First there was AI images, then there was AI text. It was only a matter of time until AI music came along—and with Suno, it is here. The AI music generator (and others such as Udio) has become my latest obsession. It is simple to use with surprisingly good results. Just enter a text prompt with a subject matter and/or musical genre, and it will quickly produce a 2-minute song of professional quality. Of course the results vary—some songs are better than others. But in a short amount of time I have already created a few gems, such as this 1980s-style electronic anthem for “Time Zone Weird” that I can’t get out of my head:
You can also enter your own lyrics, which is especially intriguing for me because I have many poems and lyrics already written, but never had the means to turn them into complete songs. I wrote before about how I enjoy creating instrumental music on Garageband (and I still do), but I cannot sing—at all. Now with the aid of AI, I can finally start combining my instrumentals and lyrics into full songs. (I plan to start releasing more of my AI-assisted songs in the future.) While I have always had a passion for listening to music, creating music has only ever been a hobby for me. As such, I don’t view AI as a threat—there is only upside for me. However, there is a growing fear of AI among professional musicians.
Many believe AI music generators like Suno will put human musicians out of work. This fear of new technology in the music industry is nothing new. Throughout history, music and technology have always been intrinsically linked. After all, technology allowed instruments to be created in the first place to play music. Later, the phonograph allowed people to record songs. Before that, you could only hear music being played live. The only way to hear Mozart was to visit him in Vienna.
Radio allowed recorded music to be broadcast around the world. Vinyl albums allowed consumers to purchase personal copies of recorded music. Vinyl was replaced by a new more convenient technology: cassette tapes—which were themselves replaced by compact discs (CDs). This proliferation of recorded music via technology put many local live musicians out of business. Why settle for hearing a knock-off of the Beatles live at your local pub when you can hear the actual Beatles at home (or in your car, or on your Walkman)?
There wasn’t a massive backlash from the music industry against these technologies because, up until that point, such technological developments benefitted the industry, the artists, and the consumers. Being able to mass-produce copies of musical recordings resulted in more money for the musical artists—at least the most popular artists—and the industry labels who distributed their albums. Then the internet came along and changed everything.
When MP3s replaced CDs, it could have potentially allowed consumers to buy even more music than ever before because they could store thousands of albums without wasting any physical space. Except Napster spoiled that idea. The peer-to-peer file-sharing service allowed consumers to download unlimited MP3s for free. The music industry had never seen a technological disruption like that before. Napster was great for consumers of music but not for creators of music. People were listening to more music than ever, but the musicians themselves were not financially rewarded for each illegal download.
Napster caused initial chaos. Record sales plummeted, hurting both artists and their labels. Obviously devastated, they responded with lawsuits and legal action. Napster was shut down but quickly replaced by other peer-to-peer filing-sharing services like Morpheus, Kazaa, and the Pirate Bay—the latter of which still stands today. Pandora’s box had been opened. There was no stopping the P2P MP3 revolution.
The music industry eventually realized they could not fight against technology (information wants to be free), so they began to work with the technology. When Apple introduced iTunes, an easy way to purchase individual MP3s for a low price ($0.99), artists started seeing profits again. Consumers were willing to pay for digital music to support the artists they loved if doing so was convenient and fairly priced. But of course, that wasn’t the end of the story. Why waste hard drive space on your computer when you can just stream music for free?
Streaming platforms contain just about every song ever recorded, so there is no longer any need for consumers to purchase albums or even individual songs. As long you have access to the internet, you have access to any song you’d like to hear at any time. This, again, was a fantastic development for music fans but less so for artists and their record companies.
Pre-internet, the bulk of a musician’s profits came from the sale of records, cassettes, and CDs. However, with the convenience of streaming platforms, consumers have stopped purchasing both physical and digital copies of music. The artists and their labels receive some money from these streaming services, but nowhere near as much as they were making before. As a result, they’ve had to adjust their business model.
Musicians started focusing more and more on live performances. Despite the decline in album sales, concert attendance numbers (and ticket prices) have risen. The internet has not been able to replace the experience of physically going to a show with other fans and seeing your favorite artists play live. Whereas before, musicians used concerts as a means to boost their album sales, now they use streaming platforms to boost their concert sales.
So far, live concerts have avoided technological disruption. Recordings of live performances are often streamed/posted for free on YouTube, but online videos fail to fully capture the experience of being there live in person. But will that last forever? If history is any indication, technology will eventually disrupt every market. It’s only a matter of time until virtual reality becomes so good that someone could watch a VR concert from the comfort of their own home and feel like they were actually in attendance.
But what about artificial intelligence? Suno and other AI music is relatively new yet already impressive, and will surely only get better in the future. Why search for new musical artists when you can just enter a prompt for the exact type of music you want to hear? Soon you may not even need to write a prompt. The Spotify algorithm could use your listening history to instantly generate a new song using AI that you are guaranteed to enjoy. Like with AI movies, each individual may listen to songs created for themselves that no one else ever hears. AI music could potentially disrupt the music streaming industry and draw listening hours from human creators, but as with movies and books, I don’t think AI will ever replace human musicians entirely.
Art is subjective, so AI can never definitively be better than humans at creating music. Even if AI is more technically proficient, some listeners will prefer human-created songs. Music is about more than the music; humans are drawn to the humans who create the music. Fans flock to concerts, even though the sound quality is inferior to the studio recording, because they want to see and hear the person in person. AI can never match that.
Sure, some of the songs Suno and Udio have generated are fun and enjoyable to listen to, but I wouldn’t call any of them “great.” I view AI music the same way as AI art and AI writing: it is “competent.” AI is a quick and easy way to achieve average results. Apps like Suno can democratize the music-making process so that more people can create songs of their own, opening up new vistas of creativity. But when I consume art (read a book, watch a movie, or listen to a song), I don’t want average; I want excellence. I seek originality, weirdness, and novelty. I seek transcendence in art—something rarely, if ever, achieved by AI.
Yes, AI is a threat to the music industry—but only as the industry currently operates. Technology is helping to eliminate the need for the middle-man in art: the publishing houses for books, the production studios for movies, and the record labels for music. This allows individual artists to connect more directly with their fans and receive the bulk of the profits themselves, rather than the industry gatekeepers. Any true artist should applaud technology that circumvents gatekeepers and middle-men, which AI has the potential to do.
The music industry will eventually learn to adapt to AI, as it has adapted to every other technological disruption throughout history. The cream of artists will rise to the top, and human listeners will always be drawn to the best human musicians. I suspect that, like every other artistic medium being disrupted by AI, the best music in the future will be created by a combination of humans and AI. Just as human artists are best equipped to maximize the use of DALL-E, and human writers are best equipped to maximize the use of Chat-GPT, human musicians will get the most out of Suno.
Only mediocre musicians should fear being replaced by AI. As a writer, I do not fear ChatGPT (it has no voice and must mimic), because it can never write in the same idiosyncratic way I do, based on my unique consciousness and life experiences—though I will happily use ChatGPT to make my writing better. Great musicians should not fear AI music generators for the same reason. Musicians can and will use AI to create better music, but no matter how good AI gets in the future, the human touch will always remain an essential part of the process.
T.Z. Barry 