YouTube Music adds new labels for AI-generated content
YouTube Music has introduced visible labels to identify tracks in which AI played a significant role, including songs generated entirely by AI or those using AI for major elements such as vocals. The labels are intended to help listeners understand how music was made as AI-generated audio becomes more common.
An “AI” label appears below the track name, and tapping it opens details about how AI was used. YouTube says similar labels will come to its video player, which also hosts music videos. Other services are taking steps too: Spotify has started labelling fake artists, while Deezer is introducing a system to filter out AI music.
- YouTube Music now marks tracks where AI played a significant role.
- Tapping the label gives details on how AI was used.
- Similar labels are planned for YouTube’s video player.
New here? Start with this
YouTube Music is part of YouTube, Google’s service for listening to songs and albums. Music can be made with artificial intelligence (AI), which can generate audio or help create parts such as a singer’s voice.
The labels are intended to give listeners information about how a track was made. YouTube is also planning similar labels for its video player, where music videos appear.
Music services are taking different approaches as AI-made tracks become more common. Spotify has begun labelling fake artists, while Deezer is developing a way to filter out AI music.
Both sides, in good faith
The strongest fair case each way — we don't pick a winner.
The case for
Clear labels can help listeners make informed choices and understand how a track was produced, especially when AI substantially shaped its sound or vocals. As AI music becomes more common, consistent disclosure may also build trust and make it easier to distinguish human performance from generated material.
The case against
Labels may imply a sharper divide between AI-made and human-made music than many creative processes warrant, since artists can use AI in limited or collaborative ways. If the criteria are unclear or applied inconsistently, the labels could confuse listeners or stigmatise experimentation without giving them useful context.