Tech stocks tumble, but the real shake-up isn’t AI — It’s pricing

US software stocks have taken a beating lately, with big names in the SaaS world getting trounced as investors price in fears that generative AI isn’t just a nifty add-on, but a business model disruptor. The sell-off has been broad and deep: the software and services index has lagged the broader market by near-record margins recently and has wiped out enormous chunks of market value amid a wave of pessimism about AI’s impact.

But here’s the real point: this isn’t simply about hype or fear. It’s about pricing power, or the lack thereof.

SaaS — Still useful, but fewer seats to sell

Traditional SaaS pricing usually looks something like this: charge per user, per month, for access to your cloud-hosted software. It worked great during the hyper-growth era, when every company was adding more tools than a Swiss Army knife. Shareholder presentations practically revolved around land and expand strategies.

Then came AI tools that promise to automate the very tasks those seats were supposed to solve. If one AI agent can draft legal briefs, summarize contracts, spin up dashboards, or write code faster and cheaper than five specialists clicking around a SaaS UI, guess what happens to your somewhere-between-$80-and-$400 per-seat subscription?

Investors are effectively repricing the future: instead of valuing recurring revenue as a sacred cow, the market is suddenly asking, “And how many of these seats will you still be able to sell in a world where AI eats workflows for breakfast?”

Declining growth, shrinking pricing power

That shift shows up in the fundamentals. Enterprise spending on SaaS has slowed, growth forecasts have come down, and companies are trimming the number of applications they deploy per organization. As that happens, per-seat pricing gets squeezed, and revenue growth — the lifeblood of SaaS valuations — slides.

Some firms are already reacting. Instead of charging simply for access (seats), they’re experimenting with:

• Outcome-based pricing — customers pay for results (e.g., “generate insights that save $X”), not logins.

• Consumption or API-based pricing — fees tied to actual usage volume (e.g., number of actions or compute cycles), similar to cloud billing.

• AI augmentation premiums — charging extra for native AI features that demonstrably reduce cost or time.

These models make a lot more sense in a world where generic AI can perform tasks previously delivered by specialized software.

Not all SaaS is doom and gloom

Let’s be clear: the market isn’t saying SaaS is dead. What it is saying is that vanilla, seat-based pricing in a world of smarter automation isn’t worth the multiples it used to command. Investors are comparing software firms against competitors that embed AI deeply — or against AI platforms that might ultimately replace the need for whole categories of niche tools.

Companies that will survive and thrive are those that:

1. Demonstrate quantifiable value: If a product can prove it saves customers money or drives revenue better than an AI alternative, pricing becomes defensible.

2. Move pricing toward outcomes: Customers are willing to pay more when the vendor shares in the upside.

3. Integrate AI as a value creator, not a checkbox: AI that genuinely expands capabilities — rather than replacing user effort — earns a premium.

Valuation reset or reality check?

Many strategists now argue the sell-off partly reflects markets pricing in a worst-case scenario that may not materialize in full. Some stocks have become oversold relative to fundamentals, and a rebound isn’t out of the question if companies execute well post-AI integration.

Still, the broader takeaway for SaaS leaders should be straightforward: stop pretending pricing models from the 2010s will carry you through the 2030s. If your product doesn’t deliver tangible business outcomes that customers can’t easily replicate with a generic AI agent, your pricing (and, by extension, your valuation) will suffer.