There’s a lot of noise right now around AI and music.
Depending on who you listen to, it’s either the end of creativity as we know it or just another tool in the long line of tools musicians have always use.
Somewhere in the middle of all that noise sits a more useful conversation. One that feels a little less reactive and a lot more grounded in how music actually works.
Now, I recently came across an article on the New Music Strategies blog called “Music AI Is Not AI Music” that cuts straight through the panic and reframes the discussion in a way that’s worth paying attention to:
The central idea is simple… AI is not “stealing” music in the way many people think it is but that doesn’t mean there aren’t real issues we need to deal with.
AI Learns the Same Way Musicians Do
If you strip everything back, what AI models are doing isn’t all that unfamiliar.
They are trained on large amounts of existing material. They analyse patterns. They generate new combinations based on what they’ve learned.
Sound familiar? That’s not far removed from how songwriters work.
You and I and every other songwriter you’ve ever heard of and admired has absorbed influences. We all have.
We listen, we internalise, we experiment, and eventually something comes out that feels like our own voice.
The article makes the point that AI isn’t storing songs and replaying them. It’s not sitting there like a hard drive full of tracks waiting to be copied and pasted. It’s working with patterns, structure, and probability.
That doesn’t make it human but it does make it a lot less mysterious than people are making it out to be.
So… Is It Plagiarism?
This is where things start to get interesting.
The argument put forward in the article is that most AI output doesn’t meet the threshold for plagiarism. Not unless it’s clearly reproducing something specific and identifiable.
And that opens up a bigger question.
If a human songwriter writes something influenced by ten different artists, we call that creativity. We call that style. We call that voice.
But when AI does something similar, we’re quick to call it theft. Now, I don’t know about you but I sense a bit of a contradiction there.
This doesn’t mean that everything AI produces is meaningful or valuable. It just means the plagiarism argument might not be the strongest place to stand.
The Myth of Pure Originality
There’s also a deeper idea running through this which is the idea that originality, in the pure sense, doesn’t really exist.
Music has always been built on what came before. Chord progressions, melodies, rhythms, lyrical themes. They get reused, reshaped, reinterpreted and as songwriters, we don’t create in a vacuum. We create in a continuum.
AI just speeds that process up and scales it in a way we haven’t seen before.
While this might be confronting it’s also very clarifying because it forces us to ask what actually makes a piece of music matter.
The Real Issue Isn’t Creative… It’s Economic
This is where the article really hit home for me. The real problem isn’t whether AI is “creating” properly.
The real problem is who benefits from all of this.
AI systems are trained on vast amounts of music. That music didn’t appear out of nowhere. It came from real people who spent years learning their craft, writing songs, recording, releasing, and building something over time.
Yet those creators are not part of the economic loop when AI systems generate new content and there lies the imbalance.
Not the act of learning. Not the act of generating. But the distribution of value and if you zoom out, this isn’t just a music problem. It’s a broader issue about how technology interacts with labour, creativity, and ownership as a whole.
AI Still Sounds Like AI
There’s also a practical reality that often gets overlooked. AI-generated music, at least right now, tends to sound like… AI-generated music.
It has patterns. It has a certain sameness. In many cases, it resembles other AI outputs more than it resembles any one human artist.
There is a reason for that though, human music carries context. It carries experience, memory, imperfection and intent and those things are harder to replicate than just chord progressions and melodies.
So while AI can generate something that resembles music, there’s still a noticeable gap between that and something that feels lived-in like a trusty pair of sneakers.
So Where Does That Leave Us?
For me, this isn’t about defending AI or rejecting it. It’s all about trying to understand what AI actually is.
AI is not a songwriter. It’s not a replacement for human experience. But it is a tool that can generate music-like outputs based on what it has learned.
The bigger question is this: If influence isn’t theft, then what is it that makes music valuable?
Is it originality? Is it execution? Or, is it something much deeper. Something tied to the fact that there’s a real person behind the song, trying to say something that matters to them?
That’s the part AI can’t replicate.
Well, at least not yet and maybe that’s where the real opportunity is. Not in competing with AI on volume or speed, but in doubling down on what makes us human in the first place.
If you want to explore this further, the original article is well worth your time: https://newmusicstrategies.com/music-ai-is-not-ai-music/
I personally believe it’s one of the more grounded takes on a topic that’s usually anything but.
Peace,
Corey 🙂
