SoundCloud addresses company’s AI policy in new statement


Panellists at the Digital Life Design women conference at Bavarian National Museum. June 30, 2011. Photo by Sascha Baumann for Getty Images


 

SoundCloud has clarified its terms of use after a social media uproar sparked by a clause in the streaming platform’s terms of use concerning AI. The offending portion, added in February 2024, gave SoundCloud permission to use uploaded content “to inform, train, develop or serve as input for artificial intelligence or machine intelligence technologies or services as part of and for providing the services.”

In their initial response to the controversy, SoundCloud said that the platform has “never used artist content to train AI models” and would not do so in the future without the expressed permission of the content’s owners. “We understand the concerns raised and remain committed to open dialogue,” the statement read. Perhaps spooked by the Duolingo exodus following the company’s “AI first” vow, SoundCloud CEO Eliah Seton has issued a new statement.

Addressed to SoundCloud’s “Artist Communit,” Seton’s statement acknowledges the confusion caused by the updated contract (“That’s on us,” he writes) and reiterates the claim that that the platform has not used uploaded content “for anything that tries to mimic or replace our work.”

Seton did leave the door open for SoundCloud partnering with generative AI models in the future, but claimed that they would have to be approved on an opt-in basis by the uploader.

“We’re revising the Terms of Use to make it absolutely clear: SoundCloud will not use your content to train generative AI models that aim to replicate or synthesize your voice, music, or likeness.”

The updated clause now reads: “We 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.”

Read the full statement below.

To Our Artist Community,

Over the past few days, there’s been a lot of conversation about our 2024 Terms of Use update which was intended to clarify how content may interact with AI technologies within SoundCloud’s own platform.

I want to take a moment to speak directly to you—our artist community—about what’s true, what’s not, and what we’re doing next.

First, let’s be clear.

SoundCloud has never 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. Period. We don’t build generative AI tools, and we don’t allow third parties to scrape or use artist content from SoundCloud to train them either.

In fact, we’ve already put protections in place like a “no AI” tag that explicitly signals content on SoundCloud can’t 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.

So, what happened?

Back in February 2024, we updated our Terms of Use to clarify how we may use AI internally to improve the platform for both artists and fans. This includes powering smarter recommendations, search, playlisting, content tagging, and tools that help prevent fraud.

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. Three years ago we expanded our AI and machine learning capabilities through the acquisition of Musiio which significantly improved how we connect creators with listeners, fuel music discovery and help rightsholders identify what’s next. Since then, through First Fans, our recommendation algorithm for Artist Pro subscribers, we’ve delivered over 7 million track recommendations to potential new listeners, helping artists get heard faster and get discovered.

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.

AI has been, and will continue to be, a key part of how we improve SoundCloud for the people who power it.

But here’s the thing, the language in the Terms of Use was too broad and wasn’t clear enough. It created confusion, and that’s on us. That’s why we’re fixing it.

What we’re doing now:

We’re revising the Terms of Use to make it absolutely clear: SoundCloud will not use your content to train generative AI models that aim to replicate or synthesize your voice, music, or likeness (see detailed terms below).

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. We don’t know what we don’t know, and we have a responsibility to give our human artists the opportunities, choices and control to advance their creative journeys.

We’re making a formal commitment that any use of AI on SoundCloud will be based on consent, transparency, and artist control.

Our position is simple: AI should support artists, not replace them. Any use of these tools on SoundCloud will continue to reflect that. You’ll see these changes reflected online within our Terms of Use in the coming weeks which can be found here.

AI is going to be a part of the changing landscape of music. It brings new opportunities, but also very real challenges. That’s why our approach will always be guided by a single principle: artist-first.

We’re going to keep showing up with transparency. We’re going to keep listening. And we’re going to make sure you’re informed and involved every step of the way. Thanks for being a part of the SoundCloud community and for holding us accountable to the values we all share.

Eliah Seton

CEO, SoundCloud