SoundCloud says it isn’t using your music to train generative AI tools
The company says it’ll let users opt out if it decides to train AI on their content in the future.
The company says it’ll let users opt out if it decides to train AI on their content in the future.
Image: The Verge / Shutterstock
Wes Davis is a weekend editor who covers the latest in tech and entertainment. He has written news, reviews, and more as a tech journalist since 2020.
The music-sharing platform SoundCloud quietly updated its terms of use in February last year, adding language that lets it train AI models on its users’ content, as TechCrunch reported. And while the company says it hasn’t used user-created content for model training, it doesn’t rule out the possibility that it will in the future.
Marni Greenberg, SVP and head of communications at SoundCloud, provided the following in a statement emailed to The Verge.
SoundCloud has never used artist content to train AI models, nor do we develop AI tools or allow third parties to scrape or use SoundCloud content from our platform for AI training purposes. In fact, we implemented technical safeguards, including a “no AI” tag on our site to explicitly prohibit unauthorized use.
Greenberg went on to say that SoundCloud’s terms of service update “was intended to clarify how content may interact with AI technologies within SoundCloud’s own platform.” She said the company uses AI for things like personalized recommendations and fraud detection, and suggests its plans for future uses of AI on its platform fall along similar lines.
When we asked about letting users opt out of having their music used for generative AI development, here’s what Greenberg had to say:
The TOS explicitly prohibits the use of licensed content, such as music from major labels, for training any AI models, including generative AI. For other types of content uploaded to SoundCloud, the TOS allows for the possibility of AI-related use.
Importantly, no such use has taken place to date, and SoundCloud will introduce robust internal permissioning controls to govern any potential future use. Should we ever consider using user content to train generative AI models, we would introduce clear opt-out mechanisms in advance—at a minimum—and remain committed to transparency with our creator community.
Hopefully SoundCloud will go to greater lengths to tell users about those opt-out mechanisms than it appears to have done for last year’s AI-related terms of use update. Tech ethicist Ed Newton-Rex, who spotted the changes reported by TechCrunch, posted that they “can’t see any emails” alerting them that the terms had been altered. I’ve contributed to SoundCloud, too, and also didn’t find any emails about the changes when I checked. SoundCloud’s terms say it will provide “prominent notice” about significant alterations to its terms, but doesn’t guarantee you’ll see that in an email.
Here’s SoundCloud’s original full statement, as provided to to The Verge by Greenberg:
