Seeing Through Tailwind CSS's 75% Layoffs: How AI Disrupts Software Business Models?

Drupal founder Dries Buytaert wrote a commentary on the massive layoffs at Tailwind, exploring how AI is impacting the business models of open-source software.
(Background: Forbes analyzes key trends for 2026 cryptocurrencies: five major trends revealing the industry’s path to maturity)
(Additional context: Bloomberg summarizes Wall Street’s 50 institutions’ expectations for 2026: AI driving global average growth of 3%, high valuation risks still warrant caution)

Tailwind Labs laid off 75% of its engineering team last week. CEO Adam Wathan recalculated revenue forecasts during the holiday break. He explained what happened in a comment on GitHub:

The reality is, 75% of our engineering team lost their jobs yesterday because AI delivered a brutal blow to our business.

Although Tailwind is more popular than ever, our documentation traffic has decreased by about 40% since early 2023.

A common narrative now is “AI is killing open-source businesses.” I don’t think that’s entirely accurate.

AI didn’t kill Tailwind’s business; it just underwent a “stress test.” Their business model didn’t pass the test, but that doesn’t mean all open-source business models have failed.

Tailwind’s business model has operated for years, relying on developers accessing their documentation, discovering Tailwind Plus (a set of pre-built UI components priced at $299), and making purchases. Traffic drives “discovery,” and “discovery” drives sales. It’s a reasonable business model, but always fragile.

Over the past year, more developers have turned to AI for code rather than reading documentation, causing their sales and marketing funnel to break.

There’s an important fairness issue here: AI companies are using Tailwind’s documentation and community-generated content to train their models. Now, these models can generate Tailwind code, answer Tailwind questions, but no longer direct anyone to visit Tailwind’s website. Value is extracted, but compensation doesn’t flow back. This troubles me and warrants broader policy discussions.

One thing I keep reflecting on is: AI will commodify (make mundane) anything that can be “fully specified.” Whether it’s documentation, pre-built component sets, CSS libraries, or open-source plugins. Tailwind’s commercial products are built on “specifications.” AI makes generating these things effortless. AI can deliver specifications, but it can’t run a business.

So, where is the value flowing now? It’s moving toward things that require “showing up” continuously, not just “defining.” Not things you define once and finish, but things that require repeated effort.

Value is shifting toward “operations”: deployment, testing, rollback, observability. You can’t instruct (prompt) a system to maintain 99.95% uptime during Black Friday shopping. You can’t just command a website to stay secure, updated, and stable at all times.

That’s why Vercel created Next.js and offers it for free. Open-source frameworks are the “medium,” hosting services are the “product.” My own company, Acquia, is the same. A large part of our business model revolves around selling Drupal-based products: hosting, search, CI/CD pipelines, digital asset management, etc. We don’t sell “describable things,” we sell “operations.”

Open source has never been a commercial product; it’s a medium to other things.

When asked how to pivot, Wathan candidly said: “Until today, I still don’t know what direction we should turn to.” I’ve written about how digital agencies might evolve, but the situation with CSS frameworks and component libraries is even more challenging. Some open-source projects are great as “features,” but not necessarily as “businesses.”

Tailwind CSS powers hundreds of millions of websites. The framework will survive, but whether the company itself survives is another matter. I support them; the world needs more successful open-source companies.

— Original source