Accueil » Streaming Music: AI Floods Platforms and Reignites the Fight Against Stream Fraud

Streaming Music: AI Floods Platforms and Reignites the Fight Against Stream Fraud

par Nathan Reed
0 commentaires
Abstract illustration of streaming music and artificial intelligence

On streaming platforms, AI-generated music keeps gaining ground. According to the latest figures published by Deezer, more than half of the new tracks uploaded each day are now produced by AI, a shift that is forcing the music industry to overhaul its detection tools and its rules for compensation.

A dramatic rise in just a few months

In April 2026, Deezer reported receiving around 75,000 AI-generated tracks per day, or 44% of daily uploads on the platform. Two months later, that volume had jumped to nearly 90,000 tracks per day, crossing the 50% mark of all new uploads for the first time. The French platform, which has made transparency about AI a selling point, was the first to publish this kind of detailed statistics and to market a detection tool for other players in the industry.

Stream fraud out of proportion

The real problem isn't so much the volume as its fraudulent use. According to Deezer, as much as 85% of the streams generated by synthetic tracks came from fraud in 2025, compared with just 8% for the catalog as a whole. Networks of bots or click farms artificially inflate stream counts to capture a share of advertising revenue and royalties, at the expense of real artists. Notably, despite these impressive volumes, AI tracks account for only 1 to 3% of actual listening on the platform, proof that the sheer quantity uploaded has nothing to do with audience interest.

  • April 2026: 75,000 AI tracks/day, 44% of uploads
  • June-July 2026: around 90,000 AI tracks/day, more than 50% of uploads
  • Up to 85% of streams on AI tracks deemed fraudulent (versus 8% on average)
  • Only 1 to 3% of actual streams come from AI-generated tracks

The measures taken by platforms

Faced with this situation, Deezer has put several safeguards in place: systematically excluding tracks identified as AI-generated from recommendations and editorial playlists, removing tracks linked to fraud schemes or entirely ignored by listeners, and, above all, bringing to market a detection tool for labels and distributors. Other streaming services are closely watching these initiatives, though none have yet released statistics as precise about their own catalogs.

What impact for artists and the economy of music?

A joint study by CISAC and the consulting firm PMP Strategy estimates that nearly 25% of creators' revenue could be at risk by 2028 due to the dilution caused by AI-generated content, or roughly 4 billion euros worldwide. That's likely to fuel the already heated debate over fair compensation for artists, just as the French recorded music market has wrapped up a tenth consecutive year of growth. The question of transparency and labeling for AI tracks, still in its infancy, looks set to become one of the industry's major challenges in the coming months, at the crossroads of technology and creativity.

Frequently Asked Questions

What counts as an “AI-generated” track according to Deezer?

This refers to a track whose composition, vocal performance, or instrumental performance was produced mainly by a generative AI tool, identified by the platform's automatic detection systems rather than declared by the rights holder themselves.

Why is fraud so closely tied to streams of AI tracks?

The low production cost of these tracks makes it possible to generate a huge number of them quickly, which makes them easier to exploit through networks of artificial streams aimed at capturing advertising revenue or royalties without any real audience.

Will AI music replace artists on streaming platforms?

Nothing suggests that at this stage: AI tracks account for only 1 to 3% of actual listening on Deezer, despite their growing share of new uploads. The real issue is mainly about revenue dilution and the need for reliable detection tools.

Sources

This article was written with the help of artificial intelligence. Editorial policy

Vous aimerez peut-être aussi

Editorial policy and use of artificial intelligence