Winners and Losers: BMG, Suno and the Cost of AI Music

Winners and Losers: BMG, Suno and the Cost of AI Music

BMG and Suno have announced a global alliance covering BMG's recorded music and publishing repertoire. Participating BMG artists and songwriters can choose to have their work included and compensated, while the agreement also settles Suno's prior use of BMG recordings and publishing works.

So, who wins? Suno wins. It gains legitimacy and access to professionally created music with the blessing of another major music company. BMG wins. It turns AI from a copyright threat into another potential revenue stream. Participating rights holders may win. At least they're being offered choice and compensation for the use of their work.

And then there is everybody else.

THE LOSERS

Session musicians lose. A technology capable of generating drums, bass, guitars, keyboards, vocals and arrangements in seconds suddenly becomes more attractive than world-class musicians—at least economically. AI doesn't have to actually be better. It only has to become good enough that "artists" decide to use it on a master.

Producers, engineers and studios lose. The same economics apply. Every AI demo or finished track potentially represents a recording that wasn't made by people. It doesn't sound as good, but look who it appeals to: the already broke musician earning fractions of a penny per stream on Spotify. When you're barely making money from your music, a cheap substitute becomes awfully tempting.

Songwriters may lose more than they gain. Think pitching songs is hard now? The same so-called "songwriters" celebrating this Suno victory may not realize their own pool has just been diluted with poets and lyricists who can put words into a prompt and claim to be songwriters.

And AI is introducing something beyond job displacement into music: dishonesty for the sake of vanity. People want to say, I wrote this. I made this. This is my song. The technology makes it increasingly easy to take credit for abilities we don't actually possess.

FANS LOSE, TOO

There's a reason so much AI music already sounds familiar. These systems learned from music humans already created, and generic prompts naturally reinforce familiar conventions. Type something like "modern country" and you've given the machine very little creative direction beyond producing recognizable characteristics of modern country. Do that millions of times and we shouldn't be surprised when much of the output starts swimming in the same musical pool.

Someone who understands music history, theory, arrangement, instrumentation and production can give AI far more specific musical direction. A musician might describe tempo, meter, harmonic movement, instrumentation, dynamics, phrasing, arrangement, production era or rhythmic feel. Someone without that vocabulary is more likely to type "modern country song about a breakup" and let the machine fill in nearly everything else.

Ironically, the musical knowledge AI supposedly makes unnecessary is exactly what separates an informed prompt from a generic one. Fans ultimately inherit the result: an endless supply of cheap music built around things we've already heard.

THE BIGGER LOSS

Then there's a question bigger than royalties or lost sessions: Why should a child spend ten years learning piano? Certainly for the love of music. That will always matter. But why pursue mastery professionally if society increasingly decides there's no economic value in having a human being master it?

What happens when fewer kids learn piano, fewer guitarists spend years developing their own sound, fewer drummers develop a feel that can't be written into notation and fewer singers discover voices nobody has heard before? AI learned from generations of humans who did exactly those things. We should think carefully before making them economically obsolete.

THE MONOPOLY WINS AGAIN

The technology company wins. The music corporation wins. The copyright owners may get another royalty stream. Meanwhile, musicians, producers, engineers, studios, songwriters and even fans absorb the consequences.

Once again, the people controlling the technology and intellectual property find a way to monetize the disruption while the people doing the actual work are told to adapt.

The winners are concentrated. The losers are everywhere.

All we can hope is that AI music turns out to be more fad than revolution—that eventually the novelty wears thin and the next rebellious trend in music becomes something wonderfully radical: five musicians walking into a room, plugging in and recording a live band again.