I recently watched a TED2025 talk by Dustin Ballard titled “Is AI Ruining Music?” and for anyone working at the intersection of music, creativity, and AI, it’s one of the most thoughtful treatments of the subject I’ve seen so far.
Ballard isn’t a technologist lecturing from the outside. He’s a musician, a creative director, and the mind behind the parody music channel There, I Ruined It. He’s also someone who clearly loves music, not just as content, but as culture, history, and emotional expression.
The talk opens with a clever misdirection: a haunting blues performance that feels deeply human, raw, and lived-in… only for Ballard to reveal that AI generated the entire thing. No man. No guitar. No past. Just a prompt.
That moment is unsettling by design. And it does what good art is supposed to do: it forces us to ask better questions.
By the end of the talk, Ballard offers three simple, but deceptively profound, questions that he personally uses as guardrails when working with AI in music:
Is it deceptive?
Is there artistic intent behind it?
How does this affect musicians?
Those questions have been echoing in my head ever since. And I want to respond to them, not theoretically, but personally, because like Ballard, I’m not watching this from the sidelines. I’m deep in it.
This is the question that makes everyone uncomfortable, and rightly so.
When Ballard plays that AI-generated blues song at the beginning of his talk, most of us feel something before we know something. The emotion lands first. The provenance comes later. And when we discover the truth, it forces a reckoning: did the song move us because of what it was… or because of what we thought it was?
That’s the danger zone.
For me, deception isn’t about whether an artist uses AI. It’s about whether AI is hidden in a way that manipulates trust.
If a song is presented as a lost archival recording, a resurrected artist, or an authentic human performance when it isn’t, that crosses a line. Not because AI was involved, but because the listener was misled about the source of meaning.
Music is relational. Even when we don’t consciously think about it, we’re responding to a story: Who made this? Why did they make it? What does it say about being human?
When AI erases or falsifies that story, it cheapens the exchange.
That’s why transparency matters, not as a legal checkbox, but as an ethical posture. I don’t believe every song needs a technical disclosure label. But I do think creators have a responsibility not to exploit emotional authenticity they didn’t earn.
AI can assist. AI can transform. AI can collaborate.
But AI should not impersonate humanity without consent or context.
This, to me, is the most crucial question of the three.
Ballard draws a sharp distinction between using AI as part of a creative process and mass-producing AI music purely for scale, speed, or monetization. That distinction matters.
Intent changes everything.
Using AI to:
Explore alternate arrangements
Prototype melodies
Translate an idea into a new style
Experiment with sound design
Overcome creative blocks
…is fundamentally different from flooding Spotify with thousands of algorithmically generated tracks designed to game discovery systems.
One is art-driven.
The other is extraction-driven.
As a songwriter, I recognize that tools have always shaped creativity. Synthesizers. Drum machines. DAWs. Sampling. Auto-tune. None of these destroyed music. They changed it. And in the right hands, they expanded what was possible.
AI is no different—except for one thing.
It collapses friction.
And when friction disappears, intent becomes the only meaningful filter.
If the goal is expression, storytelling, connection, or experimentation, AI can genuinely extend human creativity. I’ve experienced that firsthand. Used thoughtfully, it can help surface ideas that still require human judgment to matter.
But if the goal is volume without voice, output without authorship, or profit without participation, then we’re not talking about music anymore. We’re talking about noise.
This is where the conversation has to slow down.
Because the fear many musicians feel isn’t abstract. It’s economic. It’s existential. It’s deeply personal.
Ballard makes an important observation: in many cases, the artists whose voices or styles he playfully borrows end up sharing his work. That doesn’t make everything okay, but it does suggest that not all AI usage is inherently adversarial.
Still, we can’t ignore the power imbalance.
When AI models are trained on massive bodies of work without consent, compensation, or attribution, musicians are right to feel violated. When AI is positioned as a replacement rather than a collaborator, the threat becomes real.
But here’s where I land, even amid the tension:
AI will not eliminate musicians.
But it will expose what kind of musicians we are willing to be.
Artists who rely solely on surface-level output—generic styles, interchangeable sounds, formulaic production—are more vulnerable. Not because they lack talent, but because those patterns are easier to replicate.
Artists who bring identity, values, lived experience, and intention into their work will become more critical, not less.
Ironically, the rise of AI may make human authorship more valuable rather than obsolete.
But only if we defend it.
That means:
Advocating for ethical training practices
Supporting consent-based voice and style use
Valuing musicians as creators, not datasets
Teaching audiences to care about origin, not just output
The future of music isn’t about rejecting AI.
It’s about refusing to let AI redefine what music is for.
I’ll admit there was a moment in the talk that unsettled me.
The blues song played at the beginning was old, weathered, full of ache. The kind of voice that sounds like it’s carrying decades of grief and grace in equal measure. I felt it immediately.
And then the truth came out.
The singer never existed.
The pain was synthesized.
The story was invented.
I didn’t feel angry.
I felt… disoriented.
Because something real happened inside me, even though the source wasn’t, and that’s when I realized what this conversation about AI and music is really about.
It’s not about whether a machine can make a song.
It’s about whether meaning can survive without a soul behind it.
Music, at its best, is a harbor.
A place where human emotion finds shelter.
Where longing, joy, regret, faith, and love are given form.
That’s why deception feels wrong.
That’s why intent matters.
That’s why musicians matter.
When AI is used as a tool in the service of expression, it can amplify what’s already human. But when it becomes a substitute for presence—for lived experience, for truth—it creates an echo without an origin.
A sound with no anchor.
Soul Harbor, the AI group I formed two years ago, still exists because I believe music should be a place of rest, not confusion. It’s my desire that Soul Harbor be a place where listeners feel held rather than manipulated—a place where tools follow values, and not the other way around.
I don’t think so. But it is revealing our assumptions.
It’s forcing us to ask whether music is just sound or whether it’s an expression.
Whether creativity is output or intent.
Whether art is scalable or sacred.
Like Ballard, I believe AI can increase creativity when human values guide it. I’ve seen it. I’ve felt it. I’ve used it in ways that deepened, not diminished, my relationship with music.
The danger isn’t that AI can make songs. The danger is that we forget why we make them in the first place. And that’s a question no machine can answer for us.
Even though AI can help shape sound, only humans can shape meaning. In the end, that’s the line I’m unwilling to cross.
What’s your opinion? Is AI ruining music? If you think so, tell us why.
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.