistics and country flags. Photo by Markus Spiske / Pexels

Here’s Where It Gets Weird: The Geopolitical Noise Isn’t Winning

The Dow rose Monday on U.S.-Iran news around the Hormuz Strait. That should matter more than it does. Shipping lanes, oil prices, global trade—these are the things that traditionally make or break market weeks.

But here’s what’s wild: Chris Larkin at E*TRADE basically said the entire month has been “a one-sided battle” between geopolitical uncertainty and AI-driven earnings growth. And AI is winning. The bulls’ enthusiasm might be misplaced—Larkin seems genuinely uncertain—but right now, investors are treating geopolitical risk like a distraction from the real story.

I think that’s reckless. And I also think it’s exactly what happens when traders convince themselves a secular trend is unstoppable.

The stock market this week is packed. Big earnings, economic data, the Musk-Altman trial kicking off with jury selection. Any one of those could pivot sentiment. But unless something genuinely catastrophic happens at Hormuz, my read is that AI momentum’s so strong it’ll absorb the bad news and keep moving north.

That doesn’t make it smart. It makes it market reality.

The Consolidation Nobody Expected

OpenAI’s capping Microsoft’s revenue share. Qualcomm’s building chips with OpenAI instead of just making mobile processors. A bank CEO’s using AI clones to replace himself on earnings calls. These aren’t separate stories.

This is consolidation happening at the application layer instead of the corporate merger layer. OpenAI’s essentially saying: we don’t need Microsoft as much as you need us. Qualcomm’s saying: we’re moving upstream into the actual AI product layer. Customers Bank’s saying: forget middle management, we’re going straight to AI.

Traditional software vendors—the ones selling enterprise applications and middleware—are getting disintermediated. They’re not failing because they’re bad at software. They’re failing because the entire value chain is reorganizing around whoever controls the AI model and the compute.

Software stocks that will survive? The ones building on top of large language models, not competing with them. The ones becoming distribution channels. The ones that realize they’re not software companies anymore—they’re AI interface companies.

The ones still trying to peddle CRM systems and ERP deployments are done. Not next year. Now.

Small-Cap Value Plays Are Actually Worth Thinking About

You’ve got financial advisors writing about whether ISCV or IWN is the better small-cap value vehicle. Fees matter. Fund size matters. But here’s what really matters: small-cap value stocks have been beaten down for years while mega-cap AI darlings went vertical.

That’s a setup for mean reversion. Not a guarantee. A setup.

The thing is, small caps don’t care about OpenAI’s partnership moves or Nvidia’s valuation. They care about actual earnings and cash flow. A lot of them are genuinely cheap. And if this market gets even a small whisper of recession fears, capital’s going to rotate hard away from $3 trillion mega-cap tech into places where the stock price actually correlates with business fundamentals.

The three-ETF buy-and-hold-forever advice floating around is actually solid—but only if you’ve got 20 years and zero need to check your account balance. Most people don’t. Most people are going to panic-sell the second the market drops 15%.

My Take: The Reckoning’s Coming, But Not Yet

I think software stocks are pricing in reality that institutional investors are still in denial about. The shift from software-as-a-service to AI-as-a-service is happening right now. The companies that own the models and the compute are consolidating power. Everyone else is becoming a feature, not a product.

The geopolitical stuff? It’s a sideshow until it isn’t. That could change Monday. Could change next week.

But the software washout? That’s structural. That’s real. That’s what happens when an entire category of business gets disrupted faster than anyone thought possible.

Nvidia’s at record highs because the market’s right—there’s going to be hundreds of billions spent on AI infrastructure in the next five years. Software stocks are beaten down because the market’s also right—most of that infrastructure is being built by OpenAI, Qualcomm, MediaTek, and whoever else gets in early. Not by the software guys.

The uncertainty I genuinely have: whether the software washout’s already priced in, or whether we’re still in the early innings of capitulation. My gut says we’re seeing the bottom formation, not the actual bottom yet. That usually happens when everyone’s admitted something’s bad and stopped caring about it. We’re not quite there.