YouTube Music Begins Marking Tracks Created With Generative AI

YouTube Music is beginning to identify songs made or substantially altered with generative artificial intelligence, giving listeners a clearer indication of how a track was produced as AI-generated audio becomes more common across streaming platforms.

YouTube announced the rollout on Oct. 14 in the United States. The new "Made with AI" designation will appear on the track’s playback screen in the YouTube Music app. Users can tap the label to see additional information about the role AI played in creating the song. YouTube plans to bring the disclosure to the main YouTube app at a later stage, but has not provided a timetable for that expansion or detailed a country-by-country launch schedule.

The policy is aimed at music in which AI has played a significant creative role, including cases where a generative model produced the composition or vocals. It is not intended to classify every song that has passed through an AI-assisted or digital editing tool. Conventional production techniques, such as routine pitch correction, are expected to be treated differently from using generative systems to create major musical elements.

YouTube said the label is designed to provide production information rather than remove AI-made music from the service. The company will rely on disclosures from music distribution partners and will also use its own detection technology and other signals when a track has not been identified by its uploader.

The system extends a disclosure approach YouTube already uses for video. The company has been expanding notices for realistic-looking footage that was generated or materially modified with AI, placing the disclosure below standard videos and overlaid on Shorts. Applying a similar concept to music makes the information available while a listener is viewing the playback interface rather than requiring a separate search through credits or metadata.

The rollout comes as streaming companies adopt different approaches to synthetic music. Spotify has been testing voluntary credits that describe AI involvement in vocals, lyrics and production. It has also introduced disclosures for fictional artist identities created with AI and has said it does not intend to include those artists in ordinary recommendation systems.

Deezer has taken a more restrictive route by using detection tools to identify AI-generated tracks and excluding them from editorial and algorithmic recommendation playlists. The French platform has also added a filtering option for listeners who do not want such material included in their listening experience. Bandcamp, meanwhile, banned AI-generated music from its platform earlier this year, according to the material reviewed for this report. Apple Music has also been reported to be considering an AI tag.

The distinctions among these policies reflect an unresolved industry question: whether AI disclosures should primarily inform listeners, limit recommendation exposure or keep synthetic recordings off a service altogether. YouTube’s initial measure falls in the first category. It does not announce a blanket prohibition or state that every labeled song will receive less exposure in search, playlists or recommendations.

For artists, labels and distributors, the change places greater emphasis on accurately documenting how a track was made. A failure to disclose generative AI use may not prevent YouTube from labeling a song if its systems identify indications of synthetic production. That gives the platform a role in deciding not only how music is presented, but also how its production history is communicated to audiences.

The policy may also affect the way music services compete over trust and discovery. A simple label can help listeners distinguish fully synthetic recordings from works that use AI for a limited element, but the usefulness of the system will depend on the detail supplied when the label is opened and the reliability of YouTube’s detection process. The company has not disclosed the technical thresholds it will use or explained how it will handle disputed classifications.

For the broader music business, the move adds another layer to the emerging infrastructure around AI-generated content. Platforms are increasingly separating disclosure from enforcement: some are providing credits, while others are filtering or banning synthetic tracks. YouTube Music’s rollout therefore represents a transparency-first approach, with the possibility of a wider policy debate as AI-generated compositions, voices and fictional performers become more prevalent.