Spotify has strengthened its policy against generative artificial intelligence (AI) in music through a new spam filtering system and stricter anti-phishing rules, the streaming platform announced Thursday in a statement.
To enhance protection in cases of impersonation, vocal impersonation will only be permitted on music available on Spotify when the impersonated artist has expressly authorized its use.
Spotify ‘shields’ itself against artificial intelligence
In addition, it will allocate more money to combat cases where content is fraudulently uploaded through another artist’s profile onstreaming services, whether it is AI-generated music or not.
And to improve the process of reviewing incorrect content so that artists can report potential mismatches even before the official release of the albums.
A new music ‘spam’ filter will also be implemented, a system that will identify uploaders and tracks that use these tactics, as well as tag them and stop recommending them.
Spotify is also involved in the development of a new industry standard for statements of AI usage in music credits.
As this information is submitted by labels, distributors and music partners, it will begin to appear on the platform‘s application.
In this way, artists and rights holders will have a clear way to indicate “where and how AI was involved in the creation of a track, whether in AI-generated vocals, instrumentation or post-production.”
This isn't about punishing artists who use AI responsibly, nor will it affect how content is prioritized or promoted on Spotify
Spotify
The goal is to “fight impersonation, ‘spam’ and cheating,” which have grown at the same rate as total payments for music on Spotify, which have grown from $1 billion in 2014 to $10 billion by 2024.
High payments that “attract malicious actors”.
In the view of Spotify, the music industry “needs a nuanced approach to transparency in the use of AI, without the obligation to classify every song as ‘is AI’ or ‘is not AI’.”
With information from EFE
