Only Humans Can Make Music: The Current State of AI in the Music Industry

Most people don’t have any issue with the use of artificial intelligence to read, write or do math. Yet, when it’s used for the creation of music, people react very differently. In my experience, the prospect of AI making music elicits much more discomfort than even other art forms such as drawings, videos, or essays. But why is it that we, the people who let AI overrun writing, social media and school, defend human made music so staunchly? What makes music so uniquely and wholly human?

In a 2025 interview, Radiohead frontman, Thom Yorke, likened AI generated music to “a screensaver of a beautiful natural landscape in a billionaire’s bunker.” In other words, as the moving photos of trees and rivers are only a recreation of the real landscape, AI generated music is a mere imitation of real human emotion. This is true because of the way artificial intelligence actually works. It’s trained on preexisting information and is physically incapable of creating anything that is truly original. Whether it’s stealing from one song, 10 songs, or 100,000 songs, it’s still stealing. While AI can solve math problems and answer complex questions, the idea that it can also create music is factually incorrect.

This is why it brings me great dismay that Spotify has just enacted a deal with music industry titans, Universal Music Group, Warner Music Group, and Sony Music Group valued at over $4.5 billion. These three labels own roughly 75% of the global market. To put that into perspective, of the 4.8 trillion songs streamed in 2024, about 3.6 trillion belonged to UMG, WMG, or SMG. And now, all three of these companies are involved in the furthering of AI’s involvement in the creation of music. Spotify claims that this deal aims to regulate the use of AI and keep it “responsible”. However, after a recent controversy surrounding the use of fully AI generated tracks to pad out Spotify-made playlists, it’s hard to believe that this is their true intention. They also cite “creating new revenue streams” as a reason for the deal, which only strengthens their reputation for prioritizing income over artists. This deal appears to mark the beginning of an AI centric era of the music industry.

Spotify is not the only entity betting on artificial intelligence. A few weeks ago, R&B singer Xania Monet signed a $3 million contract with Hallwood Media. And it’s easy to see why they would want Monet on their label: her first single, “How Was I Supposed to Know?” climbed to the 22nd spot on Billboard’s Digital Song Sales chart, and her entire catalogue has grossed an estimated $52,000 to date. But there’s a catch: Xania Monet is not real. She is an AI bot created by Mississippi woman, Talisha Jones, to turn her poems into music. That’s right, a music chart that is currently dominated by the likes of Taylor Swift and Morgan Wallen, is also occupied by a fake person. She–rather, it–is a derivative of AI music service, Suno, which allows you to create a song simply by describing it. Numerous artists have spoken out against Monet’s signing, including Grammy-nominated R&B singer, Kehlani, who said: “Nothing and no one on earth will ever be able to justify AI to me, especially not […] AI in the creative arts in which people have worked hard for.”

It is the artists themselves who offer a glimmer of hope in this story. While label executives continue to push the usage of artificial intelligence, artists are progressing in the opposite direction. Ditto Music discovered that use of AI in the creative process is down almost 12% from 2023, and use of AI for songwriting specifically is down 15%. Furthermore, Ditto discovered an almost 20% rise in concern over AI’s “lack of individual creativity”. Artists’ support of AI is dwindling, and it seems that there are enough people against it to combat its growth in the industry.

Only humans can make music, because only we can understand its true meaning. Our emotions, imperfections, and individuality are what bring songs to life. If we want to keep music alive, it is imperative that we keep AI far away from the industry. There are a few ways you can help with this. First, and most simply, you can avoid listening to AI music. The best ways to do this are to steer clear of Apple Music or Spotify branded playlists. Additionally, Ircam Amplify has created a music-oriented artificial intelligence detector. Another more involved way of supporting the cause is backing the NO FAKES Act, a bipartisan bill that includes legislation on the use of AI in music. While the ship has sailed for many creative mediums, AI is still in the early phases of integration in music. And while artists roll back on its usage, we must hope that labels and streaming platforms will follow suit.

Leave a comment

What to read next