Deezer, the French music streaming giant, is aware that generative AIs may pose an existential problem for artists and musicians. The company actively fights against their possible proliferation.
Deezer is a pioneer and “opens the way to artificial intelligence detection to protect the future of music streaming”. According to the executives, “In a world where AI-powered music is rapidly taking off, Deezer is reinforcing its commitment to help artists better monetize their musicfight fraud and create a better user experience for fans.”
Streaming platforms host hundreds of thousands of new tracks every day, and the advent of generative AIs will only inflate those numbers. According to Jeronimo, CEO of Deezer, it is therefore “increasingly important to prioritize quality over quantity and championing true artists who create high-value content […] Deezer’s goal is toeliminate illegal and fraudulent contentincrease transparency and develop a new remuneration system”.
Deezer designs tools to detect songs created by an AI
Far from being a detractor of artificial intelligence, he specifies: “the use of Generative AI has massive benefits […] We have an opportunity to get it right early on in the AI revolution and not make the same mistakes as the social media giants who let “fake news” flood their platforms. We owe it to the artists and the fans.”
To do this, Deezer is creating a set of tools to fight against content created by an AI. Ironically enough, it is still thanks to AI that Deezer will detect music created by… AI. Content identification will primarily be based on proprietary “Radar” technology. The latter can identify any song even when the signal is distorted or its tempo is changed. If a track is “flagged” as having been created artificially, specific metadata are applied to it.