I've been reading about something people are calling "spite-coding," and I think it could become a real problem for software companies. Basically, developers are using AI to build alternatives to software they're tired of paying for or dealing with.

One developer, Brandon Thomas, has released free, open-source versions of tools modeled after Photoshop, Illustrator, Premiere Pro, Word, Excel, and PowerPoint.

Another developer got tired of waiting 30 seconds for Plex to load on his LG TV. With Claude Code, he built his own client that reportedly loads the same screen in about 3 seconds.

Now, these aren't all polished replacements. Some of the ArtCraft applications are still in early development, and building software still requires technical knowledge. But I think the bigger story is what happens when the cost of creating software keeps falling.

For years, companies have relied on subscriptions, proprietary ecosystems, and the difficulty of building competing products. AI could start weakening some of those advantages.

I don't think Microsoft or Adobe is going anywhere tomorrow, especially with enterprise customers. But I do wonder how much pricing power these companies will have if small teams can build decent alternatives at a fraction of the traditional development cost. Could AI eventually do more damage to software subscription businesses than it does to traditional jobs?

ArtCraft open-source projects: https://github.com/storytold

Developer's Plex project: https://www.reddit.com/r/ClaudeCode/comments/1x043yp/i_used_claude_code_to_reverseengineer_an_lg_tv/