The SaaSpocalypse

I was reading earnings coverage on a Tuesday morning when my feed turned into a disaster film. Anthropic had released a plug-in for in-house legal teams. One tool. One vertical. And within hours, legal software stocks dropped 12 to 20 percent. Thomson Reuters fell 16 percent. By Wednesday, roughly $285 billion in software market value had vanished.¹

A trader at Jefferies, Jeffrey Favuzza, gave it a name. "The SaaSpocalypse." He described the trading floor as pure panic: "Trading is very much 'get me out' style selling."¹ That phrase stuck. Not "sell if it drops further." Just get me out.

One product announcement. $285 billion in smoke.

The thing that stuck with me wasn't the selloff. Markets panic all the time. It was what triggered it. One AI tool for lawyers. That's all it took for investors to look at the entire software industry and think: maybe none of this is safe.

Nothing had actually broken. Nobody canceled contracts that week. The products still worked. The revenue hadn't changed. But the story about those products had changed overnight. And if you're publicly traded, the story is the price.

What Vembu Said

The product still works. The narrative walked out of the room.

I was still thinking about it a few days later when I came across a post from Sridhar Vembu, the founder of Zoho. He wrote something I haven't been able to shake: "An industry that spends vastly more on sales and marketing than on engineering and product development was always vulnerable. AI is the pin that is popping this inflated balloon."²

That landed differently than the panic coverage. Vembu wasn't talking about whether AI could replace software. He was talking about what the software industry had become. A lot of these companies spend more convincing you to subscribe than they spend building what you're subscribing to. That model works as long as switching feels expensive. Once people believe switching might get easy, everything shifts. Not because AI replaced anything yet. Because it made replacement imaginable.

Nvidia's Jensen Huang called the whole selloff "the most illogical thing in the world."² Arm's CEO called it "micro-hysteria."² They're probably right that most of these companies will still be here in five years. But I don't think the market was pricing whether software would die. It was pricing whether the certainty was real. That's a different question. And it doesn't need an answer to do damage.

The Part That Stayed Quiet

By Friday, the Dow had bounced over a thousand points and crossed 50,000 for the first time. The SaaSpocalypse was already fading. Which is the part that stayed with me.

The panic left, but the question Vembu asked didn't. Is an industry built more on marketing than engineering ready for a technology that makes its output easy to replicate? Nobody answered that. Everyone just moved on to the next headline.

I wonder how many of these weeks we'll watch before we notice what's happening. The fear shows up, does real damage to real companies, and then evaporates before anyone has to sit with what it revealed. The SaaSpocalypse lasted about a week. What it named is still there.

---

References

Footnotes

1. Vlastelica, R. (2026). "'Get me out': Traders dump software stocks as AI fears erupt." Bloomberg via Yahoo Finance ↩ ↩²

2. Griffiths, B. (2026). "Here's what smart people are saying about a software apocalypse." Business Insider ↩ ↩² ↩³