Spotify’s New AI Policy: Labeling AI Music & Fighting Spam
▼ Summary
– Spotify is adopting the DDEX industry standard to require labels and distributors to provide detailed, standardized disclosures about AI usage in music credits.
– The company explicitly bans unauthorized AI voice clones, deepfakes, and vocal impersonations, stating such content will be removed from the platform.
– A new music spam filter will be launched to combat AI-enabled spam tactics, such as mass uploads and SEO manipulation, by tagging and stopping the recommendation of these tracks.
– Spotify will work with distributors to prevent “profile mismatches,” a fraudulent scheme where music is uploaded to another artist’s profile.
– Executives emphasized that the policy aims to stop bad actors, not punish artists who use AI responsibly and authentically in their creative workflow.
Spotify has unveiled significant changes to its approach on artificial intelligence in music, focusing on greater transparency for listeners and a stronger stance against platform manipulation. The updates introduce a system for labeling AI-generated content and new tools to combat the rising tide of spam facilitated by AI technologies. This move aims to clarify the platform’s rules, particularly around unauthorized voice cloning, which is now explicitly prohibited.
A central component of the new policy is the adoption of an emerging industry standard, DDEX, for identifying AI involvement in music production. Labels, distributors, and other partners will use this framework to submit detailed disclosures directly within song credits. This system is designed to provide nuanced information, specifying if AI was used for vocals, instrumentation, or post-production work. Sam Duboff, Spotify’s Global Head of Marketing and Policy, explained that this avoids a simplistic “AI or not” classification, acknowledging that AI use exists on a spectrum within creative workflows.
Alongside the labeling initiative, Spotify is preparing to launch a new music spam filter later this fall. This tool is intended to identify and tag spammy uploads, preventing them from being recommended to users. The company recognizes that AI has made it easier for bad actors to engage in mass uploads, create duplicate content, and manipulate search algorithms. The filter will be rolled out gradually, with its capabilities refined over time based on evolving spam tactics.
The platform is also tackling “profile mismatches,” a fraudulent practice where music is uploaded to another artist’s profile without permission. Spotify plans to collaborate with distributors to address these incidents more effectively, ideally stopping them before the content becomes publicly available.
Despite these protective measures, Spotify executives were clear that they support legitimate and responsible uses of AI. Charlie Hellman, Spotify VP and Global Head of Music, emphasized that the goal is not to penalize artists using AI creatively but to aggressively stop those who exploit the system. The company believes that AI tools can empower artists, but that a safe and transparent ecosystem must be maintained.
These policy shifts arrive as AI-generated music becomes more prevalent across the industry. A recent estimate from rival service Deezer suggested that approximately 18% of daily uploads, or over 20,000 tracks, are fully AI-generated. While Spotify did not share its own specific metrics, Duboff noted that music catalogs across streaming services are largely identical, as content is typically delivered to all platforms simultaneously. He also pointed out that an upload does not equate to listenership or revenue, reinforcing the idea that the focus is on quality and authenticity, not merely the presence of AI.
(Source: TechCrunch)
