AI Generated Music Is No Longer a Niche Experiment
On major music streaming apps, AI generated music has quietly moved from novelty to industrial scale. Deezer reports it now receives around 75,000 AI generated tracks per day, about 44% of all daily uploads. Its in-house AI detection tool, active since 2025, has already tagged more than 13.4 million tracks as synthetic audio. So far, listeners are not gravitating to this content in large numbers: Deezer says AI tracks account for only 1–3% of total streams, and many of those are demonetized. But platforms worry less about today’s listening share and more about tomorrow’s catalog balance and data integrity. When nearly half of new uploads are synthetic, it raises urgent questions: what happens to discovery, to playlist recommendations, and to the visibility of human artists when algorithms must sift through a growing ocean of machine-made music?
How AI Floods Distort Playlists and Discovery
Music streaming apps rely on algorithms to power playlist recommendations, auto-mixes, and background listening stations. These systems read signals such as play counts, skips, and completion rates to decide what to surface next. When catalogs are packed with low-effort AI generated music, those signals become easier to game. The World Intellectual Property Organization has warned that bot-driven streams can skew recommendation algorithms, making it harder for genuine artists to gain visibility and warping the data that drives touring and marketing decisions. At scale, this means users may encounter more generic, AI-made filler in radio-style stations or obscure playlists, even if they never search for it. The concern is not that one viral AI hit replaces a human song, but that millions of near-identical tracks quietly crowd the system, nudging discovery away from real creative voices and toward whatever content is most easily mass-produced and manipulated.
Fraud, Fake Artists, and the Battle Over Royalties
The surge in AI uploads is closely tied to music royalties fraud. One common scheme involves uploading a relatively small catalog and using automated bots to stream it repeatedly. Increasingly, fraudsters are going bigger: flooding platforms with millions of AI generated songs, then streaming each track just enough to trigger royalty payouts. Musicians like Michael Smith have reportedly earned close to USD 10 million (approx. RM46,000,000) by uploading thousands of AI-made tracks and deploying bots to play them in small increments. AI tools now generate not only the music but also the bots themselves. A study by CISAC and PMP Strategy warns that up to 25% of creators’ revenues, equivalent to billions of dollars, could be at risk within a few years. To push back, Deezer has introduced an artist-centric model that caps streams from individual users at 1,000, sharply reducing royalties when that threshold is exceeded in an effort to blunt bot activity.
How Streaming Platforms and Regulators Are Responding
Streaming platforms are under pressure from both artists and regulators to prove that their systems are fair. Deezer has already built AI detection tools and a new remuneration model designed to limit fraud. More broadly, services are exploring content labeling for AI generated music, automated detection of synthetic audio, and policies that throttle or demonetize suspicious uploads and streams. At the same time, political scrutiny of streaming practices is rising. In Texas, the attorney general has launched an investigation into major music streaming apps over alleged payola-style promotion schemes, focusing on how songs and artists are boosted in editorial and algorithmic playlists. While that probe is not specifically about AI, it underscores a larger concern: when algorithms decide what people hear, any form of artificial manipulation—from bribes to bots—can distort the musical landscape and undermine trust in recommendation systems.
How Listeners Can Support Human Artists and Spot the Upside of AI
Fans are not powerless in this reshaped ecosystem. If you want to prioritize human-made music on music streaming apps, start by leaning on trusted, human-curated playlists rather than generic auto-generated ones. Follow artists directly, add full albums to your library, and save tracks you discover at shows or through social media to send clearer signals to recommendation engines. Where credits are available, check for named songwriters, producers, and featured musicians; anonymous or obviously templated profiles can be a red flag for low-effort AI catalog spam. At the same time, not all AI generated music is harmful. Many artists use AI as an assistive tool—for sound design, demoing, or experimental side projects—without replacing their creative role. For listeners, the upside lies in supporting creators who use AI transparently and thoughtfully, while resisting the passive drift toward faceless, spam-driven background sound.