Software Stocks Slide Again as AI Spending Doubts Mount

The software sector is experiencing its second major selloff in recent months, as investors grow increasingly skeptical about whether massive AI infrastructure investments will translate into sustainable revenue growth for enterprise software companies.

The renewed decline comes as CFOs across the industry face mounting pressure to justify AI spending while demonstrating concrete returns. The pattern mirrors broader concerns about whether the current wave of AI investment represents genuine transformation or another cycle of overhyped technology promises.

For finance leaders, the selloff signals a market recalibration around a question they've been asking internally for months: When does the AI demo become actual margin improvement? The timing is particularly awkward—many software companies spent 2024 and early 2025 pitching AI features as justification for price increases or new product tiers, but the revenue impact remains murky at best.

Here's the thing everyone's missing: This isn't just about software multiples compressing. It's about the fundamental tension between two competing narratives that can't both be true. Either AI is going to automate away enough work that software companies need fewer engineers (good for margins, bad for growth), or AI is going to create entirely new categories of spending (good for growth, uncertain for margins). The market is starting to suspect the answer might be "neither, actually."

The selloff also reflects a more practical concern for corporate buyers. As one CFO put it in a recent earnings call (paraphrasing the general sentiment): "We're being asked to fund AI pilots across every department, but we're not seeing the productivity gains that would justify expanding those pilots into full deployments." That's the kind of comment that makes software investors nervous, because it suggests the sales cycle isn't shortening—it's lengthening.

(This is, I should note, exactly what happened with "digital transformation" circa 2019-2020. Lots of consulting fees, lots of new software seats, and then a collective realization that maybe the ROI calculations were... optimistic.)

The pattern is particularly visible in the disconnect between infrastructure spending and application-layer revenue. Hyperscalers are building out massive AI compute capacity, but the software companies that were supposed to monetize that capacity are struggling to demonstrate why customers should pay more for AI-enhanced versions of tools they already own.

What makes this selloff different from typical sector rotations is the speed at which the narrative shifted. Six months ago, any software company without an "AI strategy" was being punished. Now, companies are being punished for having AI strategies that don't show clear paths to profitability. The market went from "show me your AI roadmap" to "show me your AI revenue" faster than most product teams could ship features.

For CFOs watching this unfold, the implication is clear: The market is done rewarding AI promises and is starting to demand AI proof. That means the internal conversation shifts from "should we invest in AI?" to "can we quantify what our AI investments have actually delivered?" And for many companies, that's a significantly harder question to answer.

The broader question—the one that's going to define software valuations for the next year—is whether AI represents a new growth driver or a margin compression event disguised as innovation. Because if AI tools make software development faster but don't create new categories of spending, then software companies just automated themselves into lower growth rates.

Which would be ironic, but also exactly the kind of outcome that makes markets nervous.