Over a single weekend, entirely from scratch and heavily “vibe coded”, I created by some distance the best word processor I have ever used. I’ve named it vibedit. I’m writing in it right now. If there is an actually productive task for generative AI, it is as a creator of bespoke tools like this. Given this new, relative ease of app development, it is easy to imagine the atomisation of software into a mist of customised personal projects, droplets as numerous as users.

I mean...not really? I can't imagine very many people bothering to spend a weekend creating their own word processor from scratch. Professional writers, and specifically the subset of professional writers who have extremely idiosyncratic preferences and needs for a word processor, might do so. But most folks are still going to use a program that someone else has made.

The real unraveling of software isn't going to come from AI enabling home hobbyists whipping up their own homebrew version of the WordPerfect they used to know and love.

Instead, it's more likely to come from AI dramatically lowering the cost of entry for competing developers to create their own word processors. Making a competing WP product that could take on Word has historically been a massive undertaking. But if it could be done by a few folks spending a couple of weeks really nailing the vibe-coding to come up with a better mousetrap, then suddenly Word faces a lot more competition.

So now any developer is able to spend their time coming up with a WP that might be better than Word in some given way that a consumer might find important - or even just cheaper, while still being better than the free word processors like Google Docs. Nearly all folks aren't going to want to make their own word processor, but plenty might be willing to give that new WP that's been circulating around a whirl.