DSP Generative AI Stances Reference Guide

Client Relations

August 21, 2026 17:10

This article catalogs the known generative AI policies and stances of digital service providers (DSPs) in Symphonic's distribution network.

Acceptance of a release by Symphonic does not guarantee acceptance, monetization,
recommendation eligibility, or continued availability on every DSP. Each platform maintains its
own policies regarding AI-generated and AI-assisted content, and those policies may change at
any time. Symphonic will honor applicable DSP requirements and may restrict delivery to
specific platforms when content does not meet their requirements.

To view Symphonic's Generative AI stance, click here.

If you are using AI tools in your music video production (e.g., AI-generated visuals, effects, or editing), please disclose this when submitting your video so we can ensure proper handling across all delivery partners.

To view Symphonic's required AI disclosure on submissions, click here.

Apple

Policy URL: https://www.billboard.com/pro/apple-music-to-label-ai-generated-music/ (official policy page pending; announced Aug 20, 2026)

Stance: Transparency and labeling.

Key points:

Apple Music will introduce visible "Made With AI" labels later in 2026.
Labels are built on AI-disclosure metadata supplied by distributors and creators.
Symphonic note: ensure AI-disclosure metadata is populated on Apple deliveries so releases are labeled correctly when the feature goes live.

Bandcamp

Policy URL: https://blog.bandcamp.com/2026/01/13/keeping-bandcamp-human/
Stance: Full ban on AI-generated music

Key points:

Music and audio "generated wholly or in substantial part by AI" is not permitted
Any use of AI tools to impersonate other artists or styles is strictly prohibited
Reserves the right to remove any music on suspicion of being AI-generated
Community-driven enforcement: users are encouraged to report suspected AI-generated content via Bandcamp's reporting tools

Beatport

Policy URL: https://www.beatportal.com/articles/1571333-beatport-expands-beatdapp-partnership-to-detect-ai-generated-music?token=v7yB5zAw1FtmN1kPEKmd5zhGz96Gkohd

Stance: Bans fully/substantially AI-generated tracks; allows AI-assisted production (announced August 2026).

Key points:

Bans tracks "wholly or substantially generated by AI."
Prohibits unauthorized voice cloning; such uploads are rejected at ingestion.
AI-assisted production (mixing, mastering, stem separation, standard processing) remains allowed.
Expanded its Beatdapp partnership to scan uploads and classify them as human-made, AI-assisted, or fully synthetic.

Deezer

Policy URL: https://www.deezer.com/explore/en-us/features/ai-labelling/

Key points:

Uses proprietary AI detection tools to identify fully AI-generated tracks
Tags detected AI-generated content with a visible label for listeners
Removes fully AI-generated tracks from algorithmic and editorial playlists/recommendations
Found that approximately 85% of all streams on fully AI-generated tracks were fraudulent. Due to the fraudulent nature, they were demonetized by Deezer
Fake streams are excluded from royalty calculations, preserving the revenue pool for legitimate artists.
Deezer has made its AI detection technology available to the wider music industry, promoting transparency and fraud prevention beyond its own platform.
Updated (July 21, 2026): fully AI-generated music now exceeds 50% of daily uploads — roughly 90,000 tracks per day — up from about 44% in April 2026.

Qobuz

Policy URL: https://community.qobuz.com/ai-charter
Stance: Published a comprehensive AI Charter (May 2026) with a zero-tolerance policy on AI-generated content fraud. Editorial curation is 100% human.

Key points:

Core conviction: "AI can be a value amplifier, never a substitute for human judgment"
Editorial curation is 100% human — Editor's Picks, Albums of the Week, playlists are all manually selected by editorial teams; articles and reviews are written by humans
AI is used operationally for limited technical support (transcription, moderation, first-level customer support), improved search, and internal operations
Red lines — Qobuz will never: generate musical content to feed their catalog or calculate royalties; replace editorial curation, artistic validation, or creative decisions; promote fraudulent or 100% AI-generated content
Supports AI as a creative tool for human artists (mixing, mastering, composition aid) when it serves an authentic artistic process
Opposes "industrial production of AI-generated content, without human creative intervention, designed to saturate catalogs or defraud royalty systems"
Prohibits "crawling" or "scraping" of catalog content on Qobuz, or using any part of the services or content to train a machine learning or AI model
User searches and listening behavior are never used to train external AI models — only used internally to improve Qobuz's own music recommendations
Zero-tolerance policy on AI-driven streaming fraud: detection/monitoring systems, removal of fraudulent catalogs, exclusion of fraudulent streams from reporting and royalty payments
Contractual clauses prohibit delivery of 100% AI-generated content
Integrating dedicated DDEX fields for artists and labels to report AI-generated content (currently being deployed by suppliers)
Acknowledges detection technologies are imperfect and calls on the broader music industry to collaborate toward common standards
Charter is a living document; version dated May 2026

Spotify

Policy URL: https://newsroom.spotify.com/2026-08-11/ai-persona-badges-transparency/

Stance: Supports AI disclosure through industry-standard metadata while focusing enforcement on impersonation, deception, spam, and abusive mass-upload behavior.

Key points:

Relies on voluntary AI disclosure through DDEX metadata.
Vocal impersonation with AI voice clones is only allowed with the artist's authorization.
A music spam filter targets mass-uploaded and duplicate content.
New (Aug 11, 2026; rolling out from mid-September 2026): an "AI Persona" label marks artist profiles that represent an AI-generated identity, set by artist self-disclosure in Spotify for Artists or by Spotify review.
The label reflects the public identity, not how the music was made — no removal or demonetization is tied to it.
By default, AI Persona profiles are excluded from editorial, algorithmic, and personalized recommendations unless a listener actively follows them.

Tencent- QQ Music / Kugou / Kumo

Policy URL: https://www.musicbusinessworldwide.com/tencent-music-took-down-over-250000-songs...

Stance: AIGC tagging + detection, licensed-only training, active takedowns.

Key points:

Tags AI-generated content using audio models and runs a content-security detection process.
Works with rightsholders to remove unauthorized AI content — 250,000+ tracks removed and 600,000+ "high-risk" copyright cases reviewed in 2025.
Trains AI models only on licensed content; AI-specific clauses added to user agreements.
Offers creation tools (e.g., "AI Songwriter") with one-click publishing to QQ Music.
Subject to China's mandatory AI-content labeling law (effective September 1, 2025).

TIDAL

Policy URL: https://tidal.com/ai-policy

Key points:

Defines AI-generated music as music "wholly or substantially generated by generative artificial intelligence"
Will accept AI-generated music that meets content integrity standards
Tags AI-generated content with a visible icon (100% AI-generated content tagged starting mid-July 2026; substantially AI-generated tagging to follow)
AI-generated music is not monetizable — royalties will not knowingly be attributed to wholly AI-generated music
AI-generated music associated with fraudulent activity (deceptive content, high-volume uploads, unusual streaming patterns) will be blocked or removed
TIDAL Upload (independent artist uploads) is held to the same standards
Does not use uploaded music to train AI models; uses AI internally for moderation and metadata tools

TikTok

Policy URLs:

Music Terms of Service (AI Music section): https://www.tiktok.com/legal/page/global/music-terms-us/en
AI-generated content guide: https://support.tiktok.com/en/using-tiktok/creating-videos/ai-generated-content

Stance: Allows AI music with mandatory labeling requirements and strict enforcement.

Key points:

Music Terms includes a dedicated section (Section 6) on AI Music usage
Requires clear disclosure for AI-generated content via captions, hashtags (e.g., "AI-generated music"), or in-app AIGC toggle
Prohibits AI-generated vocals intended to create a false impression that a real person's voice is being used
Integrated C2PA Content Credentials (January 2025) — first major platform for automatic AI content detection
Enforcement intensified late 2025: removed over 51,000 synthetic media videos and permanently banned 8,600 accounts for AI policy violations
AI-assisted text workflows (scripts, captions, hashtags) are exempt from labeling — only visual/audio media require disclosure

Traxsource

Policy URLs:

Stance: Two-pronged policy — prohibits use of Traxsource content for AI training and provides transparent labeling of all uploaded music. Effective February 19, 2026.

Key points:

Principle 1 — No AI training on Traxsource content: Using any music on Traxsource as training data for AI models or services is a direct violation of their Terms of Service; Traxsource cannot grant or deny that authorization on behalf of rights holders
Principle 2 — Transparency through labeling: All music uploaded to Traxsource is scanned and labeled to indicate the level of AI content
Partnered with two industry-leading AI detection companies that power a three-part system: flagging fully AI-generated music for removal, identifying AI-assisted tracks for labeling, and confirming human-made work for a "Human-Made" tag
Fully AI-generated music may be removed from the platform
AI-assisted music is labeled (not removed) — defined as tracks made with help of AI tools but driven by a human artist
"Human-Made" tag is applied to confirmed fully human creations, displayed visibly to buyers and DJs
Draws a clear distinction between "AI as a production tool" (human-driven creative process using AI to assist — comparable to synthesizers, samplers, Auto-Tune) vs. "fully AI-generated music" (prompt-to-song with no meaningful human creative input)
Acknowledges that accurate AI detection is not yet perfect and that hybrid workflows make detection challenging; includes a rebuttal process for content providers who believe their work was labeled incorrectly
Labels classify music as either "Human" or "AI Assisted"
Traxsource reserves the right to override third-party labeling based on additional information at their sole discretion

YouTube

Policy URL (creator disclosure): https://support.google.com/youtube/answer/14328491
Stance: Mandatory AI disclosure for video creators; distributed music follows a different pathway.
Key points:

Important distinction: YouTube's AI disclosure policy (the toggle in YouTube Studio, the "inauthentic content" monetization rules) primarily applies to creators uploading videos directly through YouTube Studio — not to music delivered through distribution partners
Music delivered to YouTube Music through a distributor (like Symphonic) appears as audio-only content and follows standard distributor terms; no separate YouTube Studio AI disclosure is required from the artist
Symphonic does not deliver AI-generated content to YouTube or YouTube Content ID. Please see our UGC delivery guidelines for more information.
Content ID rules still apply: Impersonation and originality policies are enforced on distributed music — AI voice clones resembling real artists or content triggering copyright matches with existing songs can be rejected or taken down
YouTube's broader AI detection systems (including C2PA metadata scanning) are part of the ecosystem and may be used to identify AI-generated content regardless of how it was delivered
YouTube has stated it may use internal systems to detect AI-generated or altered content and automatically apply labels
YouTube CEO's 2026 priorities include managing "AI slop" across the platform

Symphonic Music Video Distribution

Symphonic accepts and delivers music videos to the partners that our clients select. However, there are important AI-related requirements and limitations to be aware of:

Apple Music and Vevo require disclosure if AI was used in the creation of the music video. Symphonic is required to disclose this information on your behalf when delivering to these partners, so it is important that you communicate any AI usage in your video during the submission process.
Videos deemed fully AI-generated are excluded from Symphonic marketing services, including playlist pitching. If your music video is determined to be fully generated by AI, it will still be delivered to your selected partners, but it will not be eligible for Symphonic's marketing and promotional support.

DSP Content Removal & Policy Enforcement

It is important to understand that each DSP reserves the right to remove, restrict, demonetize, or decline content that does not comply with their individual platform policies — including their AI-related policies. This can happen at any time, with or without prior notice, and applies to content that has already been delivered and is live on the platform. Removal decisions are made at the sole discretion of each DSP and are outside of Symphonic's control.

Because AI policies vary from platform to platform, a release that is accepted on one DSP may be flagged, labeled, demonetized, or removed on another. We strongly recommend reviewing each DSP's policy before distributing content that involves generative AI tools, and ensuring that your releases comply with the specific requirements of each platform you are distributing to.

If your content is removed or flagged by a DSP for AI-related policy violations, Symphonic will work with you to understand the situation, but we cannot override a DSP's content decisions.

Misuse of Generative AI: Voice Cloning, Likeness & Impersonation

Across the industry, one of the most consistently enforced areas of AI policy is the misuse of generative AI to replicate, imitate, or impersonate real artists. This includes but is not limited to:

Voice cloning or vocal deepfakes — Using AI to generate vocals that resemble or imitate the voice of a real artist without that artist's explicit authorization
Unauthorized use of likeness — Using AI to generate content (audio, visual, or otherwise) that creates a false impression that a real person is involved in or endorses the work
Style impersonation with intent to deceive — Releasing AI-generated music under an artist name, profile, or brand designed to mislead listeners into believing the content is from an existing artist
Fraudulent delivery to artist profiles — Uploading AI-generated content (or any unauthorized content) to another artist's existing profile on a DSP

Nearly every major DSP — including Spotify, TIDAL, YouTube, TikTok, Deezer, Bandcamp, and Apple Music — has policies that prohibit some or all of these practices. Violations can result in content removal, demonetization, account suspension, or permanent bans, depending on the platform.

Symphonic does not permit the distribution of content that uses AI to impersonate, clone, or misrepresent real artists without proper authorization. If you are using AI voice or likeness tools in your music, ensure that you have obtained all necessary permissions and that your content complies with both Symphonic's policies and the policies of each DSP you are distributing to.

This document is based on publicly available information as of August 21, 2026. AI policies across DSPs are evolving rapidly and should be reviewed periodically for updates.

DSP Generative AI Stances Reference Guide

Comments

Articles in this section