Spotify Is Finally Trying to Combat AI Inefficiencies.
Did you know you can configure Google to filter out junk? Follow these steps to improve your search results, including adding my work on Lifehacker as a preferred source .
This morning, Spotify announced in a blog post measures to combat the overabundance of AI-generated music on its streaming platform. According to the company, action will be taken against unscrupulous artists and content farms that “push garbage into the ecosystem.” Spotify reports that it has already removed over 75 million such tracks over the past year, and more significant changes are on the way.
Spotify says it will crack down on music copycats over the next few months, roll out a new spam filtering system, and collaborate with others in the music business to develop an industry standard for AI-powered artist disclosure.
How Spotify’s AI-powered music policy will work
According to Spotify, the availability of AI tools has made it easy to create music deepfakes—AI-powered imitations of existing artists’ voices. The company says it will remove tracks that “imitate another artist’s voice without their permission, whether through AI-powered voice cloning or any other method.”
The ban applies to both tracks uploaded by a user who clearly impersonates another artist, and tracks labeled as an “AI version” of another artist—unless, of course, the track was created with the original artist’s permission.
Spotify is also combating bulk uploads, duplicates, SEO hacks, artificial track shortening, and other spam attacks on its platform. The company’s new spam filter, launching this fall, will identify users and tracks using such tactics and then “flag them and stop recommending them.” According to Spotify, the ultimate goal is to prevent unscrupulous users from receiving royalties that would otherwise go to professional artists and songwriters.
Spotify is working on an AI-powered music labeling standard.
Spotify also pledged to help develop a comprehensive industry standard for disclosing the use of artificial intelligence in music creation. Labeling AI in music credits is a much more complex task than Spotify’s other new initiatives: music creation uses a wide variety of technologies, and there’s a huge difference between a track that’s entirely generated from suggestions and using autotune to create slightly off-key vocals.
Spotify says the initiative requires “broad industry coordination,” so it’s partnering with companies like Labelcamp, NueMeta, Revelator, and SonoSuite through music industry standards firm DDEX to develop an industry standard for AI-powered labeling.
But much remains to be done.
Spotify’s new initiatives don’t ban AI-generated music or require its labeling. The company states that it wants to treat all music “equally, regardless of the tools used to create it,” which presumably allows Spotify to continue promoting playlists clearly created by AI, such as ” Jazz for Study ” and ” Lo Fi Chill ,” which consist primarily of “artists” like The Midtown Players , Ourchase , and The Tate Jackson Trio , which bear all the hallmarks of AI creations but are officially “verified” by Spotify.
To be fair, I did a similar search for AI-generated playlists and artists a few months ago. Finding them is a bit harder now than it used to be. However, until Spotify stops filling its playlists with AI-generated garbage, its promise to combat the “AI scum” rings hollow.