45 Years of Music Production — and AI Just Removed the Final Bottleneck

Most people know me as a software engineer.

Automation.
Systems.
Infrastructure.

That’s the visible side of my work at Subdomain Systems.

But there’s a lesser-known fact:

I’ve been producing music for 45 years.

Long before AI.
Long before streaming platforms.
Back when creating a single track required studios, engineers, tape machines—and most unpredictably—human vocalists.

And that last part has always been the hardest variable in the system.

The Hidden Bottleneck in Music Production

Modern music production looks very different than it did when I started.

Today, most music is created inside a DAW — a Digital Audio Workstation.

A DAW essentially turns a computer into a full recording studio.

With it, a single producer can:

• Record multiple tracks
• Layer instruments and vocals
• Edit performances in real time
• Adjust timing and pitch
• Mix and master inside one system

For producers like me, this was a game changer.

Instead of relying on large studios and teams of engineers just to capture a performance, an individual producer could now multitrack, shape, and refine a song directly inside the machine.

Ideas could move from imagination to production far faster than ever before.

But anyone who has produced music for a long time eventually learns something uncomfortable:

The technology isn’t the bottleneck.
The workflow isn’t the bottleneck.

The bottleneck is people.

Not because of a lack of talent—the talent is rarely the problem.

But the day-to-day reality of production often looks like this:

• Scheduling conflicts
• Vocal fatigue
• Endless retakes chasing the perfect performance
• Creative disagreements
• “That’s not really my style” conversations

A producer can hear the finished song in their head long before it exists.

But historically, translating that vision into reality still required coordinating multiple humans, studios, and schedules.

And we humans—brilliant as we are—remain the most unpredictable component in any system.

Five Years of AI Vocal Experiments

I started experimenting with AI vocals more than five years ago.

Early models were primitive.

Tone control was limited.
Phrasing could feel mechanical.
Artifacts were common.

But even then, one thing was obvious:

This was going to change music production.

The technology simply needed time to mature.

That moment has now arrived.

And if I’m being completely honest — it’s a little unsettling.

Because once AI can compete with one of the most human art forms we have — the human voice — the implications become enormous.

Singing is not just sound.

It’s emotion.
Expression.
Identity.

For centuries it has been one of the most human forms of artistic communication.

So if AI can convincingly step into that role…

Then the ceiling for what AI might eventually surpass becomes difficult to define.

Not just in music.

In almost every creative domain.

And that realization is, frankly, beyond a little scary.

Because once a machine can convincingly carry emotional performance — the rest of the creative stack becomes fair game.

The limits of what AI might exceed start looking… well, almost limitless.

A Demonstration: I AM AI

To illustrate the magnitude of this shift, I recently produced a track titled:

“I AM AI.”

Every vocal on the track is 100% AI generated.

No studio vocalist.
No session rapper.
No human singer.

Just a producer, a concept, and a machine capable of executing the vocal performance exactly as envisioned.

The result isn’t a demo.

It’s a finished production.

Listen here:

I AM AI →

The Sound of the Experiment

The track intentionally avoids modern pop polish.

Instead it lives in a darker, heavier space.

Soulful grunge art rock with cinematic weight and raw analog texture.

Heavy distorted guitars stretch out with wide sustain and layered feedback.
A gritty melodic bass anchors the groove.
Loose acoustic drums swing slightly with a subtle funk pocket.

The vocal delivery begins restrained and intimate—almost fragile—before building into explosive cathartic intensity.

The atmosphere feels like a confession echoing through an empty cathedral.

Dark.
Sacred.
Reflective.

The emotional arc moves through struggle, inner war, redemption, and breakthrough light.

Tempo sits in the slow-to-mid range, minor key, with dramatic quiet-to-loud crescendos leading to an overwhelming climax.

There’s no glossy radio sheen.

No trap drums.

No pop production tricks.

Just something raw, imperfect, and alive.

Which makes the core idea even more interesting.

Because the voice delivering that emotion?

Is artificial.

What AI Just Changed

Artificial intelligence didn’t just add another tool to the studio.

It removed the final constraint.

AI can now:

• Sing
• Rap
• Harmonize
• Adapt to genre
• Deliver unlimited takes
• Never get tired
• Never complain
• Never miss a session

From a production standpoint, this is the equivalent of moving from hand tools to industrial machinery.

The creative bottleneck disappears.

The producer’s imagination becomes the only limit.

The Producer Becomes the Architect

This shift will feel familiar to anyone in software engineering.

The work moves from manual execution to intent orchestration.

Instead of coordinating singers, studios, and session schedules, the producer designs the system:

Tone
Emotion
Delivery
Genre
Structure

The machine executes.

And when execution becomes frictionless, creation accelerates.

Where This Experiment Continues

For those interested in hearing where this technology goes next, the latest collection of AI vocal productions from B BRANT can be found here:

Full Send → https://bbrant.hearnow.com/

Hidden throughout that album are traces of a much longer story.

Those curious enough to dig will find breadcrumbs leading back through more than five years of AI vocal experimentation—early prototypes, evolving techniques, and the gradual refinement that led to the sound heard today.

The Real Takeaway

This isn’t about replacing artists.

It’s about removing friction from creativity.

The same way AI is transforming software development, design, and automation, it is now transforming music production.

And for those of us who have spent decades building systems—whether in code or in the studio—the pattern feels familiar.

First the tools get faster.
Then the tools get smarter.
Eventually the tools start participating in the creative process itself.

We are now entering that third phase.

When the tools become powerful enough…

Creation scales.

And when creation scales, something else happens.

The distance between idea and execution begins to collapse.

A melody imagined at midnight can become a finished production before sunrise.
A producer no longer waits for the right voice, the right session, or the right moment.

The machine simply executes the vision.

That changes the creative landscape in ways we are only beginning to understand.

Because once the friction disappears, the limiting factor is no longer technology…
and no longer coordination.

The only remaining constraint is imagination.

And when imagination becomes the only bottleneck, the world fills very quickly with things that previously could not exist.

Which raises a quieter, more unsettling realization.

If AI can already compete with one of the most human forms of expression—the human voice—then the boundary between human creativity and machine capability is no longer where we thought it was.

And once that boundary moves, it rarely moves back.

When the tools become powerful enough…

Creation scales.

And when creation scales,

the people with the clearest vision win.