We missed this earlier: Music streaming service Deezer revealed that roughly 50,000 fully AI-generated tracks are being uploaded every day on its platform, accounting for almost 34% of total daily deliveries.
Notably, this revelation came as part of a wider survey aimed at understanding perceptions and attitudes towards AI-generated music. For context, Deezer and market research firm Ipsos conducted this survey across eight countries, namely the United States (US), Canada, Brazil, the UK, France, the Netherlands, Germany and Japan.
The survey’s results pointed clearly towards the desire to tag 100% AI-generated music and make sure that artists and songwriters receive fair treatment and remuneration if their music is used to train AI models.
Notably, this survey came out close on the heels of a US-based rapper legally pulling up music streaming giant Spotify over allegedly ignoring the activities of people who artificially inflate their streams with bots.
The Deezer AI Music Survey At A Glance
Initially, all the 9000 survey participants listened to three tracks and then determined whether or not they were fully AI-generated. Interestingly, 97% of the respondents failed at this task. Meanwhile, 80% of participants agreed that fully AI-generated music should be clearly labelled for listeners.
Notably, Deezer claims that even though fully AI-generated music currently accounts for only a small fraction of streams on its platform—around 0.5%—it is evident that the primary purpose of uploading such tracks on streaming platforms is fraudulent. Furthermore, the France-based music streaming service reveals that it excludes AI streams from the ambit of royalty payments whenever it detects stream manipulation of any kind whatsoever.
Pertinently, Deezer is the only music streaming platform to label 100% AI-generated music clearly for its listeners. And the music service claims that its AI music detection tool can detect fully AI-generated music even from the most prolific generative AI models, such as Suno and Udio.
Furthermore, Deezer says that it has made significant progress in order to detect AI-generated content without a specific dataset to train on.
AI-Created Music’s State of Play Across the Industry: Spotify In Focus
While there is no official statistic available with respect to the total number of AI-generated songs available on the Spotify platform as of now, the Stockholm-based music platform does claim that it has removed over 75 million ‘spammy’ tracks in the 12 months leading up to September 2025.
Additionally, in the same month, Spotify announced the rollout of enhanced and more robust AI protections for artists, songwriters, and music producers. For context, these include:
- Stronger impersonation rules
- A music spam filter
- AI disclosures for music with industry-standard credits
However, in October 2025, Spotify collaborated with the likes of Sony Music and Universal Music, among others, to build “artist-first” AI products.
Importantly, this highlights Spotify’s multi-faceted approach to dealing with AI – from policing its very misuse to developing licensed AI tools that prioritise consent, credit, and fair compensation for artists
How Does AI-Generated Music Hurt the Industry and Artists?
To actually understand how AI-generated music truly wreaks havoc for the music industry worldwide and the artists who are part of it, one must examine a real-life incident that occurred last year.
In September 2024, a musician in the US was accused of using AI tools and thousands of bots to fraudulently stream songs billions of times.
Importantly, these artificial streams resulted in this person claiming around $10 million in royalty payments over several years. Meanwhile, prosecutors said that this is “the first criminal case of its kind that they have handled”.
“Through his brazen fraud scheme, Smith (the accused) stole millions in royalties that should have been paid to musicians, songwriters, and other rights holders whose songs were legitimately streamed,” a US attorney told the BBC.
Also Read:
Support our journalism:
For You
Source link
