"If it sounds good, it is good"

The impact of AI on music production

If you wonder whether artificial intelligence will replace “real” music. The answer is actually quite simple. Yes. Wherever you cannot hear the difference, AI generated music will replace human made music. Background music in shops, waiting room melodies, generic muzak: perfect territory for AI. Nobody notices the difference, and to be honest, it does not matter much. Muzak created by AI will sound just as functional as muzak created by people.

But music is so much more than that. Music connects, moves, evokes memories and shapes culture. And that is where things become far more complicated for AI. Not impossible, but certainly challenging. Duke Ellington once said, “If it sounds good, it is good.” He meant that the quality of music is defined by its impact rather than by rules, genre or technical correctness. If it resonates, inspires or touches you, then it is good. Full stop.

And that is exactly where it becomes interesting, because the examples of AI creating something truly meaningful are still very scarce. AI music is mostly still a gimmick. Take The Velvet Sundown, often presented as the first fully AI generated band. Fun as an experiment, but it is mainly the novelty that attracts attention, not the music itself. It has not built a genuine fanbase or cultural footprint.

AI is great for creators that want to express a message but can’t make music themselves. That is precisely where AI excels: enabling people to make a song who otherwise never could. Suno and Udio are perfect for that. If you have not tried them yet, you really should. In a few minutes you can make a funny little song for your best friend’s birthday. It is wonderfully playful and incredibly accessible. But it remains what it is: output from a simple prompt. A meme. A gimmick.

For anything beyond that, you need real creators. Talent, taste, nuance, context. Music professionals do not need AI to write a strong new track, because that creative ability is already theirs. For them, AI is not a replacement but a catalyst. A tool that helps them test ideas more quickly, explore arrangements, build variations, manipulate stems and make creative directions more tangible. These professionals are waiting for the next generation of AI tools. Systems built not for casual users but for makers. Systems that accelerate, deepen and enrich their workflow.

Big opportunity, no rulebook yet

For brands, AI becomes especially interesting once music can be deployed dynamically. We are now in a world where a message can adapt in real time to the moment of the day, the context or even the personal taste of the receiver. A commercial or social ad no longer has to be static. Music can shift with it. That presents a huge opportunity for brands that dare to look beyond the default version of a soundtrack.

Yet we are still far behind in setting the rules that govern all of this. Rights, ownership, transparency: the foundation is not in place. It feels as if the technology is moving faster than the legal and ethical frameworks can ever keep up.

If you want to avoid surprises, it is wise to hold off on using AI generated music in advertising for now, or at least make sure you are properly informed. The genie is out of the bottle, but until the legal fog clears, we operate in a world where tremendous potential and significant uncertainty exist side by side.