Adam Neely made this video about AI and music, what do you think?
My thoughts on this:
For some end users, the origin story does not matter. If the result does what they need and no one was harmed or cheated in producing it, most people will not spend energy caring about the process.
They just want the thing. Training-data ethics and legality matter, but for simplicity let's look over that for a moment.
Programming makes this obvious. For some of us, writing code is craft and identity - it is an art form. For others, it is a tool: solve the problem, move on.
Music is similar. Some people make music because the act of making and performing it is the point. Others make music because they need an output - for a game trailer, a podcast intro, a vibe in the background, a living.
If AI gets reliably good at turning an intention expressed in language into an output that matches the creators intent and what they imagined well enough, then in outcome-driven contexts the default will be to use it.
Not because it is morally superior, but because it is cheaper, faster, and more predictable than a labor-intensive workflow.
People can reject that for artistic reasons, for branding reasons, or because they value skill and community. Those remain choices, not inevitabilities the market will automatically reward.
And as a listener (I am not talking about myself here, I am generalizing) when a song genuinely hits, I usually am not auditing the process in those initial moments.
Just like most people do not think about the algorithms behind the software they depend on daily. The craft still matters, and there will always be a prestige layer where process is part of the value.
But in the commodity layer of recorded music, a lot of consumption behaves like output is what matters. Im describing incentives more than endorsing them.
TLDR: If the training-data/copyright issue is actually solved (consent, licensing, compensation), then I don't see a principled problem with this - it will be a powerful tool and can push music forward. And the old tools and craft-based workflows are still there for anyone who values them.