Creating Music With Large Language Models

I think that, honestly, this is just the teething problems. We're trying to shackle this crazy thing that's just so dynamic and we barely understand it ourselves. It's gonna take quite a while, I think, to get this to a point where it's like 80% or even 90%, just you know that it's not gonna do anything weird or funny. The thing that is harder is when it goes crazy and loses its mind and says all these random things. That's just the nature of large language models at this stage.