Artificial Intelligence (AI) is the talk of the town everywhere. With AI tools displacing jobs that only humans could do before, it’s natural to wonder if machines can copy the most human activity of all – art.
There have already been numerous examples of AI seeping into the music industry. Of course, computers have been part of the musical process for decades now. In 1957, Lejaren Hiller and Leonard Isaacson composed the Illiac Suite with an ILLIAC computer, predicated on rules from music theory. This would mark one of the most important developments in computerizing music.
Years later, musicians like David Bowie used simple software to aid their lyric writing process. In the late 1990s, Bowie used the Verbasizer, a lyric-shuffling software that helped him generate words for his songs. It wasn’t very sophisticated, but it was an early example of man collaborating with artificial intelligence.
Fast forward to today, and it’s estimated that 60% of recording artists already use AI technology in creating their work. The market for generative AI in music is expected to reach $2.6 billion by 2032. While some artists embrace the change and look to use AI in whatever way they can, others are wary and want to stay far away from AI and its implications for their creative work.
Nowadays, AI is showing up in different ways around the music world. Francois Pachet created the first pop album composed by AI in 2018 called Hello, World. Singer-songwriter Holly Herndon made an album called Proto, where she harmonizes with herself – her AI generated self.
Musician Ash Koosha developed an AI pop star named Yona that has performed on stage and writes its own lyrics. While many are nonsense, there are a few vulnerable lines Yona sings that shock even its creator
These are just some of the more experimental examples. Many artists already use AI to help them with more mundane tasks, like generating ideas and producing their work.
In 2023, a song called “Heart on My Sleeve” featuring both Drake and the Weeknd went viral on TikTok and numerous streaming platforms. Shortly after, it got taken down at the request of their record labels. Why?
Neither Drake nor the Weeknd had anything to do with it.
This song was made entirely by AI. According to the “Ghostwriter” user who had posted the song, they had fed a generative AI tool Drake’s work without express permission so it could learn his style and then replicate it with his voice. Needless to say, this created a lot of problems. Whether or not the song actually was created using AI, it woke up the world to the AI surge in music – and the numerous copyright problems that come with it.
.jpg)
Is the song an original work? Did they violate Drake or the Weeknd’s rights of publicity, which protect them from unauthorized uses of their image, likeness, or voice? Who controls the rights to AI generated songs? These thorny questions show the problems for modern artists in a world being quickly conquered by AI.
While the artists themselves didn’t speak out extensively about the incident, Drake made it clear through social media he isn’t pleased about all the AI generated songs with his voice. When someone posted his AI cover of the song “Munch” by Ice Spice, he wrote on his Instagram story that it was “the last straw.”
Other musicians have strong opinions about this rapid growth of AI in music, which we discuss below.
Earlier in the year, the ARA sent out an open letter, signed by over two hundred artists (from Billie Eilish to Nicki Manaj to the Imagine Dragons), that was critical of AI in music. Their letter says that “We must protect against the predatory use of AI to steal professional artists’ voices and likenesses, violate creators’ rights, and destroy the music ecosystem.”
While they acknowledge that AI can advance human creativity, the ARA takes a firm stance against AI being used to devalue their work. They point out in the letter that companies are already training AI models with artists’ work without permission. Their goal is to replace human work with a vast pool of AI generated, royalty-free music, which would be catastrophic for working artists who aren’t making millions of dollars.
One of the most outspoken supporters of AI in music is Canadian electro pop artist Grimes. Last year, she launched an AI tool called Elf.tech, which allows users to record their own voice and then change it into her own.
To encourage her audience to use the program, she said that she’ll split royalties 50-50 with anyone who is able to produce a hit song using her AI tool. While other artists might disagree, Grimes wants to “kill copyright” and open source art so that it’s accessible to anyone.
When he was interviewed by Billboard for the 50th anniversary of Hip Hop cover, Lil Wayne was asked briefly about his thoughts on AI in music. His answer? AI can never truly duplicate a person.
Someone had told the rapper that AI coud make music with a voice that sounded exactly like him, but Lil Wayne said that “it’s not me, because I’m amazing. I’m like, is this AI thing going to be amazing too?”
If you ask Lil Wayne, nothing can duplicate original, organic talent and a healthy dose of hard work.
Mega star Pharrell Williams takes the realist route to AI in music. He hasn’t come out and said that AI in music is bad, but he’s resigned himself to its inevitable rise to center stage.
Williams said the development of AI is “human nature” and that “there’s not much you can do, so buckle up.” His view reflects many who see that AI can’t be restrained forever, and that it’s best to figure out how to make the best of a changing world.
A Los Angeles electro pop band, YACHT saw unprecedented success when they started employing AI in their creative process. Before creating their album Chain Tripping, the band trained an AI machine on all their work to date. Then they told it to do its magic.
Most of the musical results from the machine weren’t helpful, but the band combed through the data and brought together the most intriguing ideas into a sensical song. What came out was a restless album that jumped from idea to interesting idea.
However, YACHT had a difficult time learning (their own?) music. The AI generator pushed them in unexpected ways, taking them out of their comfort zone and altering riffs and chords they were used to playing. It forced them to break familiar habits, learn new skills, and try something different.
The result was the band’s first Grammy nomination for best immersive audio album.
The famous singer shared his thoughts on Artificial Intelligence during a conversation on Audacy Live. Sheeran said he doesn’t know why we need AI.
“If you’re taking a job away from a human being,” he said, “I think that’s probably a bad thing.” Ed Sheeran points out that everyone needs a job in a functioning society, and using AI to replace people means that everyone will be out of work.
Already, some government bodies are taking action on protecting the rights of musicians. In March of this year, Tennessee became the first U.S. state to sign AI protection for artists into law. The Ensuring Likeness Voice and Image Security (ELVIS) act aims to keep musicians from being hurt by unauthorized use of AI.
It can be fairly said that the impact of AI on music is both overstated and wildly underappreciated. While music artificial intelligence is still unable to produce consistently coherent results or even begin to replace humans, it has become an invaluable tool to tens of thousands of artists and is developing at an unprecedented rate.
Because of this rapid growth, it’s difficult to project what AI in music will look like in a few decades. Will the futurists like Grimes prove correct, and AI will become the leading producer of music? Or is Lil Wayne right – you can never reproduce human genius?
Only time will tell.
