Post #1421412

2026-04-19 21:16 UTC

The way people talk about purported benefits of generative AI for &quot;democratizing&quot; art should probably make us reflect on how our world perceives all creative endeavors in an increasingly professionalized and commercialized way. Which has been going on for a long time. Consider music. Before the phonograph, creating music wasn&#39;t exclusively or primarily the work of professionals. Professional musicians of one sort of another have existed for a *very* long time, but the average person&#39;s experience of music often involved them or other non-professionals singing and/or playing an instrument. If you listen to some 1920s field recording of a folk song in a remote part of Appalachia, you might not care to buy an album by that performer even if they had a modern recording studio. Most people weren&#39;t experts at singing or playing instruments. But they didn&#39;t need to be. A lot of resources on learning instruments (especially &quot;classical&quot; instruments) assume you want to get good enough to perform professionally. But you don&#39;t have to. You don&#39;t need an AI to play or compose a song. There are plenty of instruments you can buy for less than what that AI will cost (at least once they&#39;re actually charging enough to not lose money). If you don&#39;t spend long hours in serious practice, perhaps no one will want to listen to your music, especially if that means paying for a recording. If you create it with AI, no one will care either. But art doesn&#39;t have to be about that.

Replies (2)

@ids1024@mathstodon.xyz 2026-04-19 21:26

Actually writing music and lyrics was starting to become a professional and commercial enterprise (to an extent, for some music) as soon as the printing press was introduced. But traditionally everyone who performed music or told stories (often the same thing) added a little (or not so little) bit of themself to their performance, intentionally and unintentionally. There wasn't such a strong binary between "creators" and "consumers" of artistic expression. All the more so in oral traditions before the printing press and before writing. In some ways, that's the very *opposite* of what one adds to a work by providing a prompt to a generative model. Often the most human aspect of art is the details. Play music for yourself and people around you, even if no one would care to pay for it. Write stories or tell them to people around you even if they aren't worth publishing. You don't have to be good at it. You don't need some special technology to do these things. You already can express things that are more "artistic" than anything out of a generative model. That's what makes you human.

Open ##1443599
@yakmacker@society.kalli.st 2026-04-20 01:45

@ids1024 The weird part is, you can still upload an album to bandcamp or play shows as a non-professional. A lot of people do it. Music recording did make the role of musicians in everyday life less necessary (because now you can just put on a playlist for your party, for example), but didn't make them obsolete. A party with live music is very clearly better than one without, even if the musicians are not pros. Ironically, we often find ourselves in a situation where there are fewer musicians than we want, which is not what you would expect after something gets automated. I think it's because what happened wasn't actually a supply shock from automation, but rather, as you say, a perception among people that it wasn't worth trying to pick it up as a hobby or a side gig if people can put on a record by professional musicians instead of listening to you play.

Open ##1443602