Within Music
Could AI Flood The Music Market?
Generative AI could increase the supply of cheap tracks, making discovery, payment and human creative value harder to protect.
On this page
- Low cost generation and scale
- Substitution and platform clutter
- Human creators and listener trust
Page outline Jump by section
Introduction
AI music raises market flooding fears because it changes the economics of supply. A human-made recording usually takes time, skill, coordination and money to write, perform, produce, mix, master and release. Generative AI can produce large numbers of acceptable-sounding tracks quickly and cheaply, especially in genres where listeners are using music as background, mood-setting or playlist filler. That does not mean every AI track is worthless, or that human musicians cannot use AI creatively. The fear is more specific: when platforms are paid by attention and catalogues are already crowded, a surge of cheap synthetic music can make discovery harder, dilute royalty pools, invite fraud and weaken listener trust in whether an artist, voice or backstory is real.
The clearest warning sign is volume. Deezer said in April 2026 that it was receiving almost 75,000 fully AI-generated tracks per day, roughly 44% of daily uploads, amounting to more than 2 million AI tracks per month. Deezer also said these tracks represented only 1–3% of total streams on its platform, but that 85% of streams on fully AI-generated tracks were detected as fraudulent and demonetised. [Deezer Newsroom]newsroom-deezer.comSource details in endnotes.
Why cheap generation changes the supply problem
The music market has always had more songs than any one listener could hear. What AI changes is the cost curve. A creator, spammer, agency or anonymous operator can generate multiple tracks in the time it would once have taken to record a demo. The result is not simply “more music”; it is a different kind of supply, optimised for speed, quantity and surface-level adequacy.
That matters because streaming platforms sort music through recommendation systems, playlists, metadata, user behaviour and catalogue ingestion pipelines. A platform can hold millions of tracks, but listener attention remains scarce. When low-cost tracks multiply, the bottleneck moves from production to filtering: which songs get indexed properly, recommended, paid, removed, labelled or trusted?
The strongest flooding fears come from three connected mechanisms:
- Scale without proportional effort: AI lowers the barrier to producing tracks that are “good enough” for ambient, lo-fi, sleep, focus, instrumental, genre-pastiche or short-form social use.
- Discovery congestion: more tracks compete for the same playlist slots, searches, recommendation surfaces and listener sessions.
- Payment pressure: if synthetic tracks attract real or manipulated streams, they can draw from royalty systems designed around human catalogues and genuine listening.
This is why market flooding is not just a taste argument. It is a platform-design problem. A million extra tracks do not harm the market merely by existing; they become disruptive when they are difficult to label, cheap to mass-upload, easy to manipulate and financially eligible under the same systems as human-made music.
Low-cost tracks can crowd the middle of the catalogue
AI music is most threatening where music functions as a commodity: background instrumentals, mood playlists, corporate videos, social clips, library music, meditation tracks, generic genre cues and low-attention listening. In those spaces, the buyer or listener may not care who made the track if it sounds acceptable, arrives quickly and costs less.
CISAC, the international confederation of authors’ societies, warned in a 2024 economic study that music creators could have 24% of revenues at risk by 2028 under unchanged conditions. The same study projected that generative AI music could account for about 20% of traditional music streaming platform revenues and around 60% of music library revenues by 2028. [CISAC]cisac.orgOpen source on cisac.org.
That distinction matters. The immediate risk is not that AI replaces every beloved artist. Fans still care about voice, personality, performance, history, identity, live shows and cultural meaning. The sharper risk is that the lower and middle layers of the market become swamped with cheap substitutes: music that is not trying to be a career, a scene or a human statement, but is good enough to fill space.
For working musicians, those middle layers are not trivial. Library placements, session work, production music, background catalogue income and modest streaming earnings can help sustain careers between tours, commissions or releases. If AI systems can generate endless “usable” tracks in those categories, the market may not collapse dramatically; it may erode quietly.
Streaming rewards make flooding financially tempting
Streaming services do not usually pay per uploaded track. They pay when tracks are streamed, and payouts are shaped by platform rules, subscription revenue, territory, rights ownership and fraud controls. That makes flooding attractive to bad actors: they do not need one AI song to become a hit if they can upload huge numbers of tracks and generate small amounts of listening across them.
Spotify’s own artist guidance defines artificial streaming as listening that does not reflect genuine user intent, including manipulation through bots or scripts, and warns that undetected artificial streams dilute the royalty pool by shifting revenue from legitimate artists to bad actors. [Spotify for Artists]artists.spotify.comfor Artists Artificial Streaming – Spotify for Artistsfor Artists Artificial Streaming – Spotify for Artists
The Michael Smith case shows how this can move from theoretical risk to criminal fraud. In March 2026, the U.S. Attorney’s Office for the Southern District of New York announced that Smith had pleaded guilty after creating hundreds of thousands of AI-generated songs and using bots to stream them billions of times, mimicking genuine consumer activity in order to obtain royalties. [Department of Justice]justice.govSource details in endnotes.
That case is important because it shows why “AI flooding” is not just about mediocre songs appearing in search results. It can combine three scalable systems: automated music generation, automated account creation and automated listening. Each part amplifies the others. The more tracks a fraudster can generate, the easier it is to spread streams thinly enough to look less suspicious; the more fake listening can be distributed, the more royalty extraction can be hidden among ordinary platform traffic.
Platform clutter is not only about bad songs
A common misunderstanding is that the market will simply ignore weak AI music. In a perfect discovery system, bad tracks would sink and good tracks would rise. Real platforms are messier. Metadata can be gamed, uploads can be duplicated, artist names can be confusing, tracks can be made artificially short, and recommendation systems can be tested by people trying to find loopholes.
Spotify responded to these pressures in September 2025 by announcing stronger AI protections, including improved enforcement against impersonation, a new spam filtering system and AI disclosures through industry-standard credits. The company said AI tools had made vocal deepfakes easier and clarified that vocal impersonation is allowed on Spotify only when the impersonated artist has authorised it. [Spotify]artists.spotify.comfor Artists Artificial Streaming – Spotify for Artistsfor Artists Artificial Streaming – Spotify for Artists
Deezer has taken a more interventionist approach for detected fully AI-generated tracks. In its April 2026 update, it said AI tracks were removed from algorithmic recommendations and editorial playlists, and that fraudulent streams were excluded from royalty payments. It also said it had stopped storing high-resolution versions of AI tracks, a sign that catalogue scale creates storage and infrastructure choices as well as cultural ones. [Deezer Newsroom]newsroom-deezer.comSource details in endnotes.
This is the platform-clutter problem in practical terms. Flooding forces services to decide what counts as legitimate music, what deserves recommendation, what needs labelling, what should be demonetised and what should be removed. Those choices are not neutral: strict filtering may protect artists and listeners, but it can also create disputes over false positives, legitimate AI-assisted work and who gets to define meaningful human contribution.
Human creators fear substitution, not just competition
Competition is normal in music. New genres, cheap home recording, sampling, drum machines and digital distribution have all changed who can make and release songs. The anxiety around AI is different because AI systems can produce outputs that compete with musicians while also being trained, at least in many contested cases, on existing human-made recordings and compositions.
The legal fight around AI music companies Suno and Udio reflects that tension. Major labels sued the companies in 2024, alleging that copyrighted recordings had been copied to train systems that could generate music competing with human artists. Reuters reported in November 2025 that Warner Music Group had settled its case with Suno, enabling licensed AI models in 2026; Reuters also noted that Udio had recently settled copyright disputes with Warner and Universal, while Suno and Udio had argued that training on copyrighted recordings was fair use. [Reuters]reuters.comOpen source on reuters.com.
For human creators, the flooding fear is therefore double-edged. First, AI music may compete in the market for attention and revenue. Second, the capability to produce that competing music may have been built from existing creative labour without permission, payment or transparent accounting. CISAC framed this as a two-front loss: unauthorised use of creators’ works by AI models, followed by substitution of human-made revenue streams by AI-generated outputs. [CISAC]cisac.orgOpen source on cisac.org.
That is why many musicians object less to “technology” in the abstract than to an industrial pattern: scrape creative work, build a generator, flood the market with substitutes, then ask the original creators to compete against the machine.
Listener trust becomes harder to protect
The market flooding fear is also a trust problem. Listeners do not only consume sound; they often form attachments to artists, scenes, stories, performances and identities. If a convincing band, voice or biography can be synthetic, listeners may start doubting what they are being offered.
A Deezer-Ipsos survey reported by Reuters in November 2025 found that 97% of listeners could not distinguish between AI-generated and human-composed songs. The same report said 73% supported disclosure when AI-generated tracks are recommended, 45% wanted filtering options, and 40% said they would skip AI-generated songs entirely. [Reuters]reuters.comOpen source on reuters.com.
The Velvet Sundown became a vivid example of this anxiety. Reuters reported that the AI band attracted one million monthly Spotify listeners before its synthetic origins were exposed. The issue was not just that listeners heard AI-generated music; it was that the project blurred music, image and identity in ways that made ordinary fan judgement less reliable. [Reuters]reuters.comOpen source on reuters.com.
That distinction is crucial. Some listeners may happily choose AI music when it is labelled, especially for background use. The trust problem arises when AI music enters the market disguised as a human act, mimics an existing artist, or uses a fictional backstory to capture attention that listeners thought they were giving to real people.
Why platforms are splitting into different policy camps
There is no single industry response because platforms serve different roles. A mass streaming service wants scale, catalogue breadth and user retention. A direct-to-fan marketplace may prioritise human identity and artist support. A music library may care about licensing clarity and cost. A social platform may care about frictionless soundtrack creation.
Bandcamp drew one of the clearest lines in January 2026, stating that music and audio generated wholly or in substantial part by AI is not permitted on the platform, and that AI tools used to impersonate other artists or styles are prohibited. Bandcamp framed the rule around fan confidence that the music they find there was created by humans. [Bandcamp Updates]blog.bandcamp.comUpdates Bandcamp’s Mission and Our Approach to Generative AIUpdates Bandcamp’s Mission and Our Approach to Generative AI
Spotify’s approach, by contrast, has focused more on impersonation, spam filtering and disclosure rather than a blanket ban. Deezer has emphasised detection, tagging, recommendation exclusion and demonetising fraudulent streams. These differences reflect different business models as much as different ethics. A platform built around direct support may see AI flooding as a threat to its core promise. A platform built around universal access may try to manage AI content rather than exclude it.
The unresolved question is whether labelling and filtering will be enough. If AI music remains a small share of actual listening, platforms may be able to contain the risk. If upload volumes keep rising and synthetic tracks become harder to detect, the cost of policing catalogues may rise sharply.
The real fear is a worse music market, not simply more music
A flooded market does not mean every listener will suddenly prefer AI songs, or that human artistry will disappear. The stronger concern is that music’s weakest economic points become weaker: discovery becomes noisier, small royalty streams become easier to divert, background music becomes more substitutable, and listeners become less sure what is human.
The harm would not be evenly distributed. Major artists with loyal fanbases, touring power and strong brands may be better insulated. Independent musicians, session players, composers for libraries, playlist-dependent artists and new acts trying to be discovered may face the greatest pressure. The market can remain full of brilliant human music while still becoming harder for many human musicians to survive in.
That is why the debate over AI flooding is ultimately about market design. The key choices are not whether AI music can exist, but whether platforms require disclosure, whether synthetic tracks are eligible for the same recommendations and payouts, whether training is licensed, whether impersonation is quickly removed, and whether fraud systems can keep up with industrial-scale generation. Without those safeguards, the fear is not a future with too many songs. It is a future in which cheap synthetic abundance makes human musical value harder to find, fund and trust.
Amazon book picks
Further Reading
Books and field guides related to Could AI Flood The Music Market?. Use these as the next step if you want deeper reading beyond the article.
The Future of the Music Business
Useful framework for evaluating AI-era music economics.
eBay marketplace picks
Marketplace Samples
Example marketplace items related to this page. Use the search link to explore similar finds on eBay.
Endnotes
-
Source: cisac.org
Link: https://www.cisac.org/Newsroom/news-releases/global-economic-study-shows-human-creators-future-risk-generative-ai -
Source: artists.spotify.com
Title: for Artists Artificial Streaming – Spotify for Artists
Link: https://artists.spotify.com/en/artificial-streaming -
Source: justice.gov
Link: https://www.justice.gov/usao-sdny/pr/north-carolina-man-pleads-guilty-music-streaming-fraud-aided-artificial-intelligence-0 -
Source: newsroom.spotify.com
Title: Strengthens AI Protections for Artists, Songwriters, and Producers — Spotify
Link: https://newsroom.spotify.com/2025-09-25/spotify-strengthens-ai-protections/ -
Source: reuters.com
Link: https://www.reuters.com/legal/litigation/warner-music-group-settles-copyright-case-with-suno-licensed-ai-music-2025-11-25/ -
Source: reuters.com
Link: https://www.reuters.com/legal/litigation/are-you-listening-bots-survey-shows-ai-music-is-virtually-undetectable-2025-11-12/ -
Source: blog.bandcamp.com
Title: Updates Bandcamp’s Mission and Our Approach to Generative AI
Link: https://blog.bandcamp.com/2026/01/13/keeping-bandcamp-human/ -
Source: support.spotify.com
Title: third party services that guarantee streams
Link: https://support.spotify.com/us/artists/article/third-party-services-that-guarantee-streams/ -
Source: cisac.org
Link: https://www.cisac.org/services/reports-and-research/cisacpmp-strategy-ai-study -
Source: justice.gov
Link: https://www.justice.gov/usao-sdny/pr/north-carolina-musician-charged-music-streaming-fraud-aided-artificial-intelligence -
Source: justice.gov
Link: https://www.justice.gov/usao-sdny/media/1366241/dl -
Source: get.bandcamp.help
Link: https://get.bandcamp.help/en/articles/15263124-bandcamp-s-acceptable-use-and-moderation-policy -
Source: newsroom-deezer.com
Link: https://newsroom-deezer.com/2026/04/ai-generated-tracks-represent-44-of-new-uploaded-music/ -
Source: facebook.com
Link: https://www.facebook.com/MixmagMagazine/posts/spotify-has-announced-a-crackdown-on-ai-revealing-it-has-removed-75-million-spam/1213225667514090/ -
Source: consequence.net
Title: spotify ai protections
Link: https://consequence.net/2025/09/spotify-ai-protections/ -
Source: apraamcos.com.au
Title: cisac ai report
Link: https://www.apraamcos.com.au/about-us/news-and-events/cisac-ai-report -
Source: prsformusic.com
Title: cisac generative ai study music creators future at risk
Link: https://www.prsformusic.com/m-magazine/news/cisac-generative-ai-study-music-creators-future-at-risk -
Source: imusician.pro
Title: spotify ai policy in 2025 artist protection transparency
Link: https://imusician.pro/en/resources/blog/spotify-ai-policy-in-2025-artist-protection-transparency -
Source: forbes.com
Title: spotify tightens ai policy and trims catalog
Link: https://www.forbes.com/sites/billrosenblatt/2025/09/26/spotify-tightens-ai-policy-and-trims-catalog/ -
Source: entertainment.slashdot.org
Title: spotify announces new ai safeguards says its removed 75 million spammy tracks
Link: https://entertainment.slashdot.org/story/25/09/25/2211230/spotify-announces-new-ai-safeguards-says-its-removed-75-million-spammy-tracks -
Source: socanmagazine.ca
Title: cisac releases study of ais economic impact on music and screen media
Link: https://www.socanmagazine.ca/news/cisac-releases-study-of-ais-economic-impact-on-music-and-screen-media/ -
Source: theverge.com
Title: spotify ai slop impersonation disclosure
Link: https://www.theverge.com/news/785136/spotify-ai-slop-impersonation-disclosure -
Source: musiccreatorsna.org
Title: cisac genai study 2024
Link: https://www.musiccreatorsna.org/cisac-genai-study-2024/
Additional References
-
Source: youtube.com
Title: AI music and the battle for royalties
Link: https://www.youtube.com/watch?v=G3cZq0k8wT0Source snippet
These videos explain how the low cost and high volume of AI-generated content overwhelm streaming platforms, creating discovery congestio...
-
Source: instagram.com
Link: https://www.instagram.com/p/DZCLxGljpQ9/?img_index=2 -
Source: musicbusinessworldwide.com
Link: https://www.musicbusinessworldwide.com/75000-ai-generated-tracks-now-flood-deezer-daily-representing-44-of-all-new-music-uploaded-to-the-platform-says-streamer/ -
Source: goldmedia.com
Link: https://www.goldmedia.com/fileadmin/goldmedia/Studie/2023/GEMA-SACEM_AI-and-Music/AI_and_Music_GEMA_SACEM_Goldmedia.pdf -
Source: reddit.com
Link: https://www.reddit.com/r/BandCamp/comments/1qbw8ba/ai_generated_music_on_bandcamp/ -
Source: facebook.com
Link: https://www.facebook.com/uchicago/posts/ai-music-is-flooding-streaming-platformsaccounting-for-about-half-of-all-new-son/1405448544947773/ -
Source: reddit.com
Link: https://www.reddit.com/r/indieheads/comments/1lmwgff/aigenerated_psychrock_band_the_velvet_sundown/ -
Source: facebook.com
Link: https://www.facebook.com/groups/2703779406444247/posts/3494933250662188/ -
Source: instagram.com
Link: https://www.instagram.com/p/DYsg9cSlpdP/ -
Source: facebook.com
Link: https://www.facebook.com/djmagazine/posts/a-new-report-shared-by-deezer-has-revealed-that-28-of-music-uploaded-to-the-plat/1327629069031267/
Topic Tree