Imogen Heap Pioneers Ethical AI Music Generation with Jen Platform

Imogen Heap Just Went Viral With 'Headlock' -- Now She's Letting You Borrow Its Vibe Via AI

Imogen Heap, whose 2005 track "Headlock" just became her biggest hit ever after a belated viral surge on TikTok, is now letting fans legally borrow its vibe for their own songs via generative AI. Heap is the first artist to sign on for a new feature from AI-music platform Jen that lets users "infuse the vibe, feel, rhythmic style, instrument textures" of licensed songs into new creations, according to co-founder and CEO Shara Senderoff. Heap also licensed four other tracks for use as what the company calls StyleFilters: "Goodnight and Go," "Just for Now," "Last Night of an Empire," and "What Have You Done to Me. The leading AI music platforms were trained on vast libraries of copyrighted songs without permission, but Jen only uses fully licensed music, which is why Heap agreed to partner with them. "The really exciting thing about Jen is it's the first service I feel fully understands the importance of waiting to get it right," she says. "They've gone the really long way around." The service pays artists 70 percent of the revenue each time someone uses their StyleFilter. Users pay $4.99 to generate up to 90 minutes of music laced with the essence of Heap's songs -- or $7.99 for a "high-strength" setting where song generations will be more obviously influenced by her work. In a press release, Jen's other co-founder, Mike Caren, frames the feature as a new business model for artists: "Think of StyleFilters like an app for music creators. You're not licensing a song, so much as you're launching a product. You become software." Heap is taking advantage of a bit of fortuitous timing: Just as "Headlock" took off on TikTok, the rights to some of the singer's most popular works reverted back to her after 20 years. "It is like this perfect storm," she says. The deal works through Heap's own technology platform, Auracles, which she's been developing for a decade. Auracles serves as a rights management system, allowing her to quickly provide Jen with all the metadata and permissions for her tracks, and she hopes other artists will see its advantages. With the notable exception of Timbaland, who is a wildly enthusiastic user of the AI-music engine Suno, vanishingly few artists, songwriters, or producers have publicly embraced AI -- instead, they've largely condemned companies who use their music without permission. "It's not that I'm just like, 'Yay, AI,'" Heap says. "I'm 'Yay, AI' with people who have ethics and morals that I align with and a future outlook of something that I feel makes sense to me and the planet.... I want to build a system that enables as much collaboration, open collaboration, as possible with trust." For Heap, traditional music industry models are obsolete. She hasn't released an album since 2014, and has no plans to return to that medium: "Never. A hundred percent. No. I just think that whole industry is baked around 'Let's make money for record labels." Her recent experiment selling song stems on a remix platform yielded $12,000 from 400 downloads at $30 each - far outpacing streaming revenue. "The same song on Spotify, maybe half a million plays, less than a thousand pounds," she says. Heap thinks more artists should find ways to embrace licensed AI. "The more open you are, the more things are likely to happen," she says. "The more [artists] that stick their head in the sand,, the more likely it's gonna go to the people who are doing it for profit and won't be thinking about them."

Imogen Heap Launches AI Music Tools With AI Platform Jen

Carlos Santana Postpones Concert After Suffering From Dehydration Grammy-winning artist and producer Imogen Heap has officially released a set of AI song generation tools through the AI music platform Jen, Heap and the company announced on Thursday, months after originally announcing their partnership late last year. Heap dropped five different "stylefilters" on Jen on Thursday, taking her songs "Headlock," "Goodnight and Go," "Just For Now," Last Night of an Empire" and "What Have You Done to Me," and making them available for users to serve as the basis for new AI song generations. The stylefilters work by taking text-based prompts users feed into Jen's system, describing the type of song they want to create. Jen then takes that description and fuses it with whichever Heap song the user picked, creating a new track with the vibe of the selected stylefilter song. Each of Heap's stylefilters are available for purchase for $4.99 ($7.99 for a higher strength version), allowing users up to 90 minutes of creations that they're free to use as they please (except to train AI models). The company said they'll be adding stylefilters for more artists in the coming months but didn't specify who would participate or when to expect their launches. Heap, who's long been an early adopter of technology to push the boundaries to create her art, tells The Hollywood Reporter she partnered with Jen as she embraced that the technology is here to stay. Heap says she wants to take a more proactive role in deciding how AI engages with her art rather than letting it be decided for her. "If you're just looking to ignore AI, well sorry, it's already happening, and this is not going away," she says. "We need to find a way to live together [with the technology], and the only way to do that is to do what we do for ourselves. Artists have to get involved. We can't let only the people who want to do it for profit or want to do it for their own gain be making the decisions." Jen first announced the partnership during the Web Summit in Portugal back in November. Jen was cofounded by music-tech executive Shara Senderoff and Mike Caren, the founder of record label and publishing company Artist Partner Group. Every stylefilter is trained only on its original song, according to Jen. Jen positions itself as an ethical means of AI music development, noting that the company only trains its software on music it's explicitly obtained licenses to use. This is a departure from other AI companies, who've often trained their models without permission from rights holders while arguing AI generation is a fair use case. In the music industry, the major record labels sued two of the most prominent AI music generators, Suno and Udio, accusing the companies of powering their AI on thousands of unlicensed songs. Senderoff says the upfront payments to artists for their filters offers a potentially faster means of getting paid and says that with stringent requirements to get licenses for their products, Jen has an opportunity to establish a better data and tracking system than what the music industry currently has available. "I think I'm building in a blue ocean," Senderoff tells THR. "I don't think I'm doing what the others are doing. This is an opportunity for music creation to redefine monetization, and that is a big difference because I licensed and did the work. We're not going to risk the original creators not being compensated because we are building this with utmost integrity and trying to create a precedent for how this should be done." Right now, Jen only produces instrumental music, but Senderoff says Jen will eventually launch a vocal product as well. Heap says she partnered specifically with Jen and Senderoff because "she was the only one doing this ethically, and I did look quite hard." "I think a lot of people want to plant their flag in the sand before they've got all this figured out," Heap says. "And maybe that works in the short term, but I don't think it'll work in the longer term because f you're not financially recompensing the people that create music, you're going to end up with a desert." When asked where she draws the line with AI creations, Heap says it's mainly about proper credit. "if my music has an essence of something that is being released into the world, it's just about recognition," she says. "The only thing I have on this planet right now, this is my output. This is my extension of myself. And I would like that still to be possible. It's beyond just financials and getting a few extra quid. It's recognition and attribution."