SoundCloud updated its Terms of Service in February 2024 to include language stating that users agree their content may be used to 'inform, train, develop or serve as input to artificial intelligence' as part of providing the platform's services. Following public backlash in May 2025, SoundCloud's CEO issued a letter clarifying that no user content had been used to train generative AI models, and the company revised its terms to explicitly prohibit training generative AI on user content without opt-in consent.

The February 2024 Terms update was discovered publicly in May 2025 by tech ethicist Ed Newton-Rex, triggering widespread concern among artists. SoundCloud stated the clause was intended to cover internal AI uses such as personalized recommendations, content organization, fraud detection, and content identification — not generative AI model training. In a May 14, 2025 CEO letter, SoundCloud revised its Terms to state: '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.' Any future use of content for generative AI model training would require an opt-in mechanism.