I hate AI slop as much as the next guy, but I will say this: I’m almost old enough to retire, and I can’t remember a time when the radio wasn’t playing completely terrible music made by uninspired employees or machines for purely commercial purposes 90% of the time.
For every Beattles, there’s a hundred Backstreet Boys. Over time, the good music stays and the not-so-good music is rightfully forgotten. That’s why music from the past always seems to be so much better than the shit that plays on the radio today. But crucially, it was the same amount of shit back then too.
AI is just the latest iterative process in the shit music industry. If there’s any good AI music, you’ll know when people still listen to it in 20 years. It’s too early to tell.
Love the idea
Where I see AI music actually serving a purpose is in the space where you have lesser celebrities who already have a following and want to release music.
Allowing them to quickly iterate through a bunch of different ideas to send to the people that actually make the music to find something that is a fit for their brand and followers so that their music becomes more personal and it, uh, or sorry, not their music, their songs become more personal because they found something that fit them without running their music staff into the ground or spending 15 hours scrolling on some beat sales site.
Outside of that, I could see a lot of people who are interested in writing a song or putting their poetry to music or something, using it as a tool to see what that could be like, but I doubt that any time soon we will hear an AI musician or AI singer or AI band actually chart and be considered to be good music.
An ai band is topmost the charts in Korea.
Cooked.
Perhaps it will be prompt artists
Honestly, I’m listening to nothing but AI music at this point – All I ever listened to was EDM and shit anyways, so it hasn’t been much different.
What service generates it?
Suno is a common one
making with suno and how the voice kept not accentuating the right part inspired this post
Yeah, I tried it out with some of my old songs and, like, it definitely took them in an interesting direction, but it wasn’t always good. Like, I think I got 20 seconds or so worth of interesting music out of 15 prompts.
Nah, in a few months I will be hearing my own voice singing all of the hit songs with ai
Edit: or my ex’s voice from a voicemail I randomly have from 15 years ago, I haven’t decided.
