There’s no shortage of AI chatter in the music world right now, but most of it lives in the land of theory, fear, and hot takes.
In this video below, producer/engineer and educator Warren Huart cuts through that noise with something much more useful: what he’s actually seeing happen day-to-day as a working professional, surrounded by artists, mixers, engineers, and songwriters who are actively making records.
His message is blunt: AI isn’t “coming.” It’s already in the workflow. And the parts of the industry built on “good enough” music are taking the first hit.
Here’s the core of what he says in the video (see below)…
1) “Not theory”: AI is already replacing paid music work (especially in sync)
Warren’s first real-world example is a friend who’s a mastering engineer doing a lot of sync work (music used for film/TV/commercial placements). He describes a large company (he doesn’t name it) that has let go of staff and contractors who were producing “soundalike” tracks for things like documentaries and TV.
Why? Because that type of music is now being generated cheaply through prompts.
This is one of the most important points in the whole video because it’s not abstract. It’s not “AI might take jobs.” It’s: jobs are already being removed in certain categories, and those categories share a common feature:
They’re “generic” by design.
He’s clear about what’s happening economically:
- AI-generated “bed music” is fast and cheap
- companies can avoid paying composers
- in many cases they can avoid royalties
- the result is “good enough” for background use
In other words: if the business goal is “fill the space with something that sounds like ___,” AI is a perfect weapon.
2) The first casualty is “generic” music (and there’s not much you can do to stop that)
Warren says he predicted this a couple of years back: generic music gets killed off first.
He gives examples like:
- “Beach Boys-style backing track”
- reggae instrumental
- mood/background documentary cues
When the requirement is “evoke a vibe, don’t distract anyone,” AI is well-suited. And that’s why this sector is being hit early.
The hard truth here (and Warren doesn’t sugar-coat it): there isn’t much you can do to prevent AI from flooding the lowest-value, highest-volume parts of the market.
So rather than wasting energy trying to “win” that game, he pivots to the better question: How do artists stay valuable?
3) AI is already being used for songwriting “starter tracks” (then humans rebuild it)
Next, Warren describes a workflow that’s happening regularly:
- Songwriters use AI to generate tracks (full song ideas, backing tracks, rough productions).
- Even if platforms remove features like stem downloads, people still extract parts (he mentions methods like using audio tools to separate elements).
- Real musicians then replay the parts and rebuild the production with human performances.
That’s a key nuance: a lot of the music you’ll hear in the future might not sound like “AI music,” because humans will re-track it, re-produce it, and polish it.
So the question “Can listeners tell?” isn’t the point. The point is:
AI is becoming the sketchpad. That means more people will produce more “finished songs” faster than ever before.
4) Expect a flood: more music, more releases, more competition for attention
Warren predicts a near-term outcome that should matter to every independent artist:
- More material will be made than ever before
- Successful artists will release more music than ever before
- AI will handle some of the “heavy lifting” at the idea stage
This ties into something that readers of this blog already know: streaming is already overloaded. Add AI-assisted creation and we’re heading toward a world where volume becomes even more extreme.
This doesn’t automatically mean “great music won’t win.” It means the environment will be noisier, faster, and more saturated.
5) The real opportunity: stop worshipping “perfection” (AI loves perfection)
Here’s where Warren flips the whole conversation. He argues that our obsession with perfection is part of what made AI so effective in music.
He calls out the modern culture of:
- tight editing
- perfectly tuned vocals
- sample-replaced drums
- copy/paste production
- ultra-polished performances
- “shred” clips that are edited to death and mimed on video
His point: when everyone is chasing the same hyper-clean standard, music becomes more uniform… and uniformity is exactly what AI can replicate.
He even gives a vivid example of how this culture shows up online: people attacking singers for being “out of tune” when they’re actually doing expressive pitch movement (he references Steven Tyler being criticized for “bluesy” note choices because listeners expect “stepped autotune”).
So his advice is basically:
If you build your music around flawless execution, you’re moving closer to what AI and the machines do best.
6) “Classic” records didn’t get there by being perfect
Warren talks about a period from the late 90s into the 2000s when DAWs took over, drum samples became standard, and everything got tightened and tuned.
His claim: a lot of records from that era don’t feel “classic” anymore because they sound like a technical template. Loud, slamming, polished… and kind of interchangeable.
He mentions how today anyone can get the same wall-of-guitars sound, the same big drums, the same tightness, using the same tools and plugins.
So what’s left?
What you bring to the party.
That’s his phrase. And it’s the line every artist should write on a sticky note and slap on their monitor.
7) The winning edge: individuality, human performance, and truth
Warren gives examples of what he values in music:
- his “crappy” slightly out-of-tune piano because it has character
- keys that sound different from each other
- “weirdness” and imperfection as identity
He connects this to artists who weren’t “perfect” in a modern sense but were massively important because they were distinct (he references names like The Velvet Underground and Iggy Pop alongside more technically immaculate music).
Then he drops the philosophy bomb (quoting a story involving John Lennon):
“Tell the truth and make it rhyme.”
That becomes his north star for what AI can’t replace:
- great performances
- great ideas
- lyrics that mean something
- music with real identity and emotion
In his view, disposable music is where AI thrives. But music with a point of view is where humans still dominate.
8) The practical takeaway for artists and producers
Warren isn’t saying “never use AI.” He’s realistic: people will use it as a tool, and in many cases it will help speed up early-stage creation.
But he strongly warns against letting it steer the entire direction of your music.
His challenge to artists is simple:
- stop chasing generic perfection
- lean into what makes you human
- write and perform like you mean it
- create work that’s hard to imitate because it comes from lived experience, not templates
Or, as he puts it: Don’t let it win.
My final thoughts
From an AMB perspective, this is the real business shift underneath the AI debate:
- Commodity music (background, soundalikes, filler) is being automated first.
- Distinct artists will still win attention, but only if they stop trying to sound like everyone else.
- The “perfect and generic” middle is the danger zone.
AI is going to flood the market with “acceptable.” That means the value of identity, performance, and meaning goes up, not down.
And that’s a strange twist of fate: the more machines show up, the more human artists need to become.
