YouTube Music is rolling out labels that show when a song was created primarily using generative AI, YouTube announced on October 14, 2026. The company said the labels, which cover cases such as a generative model creating a track’s composition and vocals, will also expand to the main YouTube app.
When a listener plays a track on YouTube Music in which AI played a significant role, the label appears directly on the player page. Tapping the label opens a simple menu with more information about the track.
YouTube said it is working with music partners and using its own detection tools to identify AI-generated tracks, applying the same labeling principles it already uses across YouTube. The company said listeners have told it they want transparency when they are hearing an AI-generated track, and described its goal as giving viewers and listeners clear information about the music they hear.
Gen AI Designations for Music Partners
Disclosure guidelines for music content in the YouTube Help Center direct music partners to report generative AI use so YouTube can accurately treat and label media that contains it. When delivering metadata, partners choose from affirmative designations: Fully Gen AI, meaning all content was created using generative AI; Partly Gen AI, meaning part of the content was created using generative AI; and No Gen AI, meaning no content was. If a partner provides no value, YouTube interprets it as unknown unless it can make a determination based on other signals.
The guidelines treat a track as Fully Gen AI when a user enters a text prompt into an AI music generator and downloads the resulting track, or when an artist uses a GenAI model to completely generate a song’s composition, instrumental performance, and vocals, iterating with multiple prompts before downloading. Partly Gen AI examples include a GenAI tool generating structural layers such as a bassline or string section beneath an artist’s own live vocals and instruments; GenAI tools used to brainstorm lyrical themes, co-write verses, or generate and iterate on melodies that are then recorded in the studio; and a song performed entirely by humans but paired with visuals produced by a text-to-video AI model.
No Gen AI covers an engineer using standard pitch-correction software such as Auto-Tune to correct minor tuning mistakes in a live vocal performance, and a producer using automated AI mastering software to adjust a track’s final equalization, compression, and sonic balance. The page notes that it is up to partners to interpret the Gen AI definitions provided by DDEX in relation to their own content.
YouTube also keeps a classification role of its own. The Help Center states that YouTube may use additional signals to augment partner declarations, and that when a partner includes no Gen AI information for a piece of content, YouTube will rely solely on those signals and may designate the content as fully or partly Gen AI.
Delivery Through DDEX, CSV, and YouTube Studio
Partners can supply Gen AI information when delivering content through DDEX ERN 4.3.2+ or CSV delivery templates pursuant to their SRAV terms; uploaders who do not use DDEX or CSV can instead mark disclosures in the attribute section of YouTube Studio. For DDEX deliveries, the guidelines point to reference resources that include an XML example demonstrating how to signal AI contribution for individual tracks on an album, as well as “AI contributor composite” examples in the DDEX 4 Composites article.
For CSV deliveries, the GenAiDeclarationCsvTemplate, available in Studio Content Manager, requires two columns: genai_declaration_asset_id, which specifies any YouTube Asset ID where the content owner holds ownership, including sound recordings, art tracks, music videos, and web videos; and contains_ai, which accepts the values “all,” “partly,” “none,” or “unknown,” with “unknown” clearing and resetting a previously misdeclared value. For new content deliveries or full metadata updates, the optional contains_ai column is supported across the Asset Update, Audio-Art Tracks, Audio-Sound Recording Upload/Update, Music Video, and Web Video templates.
The Disclosure Framework Behind the Labels
The announcement points users to YouTube’s broader “How this content was made” disclosures. Under that framework, a “Made with AI” disclosure indicates that a video contains content meaningfully altered or synthetically generated with Gen AI tools. It can appear in the video player or in the expanded description when a creator manually discloses AI content in the Studio workflow, when creators use YouTube’s own generative AI tools such as DreamScreen, when valid Content Credentials data indicates the entire video was made with AI, or when YouTube, in some cases, proactively applies a label to undisclosed content to reduce the risk of harm to viewers.
YouTube carries forward disclosures available with secure Content Credentials (C2PA) 2.1 or higher that indicate the entire video was made with AI, and where those credentials include a signing authority, an “Info from” attribution beneath the disclosure names the source that validated the information.
In a May 27, 2026 post, the company said it had been labeling content based on creator disclosure since 2024 and announced it was moving the disclosure label for photorealistic and meaningfully AI-altered or generated content to a position directly below the video player on long-form videos and onto Shorts as an overlay. The post described that as the single label format for such content on YouTube, and said disclosures for unrealistic, animated, or slightly altered content can be found in the expanded description.
The same post said that starting in May 2026, YouTube was rolling out internal signals to help identify AI-generated content, automatically applying a label when its systems detect significant photorealistic AI use and the creator has not specified whether AI was used. Creators who believe their content was incorrectly identified as AI-generated can update the disclosure status in YouTube Studio, though disclosures remain permanent in a handful of cases, including content created with YouTube’s own AI tools such as Veo or Dream Screen and content containing C2PA metadata indicating it was fully generative AI. YouTube stated that a disclosure label alone does not change how a video is recommended or whether it is eligible to earn money.
Credit: Source link
