Bandcamp Bans AI Music Outright as Streamers Split Over Labeling
Bandcamp has banned AI music entirely, Deezer says AI tracks exceed 50% of peak daily uploads, and Spotify will tag AI artists from September. Each platform is drawing its own line.

Programme
- Bandcamp announced a total ban on AI-generated songs in January, a first for online music platforms.
- Deezer reports AI uploads exceeded 50% of daily tracks at peak level in June 2026, up from 10% in January 2025.
- Spotify's 'AI Persona' label for AI-generated artists begins rolling out in mid-September.
- Tidal announced in June it will not monetize AI-generated content and will hold distributors responsible for identifying it.
- Apple Music's Made With AI labeling system is set to roll out by the end of 2026.
Bandcamp announced a total ban on AI-generated songs in January — a first for an online music platform — as major streaming services fragment into sharply different policies on the flood of AI-made tracks now dominating daily uploads.
The scale is measurable and growing fast. Deezer reports that daily AI track uploads exceeded 50% at "peak level" in June 2026. Apple Music told Billboard in April that "more than a third" of monthly uploads are 100% AI-generated. Qobuz noted in January that 34% of new daily uploads are AI-made. Even those figures may undercount the total, because partially AI-generated music is far harder to detect, and no service can tell when AI wrote lyrics or melodies later recorded by human musicians.
What does each platform actually do?
The answer differs sharply by service, and Billboard is tracking the policies as an updating list.
-
Bandcamp: The strictest stance. A total ban on AI-generated songs, whether AI is used in all or a "substantial part" of the creative process. It also bars AI tools used "to impersonate other artists or styles." iHeartRadio set a precedent months earlier, banning AI music "that features synthetic vocalists pretending to be human" across its nationwide radio stations.
-
Deezer: The most outspoken service. It began publishing detection figures in January 2025, when 10% of daily uploads were fully AI-generated. It labels fully AI tracks with its in-house detector, excludes them from algorithmic and editorial recommendations, and as of July removes AI tracks suspected of streaming fraud or unstreamed for six months or more. Deezer has said up to 85% of streams on fully AI tracks were fraudulent.
-
Spotify: The most layered approach. Policies against spam, impersonation and streaming fraud were announced last year. In April 2026 Spotify let creators disclose AI use in credits, down to a single guitar lick or the entire production. It added verification badges for human creators and is rolling out Artist Profile Protection, which locks artist pages against unauthorized uploads. Its boldest move came in August: a new label, "AI Persona," applied to AI-generated artists from mid-September, with their music excluded from personalized recommendations.
Apple Music: Visible labels are coming. The Made With AI system, announced in August, rolls out by the end of 2026. Songs tagged as "materially generated using AI" by content providers will carry the label. A March requirement already obliges providers to disclose AI use in artwork, recording, composition or music videos. The system relies on disclosure, not automated detection — though Apple Music head Oliver Schusser has told Billboard the company runs an internal detector to monitor AI content's rise.
SoundCloud: A ban with an embrace attached. It bars AI content that impersonates artists or clones voices, and requires AI uploads to meet the same standards as any other music. It says it does not train generative AI on its catalog and will not without prior consent. In a November 2026 open letter, CEO Eliah Seton promised protections against users who spam the platform with AI slop, stronger defenses against streaming manipulation, and blocks for violators. Yet SoundCloud also partners with licensed AI music companies, including ElevenLabs and Moises, giving AI users a direct pipeline to share work.
Tidal: Some of the industry's strictest rules, announced in June. Tidal will not monetize AI-generated content and makes distributors responsible for identifying it before upload. Its own labeling began in mid-July, starting with fully AI-generated music and expanding to "substantially AI-generated" tracks as detection improves. "The responsibility to identify and tag AI-generated content should not rest with Tidal alone," the company said in a press release. "We expect — and will begin to enforce — that content distributors identify AI-generated content before it reaches our platform."
Qobuz: A "human-first stand on AI-generated music," announced in January. Its editorial curation — playlists, articles, reviews — will remain "100% human," and its Discover page will "filter out AI-generated tracks." Qobuz will identify fraudulent streams, exclude them from the royalty pool, and now has contractual clauses prohibiting delivery of 100% AI-generated music, plus dedicated DDEX fields for reporting AI content.
YouTube: Requires disclosure of generative AI use in uploads, including music, and adds a transparency label. Beyond that, it has no AI-music-specific policies. Parent company Google is pursuing its own AI music model, Lyria 3, after acquiring ProducerAI earlier this year.
Amazon Music: Systems exist to prevent impersonation, deception and spam, but no AI-specific policies. "The broader music community is defining what labeling standards should look like, and we're part of those conversations," Amazon said in a statement. "As those discussions continue, we'll continue to explore ways to help customers make informed choices about what they're listening to."
Why enforcement is hard
Every system faces the same traps. AI-detection tools risk false positives. Self-disclosure invites deception at worst and innocent omission at best, given how easily AI can enter any stage of the recording process.
What does this change for artists and listeners?
For now, it creates a patchwork: an AI track banned on Bandcamp may be labeled on Deezer, tagged as an AI Persona on Spotify, or pass unmarked on Amazon Music. Until regulators or the industry converge on shared labeling standards, the platform you choose determines how much AI you hear — and whether you know it.
Source: artsjournal.com
More from Marcus Bennett
Show full bio
Staff writer covering people, launches and day-to-day practice in arts in the broadest sense at Arts & Voices.
1,586 articles



