Are AI Wedding Songs the Future—or a Red Flag? A DJ’s Honest Take on Music, Emotion, and Authenticity

Artificial intelligence can now write wedding songs on command—from first-dance ballads to cocktail hour instrumentals—but the question no one seems to ask is the most important one: should it? As a wedding DJ who's watched music evolve through vinyl, CDs, MP3s, streaming—and now AI—I have thoughts…big ones. If you've ever wondered whether AI belongs at a wedding, or whether robot-generated music can truly replace human emotion, you're going to want to read this. 

The Delta Blues: Where folklore, fear, and music collide.

December 4, 2025

In my first conversation with the couple, they talked at length about their love of the horror genre and how they wanted to lean into that for their Halloween wedding. I remember being so incredibly excited. I am also a huge horror fan, and the prospect of a genuinely frightening wedding thrilled me.

I don't remember how we got there, but that first conversation soon turned to the Ryan Coogler film Sinners. The couple had only recently seen the film, but they could not stop talking about its influence and how much they loved it. Frankly, who didn't? It was one of the best horror films of the last 20 years. And what a soundtrack! Among its many accomplishments was introducing a younger audience to the blues. Not blues as a museum piece. Not blues as a chapter in a music history textbook. The real thing. Raw, emotional, deeply human music that serves as the beating heart of the film. They told me that they were now blues fans and they wanted to incorporate the genre into their playlist. I was not surprised. I suspect many younger viewers encountered artists and traditions through Sinners that they otherwise might never have explored. As somebody who has loved the blues for most of his life, I enthusiastically awaited their list of must plays.

The Mississippi Delta is the birthplace of legendary, horror-themed blues, and the blues genre has always possessed a complicated relationship with darkness. It is deeply rooted in folklore, death, and the supernatural. Long before Hollywood discovered horror, blues musicians were filling songs with devils, crossroads bargains, restless spirits, bad omens, and unseen forces lurking just beyond the edge of the firelight, and the greatest horror-themed Delta blues songs masterfully combine eerie lyrics, ghostly slide guitar, and haunting vocals. Robert Johnson's "Me and the Devil Blues," for instance, is the ultimate crossroads tale. The song kicks off with the chilling lyric: "Early this morning, when you knocked upon my door, I said, 'Hello, Satan, I believe it's time to go.'" It perfectly captures the mythos of Johnson supposedly selling his soul to the devil in exchange for his supernatural guitar skills. Similarly, Johnson's "Hellhound on My Trail" recounts Johnson's attempt to outrun the devil when he comes to collect his due. That track still sounds unsettling nearly a century after it was recorded. Skip James's "Devil Got My Woman" is another classic example. Driven by eerie falsetto vocals and an unsettling minor-key tuning, the track features a harrowing, depressive atmosphere as James sings about forces beyond his control claiming his love and soul. Son House gave us "Death Letter Blues," a haunting retelling of receiving a letter announcing the death of a loved one complete with frantic slide guitar and visceral, trembling vocals make you feel like you are standing right in front of a grave. And then, tere is Howlin' Wolf's "Evil (Is Going On)" which practically drips with dread from the opening note. It is an absolute masterpiece of dark, menacing blues. Written by the legendary Willie Dixon, the track is a masterclass in psychological horror and paranoia. It carries all the dread, shadows, and superstition of the old Mississippi crossroads. 

Finally, I received their song requests. To say I was disappointed would be an understatement. None of these tracks were listed. Instead, I found something entirely different.

Every couple leaves fingerprints throughout their planning forms. Sometimes it is an unusual first dance selection. Sometimes it is an obscure song hidden deep inside a dinner playlist. Sometimes it is a must-play request that tells me something about them before we have even met. This couple's fingerprint appeared immediately. Sitting at the very top of their must-play list—the very first song they wanted played when the dance floor opened—was a blues version of Michael Jackson's "Thriller."

Not "Thriller."

A blues version of "Thriller."

Naturally, I was intrigued.

It was not the one of the tracks I had expected. But, I thought to myself, maybe it's even better. I have spent most of my life around music, and one of my favorite things in the world is hearing a familiar song transformed into something unexpected. Some songs become so deeply intertwined with a particular arrangement that it becomes difficult to imagine them existing any other way. "Thriller" is certainly one of those songs. Few recordings are more closely tied to their production. Remove the iconic groove, the synthesizers, Vincent Price lurking somewhere in the darkness, and much of what people think they know about the song disappears with it. The idea of stripping all of that away and rebuilding it as traditional blues seemed unusual enough that I immediately wanted to hear it.

When I finally sat down to listen to the recording, I expected a novelty.

What I heard was something else entirely.

The opening guitar emerged alone, dry and weathered, with just enough grit in its tone to sound as though the amplifier had already survived decades of smoky bars and hard miles. The familiar urgency of Michael Jackson's original disappeared almost immediately. Gone was the sleek pop production. Gone was the sense of spectacle. In its place sat a sparse Delta blues arrangement built on patience and atmosphere. Every chord seemed willing to linger for an extra moment before reluctantly giving way to the next. The song moved slowly, confidently, and without apology.

What impressed me most was its restraint. The arrangement never seemed interested in showing off. The guitar occupied a narrow lane, offering sparse fills and mournful responses to the vocal without ever demanding attention for itself. A subtle slide guitar drifted in and out of the mix like fog rolling across a Louisiana bayou at dusk. The harmony remained deceptively simple, allowing the tension to emerge from the relationship between the melody and the singer's phrasing rather than from complicated chord changes. Beneath it all sat a slow pulse that felt more implied than performed. The groove did not push forward. It stalked.

And then there was the voice.

As every great blues musician understands, the voice is the most important instrument in the room. Everything else exists to support it. The singer delivered the lyrics with a gravelly weariness that sounded decades old. There was grit in the phrasing and weight behind the words. The melody occasionally bent against the harmony in that uniquely blues fashion, creating moments of dissonance that never quite resolved the way modern pop music expects them to. The result was haunting. Not Halloween haunted. Human haunted. The kind of haunted that sounds as though the singer has survived something difficult and is still carrying pieces of it around with him.

What fascinated me most was how naturally "Thriller" fit inside the genre. The song's DNA had always contained darkness. Strip away the iconic production and what remains is a story about fear, paranoia, pursuit, and dread. In Michael Jackson's hands, those emotions became theatrical. In the language of the blues, they became intimate. The monsters no longer felt cinematic. They felt personal. The shadows felt closer. The danger felt real. By the second chorus, I found myself wondering why nobody had thought of the concept sooner. "Thriller" as Delta blues is a genuinely brilliant idea.

The more I listened, the more I admired it. The more I admired it, the more impressed I became. By the time the song ended, I was already mentally crediting some unknown musician for a remarkably clever reinterpretation of Michael Jackson's original. In fact, one of the first things I did afterward was try to learn more about the artist responsible for it. I wanted to hear other recordings. I wanted to know whether this was part of a larger project. Mostly, I wanted to give credit where credit was due because transforming "Thriller" into a convincing Delta blues recording is exactly the sort of creative risk I tend to celebrate. Instead, I discovered that there was no artist to find. The performance I had just spent several minutes admiring had never passed through a recording studio, never emerged from a rehearsal space, and never existed inside the imagination of a musician. The singer whose voice sounded weathered by experience did not exist. The creative instincts I thought I was hearing belonged to no one. What I had assumed was interpretation turned out to be generation. The entire recording had been created by artificial intelligence. I remember sitting there for a moment staring at the screen, not angry or offended so much as genuinely confused by my own reaction. The song was still good. Everything I had admired about it remained exactly as it had been five minutes earlier. Nothing about the recording itself had changed. Yet somehow the knowledge that no human being had actually created it altered the way I heard it, and I found myself carrying that feeling long after the song had ended.

To be clear, the source of my discomfort was not the idea of reimagining a familiar song.

Quite the opposite.

I have always loved artists who take existing music and transform it into something entirely new. In many ways, that is what great music has always done. One generation borrows from another. One artist hears a familiar melody and discovers a different way to tell the story. The best cover songs do not simply reproduce what already exists. They reveal something that was there all along, waiting for the right artist to uncover it.

For years, Scott Bradlee's Postmodern Jukebox has built an entire musical project around that very idea. Modern pop songs become jazz standards. Rock songs become ragtime numbers. Hip-hop becomes swing. Familiar melodies are transported into entirely different musical worlds, often revealing qualities that were hidden in the original recordings. The appeal is not novelty for novelty's sake. The appeal is discovery. The listener hears a song they already know and suddenly understands it from a different perspective. That transformation is what makes the experience worthwhile.

Some of my favorite recordings fit that description perfectly.

When Regina Spektor performs "While My Guitar Gently Weeps," she approaches the song from an entirely different emotional angle than George Harrison. The original carries a quiet sadness, but Spektor's version feels more fragile, more exposed, as though the song itself has been stripped of its defenses. A similar transformation occurs in KT Tunstall's interpretation of "Boys of Summer." Don Henley's original is built upon motion and memory. Tunstall slows everything down and allows the loneliness to linger. What once felt nostalgic begins to feel isolated. The scenery remains the same, but the emotional landscape changes completely.

The darkness deepens with Disturbed's version of "The Sound of Silence." Simon and Garfunkel's original feels intimate, almost conversational, as though the listener has stumbled upon a private reflection. Disturbed turns the same song into something much larger and far more ominous. David Draiman's voice carries the weight of a prophecy. The lyrics no longer feel personal. They feel apocalyptic. The result is unsettling in a way the original never attempts to be.

And then there is Chris Cornell.

The first time I heard Cornell's version of "Billie Jean," I remember feeling almost disoriented. Michael Jackson's original is so deeply embedded in popular culture that it is difficult to separate the song from the recording itself. The groove is iconic. The production is iconic. The performance is iconic. Yet Cornell strips all of it away and forces the listener to confront the lyrics without the distraction of movement. Suddenly the song becomes something darker than I had ever realized. The story is no longer carried by rhythm. It is carried by dread. What emerges is a portrait of manipulation, accusation, obsession, and paranoia. The monsters inhabiting the song are no longer supernatural creatures lurking in the shadows. They are people. Real people. The darkness becomes more intimate, and therefore far more frightening.

That transformation fascinates me because it could only have happened through interpretation.

Somewhere along the way, Cornell heard something inside "Billie Jean" that most of us had overlooked. He recognized an emotional truth hiding beneath the surface and built an entirely new performance around it. Every choice serves that discovery. The slower tempo. The sparse arrangement. The lingering pauses. Even the silences feel intentional. Listening to the recording, you can almost hear the artist thinking his way through the song.

And that is precisely why the blues version of "Thriller" unsettled me.

Both recordings accomplish something remarkably similar. Both take familiar material and reveal a darker emotional core. Both strip away the production that originally defined the songs. Both leave the listener hearing something entirely different than what existed before. Had I encountered the two recordings back-to-back without any context, I would have praised them for exactly the same reason: they felt creative.

The difference is that Chris Cornell leaves fingerprints all over "Billie Jean."

His experiences are there. His instincts are there. His influences are there. His interpretation is there. The recording becomes a conversation between two artists separated by time, style, and circumstance but connected through the same piece of music.

The blues version of "Thriller" creates the illusion of that same conversation.

The longer I sat with that realization, the more I found myself wondering whether the song had disturbed me for a reason entirely unrelated to technology. Perhaps what bothered me was not that artificial intelligence had created something convincing. Perhaps what bothered me was that it had created something that sounded so convincingly human.

And that question led me somewhere I did not expect.

It led me back to Elvis Presley.

At first glance, Elvis Presley might seem like an odd place to look for answers about artificial intelligence.

Yet the more I thought about the blues version of "Thriller," the more I found myself returning to a question that has followed popular music for generations. What happens when one artist takes something that belongs to another culture, another tradition, or another musical world and makes it their own? It is hardly a new question. In fact, it sits at the very heart of American music.

Few examples illustrate the debate better than "Hound Dog."

When most people hear the title today, they think of Elvis. They think of the swagger, the television appearances, and the explosive energy of the 1956 recording that helped turn him into a cultural phenomenon. What many listeners do not realize is that the song belonged to Big Mama Thornton first. Written by Jerry Leiber and Mike Stoller and recorded by Thornton in 1952, "Hound Dog" emerged from an entirely different musical tradition than the one Elvis would later bring to mainstream audiences. The two recordings share lyrics and little else. Thornton's version is tougher, bluesier, and considerably more confrontational. Elvis transformed the song into something entirely different.

For decades, that transformation has fueled difficult conversations about race, appropriation, influence, and ownership.

Some critics have argued that Elvis benefited from a system that routinely rewarded white performers for music that originated in Black communities. There is truth in that criticism. The history of American music contains countless examples of Black artists receiving less recognition, fewer opportunities, and smaller financial rewards than the white musicians who later popularized similar sounds. Ignoring that reality would be dishonest.

Yet reducing Elvis to a simple story of theft misses something equally important.

Elvis did not discover Black music from a distance. He grew up surrounded by it. He listened to it. Studied it. Loved it. The gospel he sang, the blues he absorbed, and the rhythm and blues records he devoured were not commodities to him. They were influences. They shaped the musician he became. When Elvis stepped into a recording studio, he was not attempting to imitate a tradition he did not understand. He was participating in one he genuinely admired.

That distinction matters.

Whether one loves Elvis or dislikes him, his recordings emerged from a human relationship with music. They emerged from curiosity, admiration, influence, and experience. They emerged from countless hours spent listening, learning, and absorbing the work of other artists. Most importantly, they emerged from a conversation.

Music has always worked that way.

Robert Johnson influenced Muddy Waters. Muddy Waters influenced the Rolling Stones. The Rolling Stones influenced countless rock bands that followed. The Beatles borrowed from Chuck Berry. Eric Clapton built portions of his career studying blues musicians whose names many listeners have never heard. Every artist enters an ongoing conversation that began long before they arrived and will continue long after they leave. The greatest musicians do not create in isolation. They respond. They borrow. They reinterpret. They argue. They celebrate. They contribute.

That is precisely what Chris Cornell was doing when he transformed "Billie Jean."

And it is precisely what the blues version of "Thriller" seemed to be doing.

The song sounded as though it belonged to that same conversation. It sounded as though somebody had spent years listening to Delta blues records. It sounded as though somebody understood why Robert Johnson's slide guitar still sounds eerie a century later. It sounded as though somebody had fallen in love with Howlin' Wolf, Skip James, Son House, Willie Dixon, and the countless musicians who built the foundation of the genre.

But there was no "somebody."

And that realization changed everything.

The blues version of "Thriller" had borrowed the language of the blues without ever participating in the culture that created it. It had absorbed the sound without experiencing the struggle. It had reproduced the style without engaging in the conversation. The recording knew what the blues sounded like. It had no idea what the blues meant.

The more I thought about that distinction, the more unsettling it became.

Not because the recording was bad. Quite the opposite. Had the song been terrible, the entire matter would have been easy to dismiss. Nobody worries about a failed imitation. Nobody loses sleep over a gimmick. The blues version of "Thriller" lingered because it succeeded. It sounded authentic. It sounded thoughtful. It sounded like the work of somebody who genuinely loved the music that inspired it.

And yet, the deeper I dug into my own reaction, the more I found myself returning to a simple truth that extends far beyond the blues.

Art has never been merely about the finished product.

When I listen to Robert Johnson, I am not simply listening to a collection of notes. I am listening to Robert Johnson. When I hear Howlin' Wolf growl his way through "Evil (Is Going On)," I am hearing a personality as much as a performance. The same is true of every artist whose work has ever meant something to me. Their recordings carry traces of their lives. Their experiences shape the music. Their triumphs shape the music. Their failures shape the music. Even when we know very little about the people behind the recordings, we instinctively understand that another human being stands on the other side of the speaker.

That relationship changes everything.

A love letter means something because somebody wrote it. A photograph means something because somebody took it. A wedding toast means something because it emerges from a lifetime of shared experiences between the speaker and the couple standing before them. Remove the human being from any of those things and something essential disappears. The words may remain. The image may remain. The performance may remain. Yet the meaning changes because meaning has always been tied to the person who created it.

The more I sat with the blues version of "Thriller," the more I realized that what unsettled me was not the recording itself. It was the absence hiding behind it. Every instinct told me there was a musician somewhere behind the curtain. Somebody making choices. Somebody taking risks. Somebody trying to communicate an idea. The arrangement sounded intentional. The vocal sounded intentional. The atmosphere sounded intentional. Everything about the recording pointed toward the existence of an artist.

There wasn't one.

Or at least not in the way we have traditionally understood the word.

And that realization opened the door to a far larger concern than a single Halloween wedding song.

The blues version of "Thriller" may have been the first AI-generated song that genuinely stopped me in my tracks, but it certainly wasn't the first AI-generated song I had encountered. By that point, artificial intelligence had already begun quietly infiltrating nearly every corner of the music industry. Entire albums were appearing online with no identifiable musicians attached to them. Streaming services were finding themselves flooded with AI-generated content. Record labels, artists, lawyers, and technology companies were all scrambling to answer questions that nobody seemed fully prepared to confront. The technology was advancing faster than the conversation surrounding it.

For musicians, the implications were immediate.

Imagine spending decades learning an instrument. Imagine sacrificing weekends, holidays, relationships, and countless hours of sleep chasing mastery of a craft. Imagine developing a unique voice, a recognizable style, or a distinctive approach to songwriting. Then imagine discovering that a machine can analyze thousands of recordings, absorb the characteristics that make your work unique, and generate something remarkably similar in a matter of seconds.

Whether one views that possibility as exciting or terrifying depends largely upon which side of the equation they occupy.

For listeners, artificial intelligence often arrives as convenience. For musicians, it frequently arrives as competition.

That distinction matters because music has never been merely entertainment. For many artists, it is a livelihood. The songs we stream while driving to work or cleaning the house represent years of labor. Every album contains thousands of decisions invisible to the listener. Every recording reflects countless hours spent learning, practicing, failing, revising, and improving. The finished product may last only three or four minutes, but the journey that produced it often spans decades.

Artificial intelligence bypasses that journey entirely.

The machine does not spend ten years learning guitar before composing a blues song. It does not play empty bars to indifferent audiences. It does not write bad songs before learning how to write good ones. It does not suffer creative droughts, experience heartbreak, lose confidence, develop confidence, or slowly discover its artistic identity through trial and error. Instead, it begins at the finish line, trained upon the work of countless musicians who endured all of those experiences on its behalf.

The more I thought about that reality, the harder it became to ignore an uncomfortable question.

If the machine's education comes from human creativity, what exactly do those humans receive in return?

That question sits at the heart of nearly every major legal battle surrounding artificial intelligence today. Musicians, authors, visual artists, filmmakers, and journalists have all raised versions of the same concern. Their work has become the raw material from which these systems learn. Entire careers have been absorbed into datasets. Entire catalogs have become training material. Yet in many cases, the creators themselves were never asked for permission, never compensated, and never given meaningful control over how their work would ultimately be used.

The issue becomes even more complicated when we remember that copyright law was never designed for a world like this.

For generations, copyright rested upon a relatively simple assumption: creative works originate with human beings. A songwriter writes a song. A painter paints a painting. A novelist writes a novel. The law concerns itself with protecting the rights of the creator because the creator is easy to identify.

Artificial intelligence disrupts that framework.

If a machine generates a song, who owns it? The person who typed the prompt? The company that built the model? The artists whose work helped train it? The answer remains frustratingly unclear because the technology has arrived faster than our ability to define its boundaries.

And if ownership becomes difficult to determine, authorship becomes even harder.

Because beneath every legal question, every copyright dispute, and every lawsuit sits a much older question.

What happens when creativity becomes detached from the creator?

The more I explore the question, the more I find myself occupying an uncomfortable middle ground. The technology fascinates me. It would be dishonest to pretend otherwise. I have seen AI-generated images that are breathtaking. I have read AI-generated passages that are genuinely impressive. And I have certainly heard AI-generated songs that made me stop what I was doing and listen. The blues version of "Thriller" is proof of that. Yet fascination is not the same thing as comfort. Those are two very different emotions, and the longer I sat with that recording, the more I realized I was experiencing both simultaneously.

What lingered in my mind was not what artificial intelligence could create. What lingered was the possibility of forgetting where creation comes from. Music has never existed merely to produce pleasant sounds. Art has never existed merely to generate content. The things we create are extensions of ourselves. They preserve experiences, perspectives, fears, hopes, failures, triumphs, and moments in time. They allow us to communicate pieces of our humanity to people we may never meet. That human connection is not incidental to the work. It is the reason the work matters.

When Robert Johnson recorded "Hellhound on My Trail," he was not creating content for future consumption. When Howlin' Wolf recorded "Evil (Is Going On)," he was not generating material for an algorithm to study. When Chris Cornell transformed "Billie Jean," he was not manufacturing product. Each artist was responding to the world around him and leaving behind something uniquely personal in the process. We often use the words art and content interchangeably, but I am no longer convinced they mean the same thing. If anything, the rise of artificial intelligence has made that distinction feel more important than ever.

That distinction matters more than ever because artificial intelligence is no longer some fringe experiment tucked away in a dusty experimental corner of the internet. If you think AI-generated songs are still a novelty reserved for curious technologists and internet hobbyists, think again. Here’s a number that nearly made me spill my coffee: as of late 2025, nearly 28% of all new tracks uploaded to the streaming platform Deezer were fully AI-generated. That’s roughly 30,000 new AI songs per day.

And before we applaud the brave new digital frontier, here’s the darker side: up to 70% of streams on those AI tracks have been tied to fraudulent activity. Streaming farms. Bots. Automated systems designed to generate plays and siphon revenue from the same royalty pools upon which working musicians depend. Suddenly, the conversation is no longer about creativity. It is about economics. It is about visibility. It is about whether real artists could still be heard in an environment increasingly crowded by algorithmically generated material.

As a DJ, that part resonates with me more than the technology itself.

Every weekend, I spend hours searching for music. I dig through playlists, revisit forgotten albums, explore new releases, and hunt for songs that might connect with a particular couple or audience. Discovery has always been part of the job. Great songs find their way into weddings because somebody created them, somebody shared them, and eventually somebody else fell in love with them. That process has always felt organic. Messy at times, certainly. Imperfect, absolutely. But fundamentally human.

Now imagine trying to navigate a landscape flooded with tens of thousands of newly generated songs every day. Imagine searching Spotify for a first-dance song and finding yourself wading through an endless sea of titles that sound almost right but somehow feel hollow. Imagine playlist after playlist filled with songs designed not to express something meaningful but to satisfy a prompt.:

• “Love Dance Wedding Ballad ft. AI-Voice #14”

  
  

• “Romantic Marriage Slow Song (No Copyright Needed)"

  
  

• “Bruno Mars-ish Wedding Song (Legally Not Bruno Mars)”

  
  

That's not a playlist—that's a graveyard of emotional counterfeits.  Everything technically functional, yes. But everything emotionally interchangeable, as well. Everything generated because somebody asked a machine to create content that sounded like music rather than music that sounded like somebody's life.

The problem is not merely that artificial intelligence can create songs. The problem is that artificial intelligence can create so many songs that it becomes increasingly difficult to separate meaningful work from manufactured noise. The danger is not that AI will replace every musician. The danger is that genuine artistry becomes harder to find beneath an avalanche of material designed to imitate it. Search results become cluttered. Playlists become crowded. Discovery becomes more difficult. The signal becomes buried beneath the noise.

That is where my concern begins to shift.

No matter how good AI gets, no matter how realistic the vocals become or how clean the production is, something remains missing: the lived experience behind the music.

Eric Clapton didn’t write “Wonderful Tonight” because a predictive model suggested it had a high emotional engagement score—he wrote it while waiting on his wife to get ready for a night out, capturing a moment so ordinary and intimate that it became universal. Pianist Jonathan Cain penned “Faithfully” on the road with Journey—wrestling with love, distance, temptation, and devotion in a way only a touring musician could. John Legend wrote “All of Me” with his future wife sitting beside him, the lyrics reflecting their real flaws and real tenderness. Alicia Keys wrote “If I Ain’t Got You” just weeks after losing her close friend and fellow musician Aaliyah—it was grief and clarity poured into melody. And when Billy Joel wrote “Just the Way You Are,” he wasn’t engineering a hit—he was writing a love letter to his then-wife Elisabeth when their marriage suffered from financial problems following a contract dispute with his label.

Real music is built from truth—sometimes messy, sometimes imperfect, always human.

AI music is built from prediction — clean, calculated, and emotionally neutral.

And perhaps that is why I keep returning to the blues version of "Thriller."

For all of the philosophical questions it raised, it was still a good song. It still made me listen. It still made me think. The problem is that for every thoughtfully constructed experiment, thousands of other AI-generated tracks are being released into the same ecosystem. Some are harmless. Some are clever. Some are forgettable. Together, however, they create a volume problem unlike anything the music industry has ever faced before.

The more I explored that reality, the more I realized that the conversation extends far beyond questions of taste. It reaches into issues of ownership, authorship, compensation, and consent. Because once artificial intelligence begins generating music that sounds convincingly human, another question inevitably follows.

Who owns the voice?

That question sounds deceptively simple until you begin pulling on the thread.

For most of modern history, authorship has been relatively straightforward. A songwriter writes a song. A novelist writes a novel. A painter paints a painting. Copyright law was built around that assumption because the creator was easy to identify. If somebody stole your work, there was little ambiguity about who had been wronged. The law might not always have protected artists as well as it should have, but at least everyone generally agreed upon where the work originated.

Artificial intelligence has complicated that arrangement in ways lawmakers never anticipated.

The blues version of "Thriller" did not emerge from nowhere. It did not materialize out of thin air. Everything I admired about that recording—the phrasing, the atmosphere, the arrangement, the vocal style, the understanding of Delta blues conventions—originated somewhere. Those sounds were learned from recordings made by real musicians. Those techniques were developed by real artists. Those traditions were built by generations of performers whose names most listeners will never know. The machine did not invent the blues. It inherited the blues.

That reality has produced a question that courts, record labels, technology companies, and artists are all struggling to answer: if artificial intelligence learns from human creativity, what obligations does it owe to the humans who taught it?

The answer becomes even murkier when money enters the picture.

A blues musician might spend decades developing a style that becomes instantly recognizable. A songwriter may devote an entire lifetime to mastering a craft. A singer may spend years refining a voice unlike any other. Yet artificial intelligence can absorb the characteristics that make those artists unique and reproduce remarkably similar results in a matter of seconds. The technology does not merely study the work. In many respects, the work becomes the foundation upon which the system is built.

It is easy to understand why many artists find that unsettling.

Imagine discovering that your life's work has become part of a machine's education. Imagine learning that the songs you spent decades writing, recording, and performing are now helping generate recordings that may eventually compete with your own. Imagine knowing that your voice, your style, or your artistic identity can be imitated by somebody who has never met you, studied with you, performed beside you, or shared any of the experiences that shaped your art. Whether one ultimately supports or opposes artificial intelligence, those concerns are difficult to dismiss.

The situation becomes even stranger when the artists themselves are no longer alive.

When I hear AI vocals mimic an artist who never sang those words—Amy Winehouse resurrected to croon a song she never wrote—it hits me wrong. It feels like borrowing a soul that isn't there. The voice sounds familiar. The phrasing sounds familiar. The emotional texture sounds familiar. Everything about the performance points toward the presence of a specific human being. Yet the human being is absent. What remains is a remarkably convincing illusion.

And perhaps that is what makes the entire situation so unsettling.

Music is one of the last sensory experiences that still feels uniquely human. The scratches in a vinyl record. The crack in a singer's voice. The slight rhythm mistake during a garage-band rehearsal. The breath taken between lines. The imperfections matter because they remind us that another person stood behind the work. Those flaws are not defects. They are evidence. They tell us that the song existed in a real place, at a real moment in time, created by a real person wrestling with something worth expressing.

Artificial intelligence challenges that relationship.

The technology can reproduce the sound of a voice without reproducing the life behind it. It can imitate the style without experiencing the struggle. It can mimic the emotion without ever feeling it. The result can be astonishingly convincing. In some cases, it can even be beautiful. Yet the beauty comes attached to a question that grows more difficult to ignore the longer one thinks about it: if the performance sounds exactly like Amy Winehouse, but Amy Winehouse never performed it, whose performance are we actually hearing?

The legal system is still trying to answer that question.

One of the strangest realities surrounding AI-generated music is that fully AI-generated works currently occupy an uncertain legal space. The United States Copyright Office has repeatedly maintained that copyright protection requires human authorship. In other words, art requires an artist. That sounds straightforward until artificial intelligence enters the equation. Suddenly we find ourselves asking whether the creator is the person who wrote the prompt, the company that designed the software, the artists whose recordings trained the model, or some combination of all three.

Nobody seems entirely sure.

Record labels are already filing lawsuits. Artists are already pushing back. Technology companies are already defending their practices. The precedents are forming in real time, often faster than lawmakers can react. Every new case seems to generate more questions than answers.

The deeper I explored the issue, the more I realized that ownership might not even be the most important question. Consent may be.

A living artist can choose whether to record a song. A living artist can decide whether to license a recording, collaborate with another musician, or lend a voice to a project. Dead artists lose that ability. They cannot object. They cannot approve. They cannot participate in the decision at all. Their voices become raw material for somebody else's creative experiment.

That reality becomes difficult to ignore once you start imagining the possibilities. Today it might be Prince singing a song he never recorded. Tomorrow it might be John Lennon. It might be Freddie Mercury. It might be anybody whose voice exists in sufficient quantity for a model to learn from. For some listeners, that may feel exciting. For me, it feels like something closer to grave robbing. The technology can recreate the voice, but it cannot recreate the person. It can imitate the sound, but it cannot reproduce the experiences that gave the sound meaning. What remains is a remarkably convincing illusion of presence.

And perhaps that is what artificial intelligence does best.

It creates the illusion that nobody is missing.

And perhaps that is why I keep returning to the same feeling I had when I first discovered that the blues version of "Thriller" wasn't the work of a musician at all.

The questions it raised continue to bother me, and talking with others in the industry, I have learned that I am not alone. As DJs, we all spend our lives working with music that comes from lived human experience — heartbreak, joy, innocence, chaos, defiance, hope.

When I drop “September” by Earth, Wind & Fire and the dance floor erupts, that’s not an algorithm succeeding. That’s history. That’s memory. That’s culture. That’s humanity. And yet — I also know AI isn't going away. In fact, projections suggest that by 2028, AI-generated music could make up 20–60% of streaming catalogs and threaten up to 23% of musician revenue globally.

So where does that leave me?

Somewhere between curiosity and resistance. Somewhere between possibility and caution. Somewhere between admiration and unease.

If a couple asks for an AI song, I’ll talk with them—honestly. We’ll decide whether it belongs in a meaningful moment or somewhere lighter—maybe cocktail hour or a novelty placement. If they insist, they win. They hire me to play the music they select after all.

But when the lights fade, the room goes quiet, and the first dance begins? I will always advocate for a song written by a human being who meant it. Because music isn’t just something we hear. It’s something we feel. And until AI can feel heartbreak, beauty, fear, longing, hope, and love…humans will always do it better.

Previous Post Return to Blog Next Post