The tool isn't the work

You're reading this on a site I built before any of the AI tools I now use every day. I designed it myself. It took me longer than it would today. I'd do it the same way again.

That's not nostalgia. AI is faster, and I'm not pretending otherwise. I lean on it constantly. The honest question isn't whether the tools got better. It's which part of the work the tools actually touched.

Speed is a feature of the tool. It isn't a feature of the work. The work is the part where you decide what to make, and which version of it is worth keeping. AI made the typing cheaper. It didn't make the choosing cheaper.

Where this site happens to be the proof

There's a small TV-noise GIF over my face in the hero. Hover it and three social links show up inside the static. I sat with that idea for a while before I built it. It isn't what a tool would have suggested. It came from wanting the page to feel a little weird, a little personal, slightly off-balance, and from no other source.

The doodle lines under the word "Scroll" each have their own animation delay. I picked the values by eye. The portrait is hand-drawn. The accent is lime green because I tried a handful of colors and lime green felt right. None of those decisions are good in some objective sense. They're mine, which is a different thing.

A faster tool would have shipped a competent version of this site months sooner. It wouldn't have shipped this site. That's the part I keep coming back to.

The strongest counter-argument

It's true that AI helps you choose, too. It will offer variations, suggest palettes, generate ten layouts in the time it takes to read this paragraph. I use it for that. It's genuinely useful.

But choosing is a skill you only get better at by doing it under your own taste. A suggestion you accept without friction is a suggestion that came out generic. The friction is the work. You can outsource the typing. You can't outsource the part where you sit with three options and notice which one feels right.

Use the tools

There's no virtue in slow. The "do it the hard way" framing is a trap, and I won't defend it. Use the tools. Use them constantly. But notice which parts get faster when you accept the help, and which parts don't. The parts that don't get faster are the parts that are actually the work.

The little TV over my face is a stupid detail. It took longer than it should have. I'd put it back tomorrow.