YouTube is in talks with record labels to license their songs for artificial intelligence tools. These programs that clone the music of popular artists have produced remarkably known songs by imitating the voices of artists like Bad Bunny, for example. The Google-owned video platform needs the record labels' content to legally train AI song generators.

The company has recently offered large sums of cash to major record labels—Sony, Warner, and Universal—to try to convince more artists to allow their music to be used in AI training software, according to several people briefed on the discussions. However, many artists continue to fiercely oppose AI-generated music, fearing it could undermine the value of their work. Any move by a record label to force its stars to participate in such a scheme would be hugely controversial.

 

Challenges and Expectations

"The industry is struggling with this. Technically, the companies have the copyrights, but we have to think about how to handle it," said an executive at a major music company. "We don’t want to be seen as Luddites." Last year, YouTube began testing a generative AI tool that allows people to create short music clips by inputting a text prompt. The product, initially called "Dream Track," was designed to mimic the sound and lyrics of well-known singers.

But only 10 artists agreed to participate in the trial phase, including Charli XCX, Troye Sivan, and John Legend, and Dream Track was made available to a small group of creators. YouTube wants to sign "dozens" of artists to launch a new AI song generator this year, said two of the people.

YouTube is seeking new deals at a time when AI companies like OpenAI are striking licensing agreements with media groups to train large language models, the systems that power AI products like the ChatGPT chatbot. Some of those deals are worth tens of millions of dollars, according to insiders.

The music deals being negotiated would be different. They wouldn’t be blanket licenses but would apply to a select group of artists, according to people familiar with the discussions. It would be up to the record labels to encourage their artists to participate in the new projects. That means the final amounts YouTube might be willing to pay the labels are indeterminate at this stage.

The deals would resemble the one-off payments made by social media companies like Meta or Snap to entertainment groups for access to their music, rather than the royalty-based deals that record labels have with Spotify or Apple, these people said.

 

Innovation in AI and Music

YouTube's new AI tool, which is unlikely to carry the Dream Track brand, could become part of YouTube's Shorts platform, which competes with TikTok. Talks are ongoing, and the terms of the deal could still change.

YouTube's latest move comes as major record labels filed lawsuits on Monday against two new AI companies, Suno and Udio, alleging they are illegally using copyrighted recordings to train their AI models. A music industry group is seeking "up to $150,000 per infringed work," according to the documents. After facing the threat of extinction following the rise of Napster in the 2000s, music companies are trying to get ahead of disruptive technology this time. Record labels are willing to engage with licensed products that use AI to create songs using their  musical copyrights and that pay them for it.

 

Artists and Copyrights

Sony Music, which did not participate in the first phase of YouTube's AI experiment, is in negotiations with the tech group to make some of its music available to the new tools, said a person familiar with the matter. Warner and Universal, whose artists participated in the trial phase, are also in talks with YouTube about expanding the product, these people said.

In April, more than 200 musicians, including Billie Eilish and the heirs of Frank Sinatra, signed an open letter. "If unchecked, AI will start a race to the bottom that will degrade the value of our work and prevent us from being fairly compensated for it," the letter states. YouTube added: "We are always testing new ideas and learning from our experiments; it is an important part of our innovation process. We will continue down this path with AI and music as we build for the future."