Commentary: Is SoundCloud Now Allowing AI Training on Artists’ Music?

On May 9th, 2025, after first being spotted by Ed Newton-Rex, the media outlet TechCrunch reported that the streaming platform SoundCloud had quietly made a significant change to its terms and conditions. According to the ‘new’ policy (which was reportedly added to the terms back in February 2024), SoundCloud reserves the right to train AI on audio that artists create and upload to the platform.

“You explicitly agree that your Content may be used to inform, train, develop, or serve as input to artificial intelligence or machine intelligence technologies or services as part of and for providing the services,” read the Soundcloud terms and conditions.

The policy includes a carve-out for content under “separate agreements” made with third-party rights holders, such as record labels. This content cannot be used, copied, or reproduced by SoundCloud or any third party. According to Newton-Rex, no explicit opt-out options were found on the company’s website at the time of reporting.

As one can imagine, the recent revelation alleging that creators may unknowingly consent to SoundCloud utilizing their art to train and develop AI technologies has drawn intense criticism from industry insiders, including artists and AI experts like the previously mentioned Newton-Rex.

The founder of Fairly Trained, a nonprofit organization that certifies generative AI companies for their training data practices respecting rightholders’ rights, has penned an op-ed in Music Business Worldwide to share his perspective on the SoundCloud situation.

“Unfortunately, even for a platform that calls itself artist-first like SoundCloud, it looks like the allure of a trove of training data is just too great to be turned down in the age of generative AI,” expresses Newton-Rex in his article.

A quick visit to a dedicated Reddit page also reveals the strong reactions from artists, creators, and fans alike.

“That’s the frightening thing about the future of music rn,” comments one user. “There’s no way I would have PAID money for this had I known,” remarks another. Some have gone as far as to say, “We need a good alternative to Soundcloud...,” or “Honestly, I wouldn’t be surprised if a mainstream artist was revealed as being completely AI… not shocked at all.”

SoundCloud’s affinity for AI, however, shouldn't come as a surprise to anyone, as the company has been embracing new developments in recent years. In 2024, SoundCloud collaborated with several third-party services to offer a total of six new AI-powered tools for generating vocals, remixing, editing, producing, and creating custom samples.

Back then, the company highlighted that it was further partnering with Audible Magic and Pex in order to remain responsible, innovative, and ethical regarding the use of creative AI tools. At that point, it stated that it will “provide all of our existing and new AI partners access to content identification solutions, ensuring rights holders receive proper credit and compensation.”

SoundCloud took five days to address the public’s understanding of the articulation of the company’s new policies, which have sparked backlash. In the official statement, SoundCloud CEO Eliah Seton categorically denied that the company had ever “used artist content to train AI models. Not for music creation. Not for large language models. Not for anything that tries to mimic or replace your work.”

“Our use of AI is focused on discovery—helping fans find new music and helping artists grow, starting with their first fans. That’s core to our mission…More broadly, we use AI to identify emerging talent, personalize the platform experience, and support real-time customer service, all designed to support human artists and engage real fans,” Seton noted in his statement.

He went on to say that the company has diligently safeguarded artists from generative AI by implementing technical protections like the ‘no AI’ tag, which explicitly signals that no content published on SoundCloud can be used for AI training. “At SoundCloud, protecting artist rights isn’t new for us, and being artist-first isn’t a slogan. It’s core to who we are and always will be. It’s in our DNA,” he emphasized.

What Seton recognized as true is that the wording of the new policy in the T&Cs was framed unclearly and ambiguously. As a result, he stated that the company would revise the section on AI use to read: “[SoundCloud] will not use Your Content to train generative AI models that aim to replicate or synthesize your voice, music, or likeness without your explicit consent, which must be affirmatively provided through an opt-in mechanism.”

It is specifically this aspect that many industry professionals found unconvincing about SoundCloud’s future engagement with generative AI tools. Newton-Rex noted that while the company denied any previous use of artists’ music for training generative AI, it also did not state that they weren’t planning to do so in the future.

“...if you weren’t planning on training on your users’ music, you would sure as hell say so. In other words, they have not ruled out using their users’ music to train generative AI models in the future—and their terms of service explicitly allow it,” Newton-Rex emphasized.

However, what startles Newton-Rex and other industry insiders isn’t merely the possibility that SoundCloud might one day decide to adopt generative AI. After all, Seton hasn’t been secretive about a potential partnership between SoundCloud and an AI generative tool. It would be unfair, then, to accuse the streaming platform of a lack of transparency. “With the landscape changing rapidly, if there is an opportunity to use generative AI for the benefit of our human artists, we may make this opportunity available to our human artists with their explicit consent, via an opt-in mechanism,” Seton emphasized.

It’s the ‘opt-out mechanism’ aspect that faces widespread criticism, though. “Opt-out mechanisms for generative AI training are hugely unfair to musicians, for a heap of reasons. A streaming platform should not be exploiting its users’ music to train generative AI models without their explicit permission. This is non-negotiable,” Newton-Rex stated.

He also criticizes the lack of communication from SoundCloud, emphasizing the necessity (both practical and legal) to inform users about significant policy changes. “If you’re going to start reserving the right to use people’s music for something totally orthogonal to what they signed up to the service for – particularly something as inflammatory as AI training – you have to tell them,” he remarked.

Finally, Newton-Rex highlights a key difference in how indie and signed artists are treated, pointing out that, according to the T&Cs, musicians without labels face significantly worse terms than those backed by a label. This is a serious claim against a company that has consistently positioned itself as artist-first, as seen in its statement about the new policy. However, it is true that in the Terms of Use, only licensed content from major labels is excluded from being used to train AI models, including generative AI. For other content, any AI-related use is permitted. The carve-out section remains in the T&Cs even after the updated wording of the policy.

“Why on earth would a platform that is supposed to pride itself on existing to serve all musicians…treat unsigned artists so poorly, and subject them, and only them, to involuntary AI training on their music?” asks Newton-Rex in disbelief.