AI music as the main focus of a Short: YouTube says to disclose it

A Short built around a generated song is a different case from a Short with a quiet bed under speech. YouTube's help page on disclosing altered or synthetic content lists four things that need a disclosure. One of them is content that "creates music that's the main focus of the video".

What the page says

The same page lists what does not need a disclosure: beauty filters, color adjustment, special effects filters, caption creation, and cloning your own voice for voice overs. It also lists production help such as a script or thumbnail, upscaling and caption creation, and says non-realistic content, such as someone riding a unicorn through a fantastical world, is exempt.

Where Sume music fits

Sume's Music 1.0 makes songs from a text prompt and runs on Google Lyria 3.5. Its docs say the request has no duration parameter, so you steer length in the prompt and then check the audio you get back. If the song carries the Short, plan to answer the disclosure question in the upload flow.

How to disclose

In YouTube Studio, go to the upload flow, find the Attributes section and answer Yes under AI use. The page says labels appear in the video player for photorealistic content and in the expanded description for non-photorealistic or animated content, and that disclosing does not limit the video's audience or its eligibility to earn money.

A short checklist

• Decide, before you generate, whether the music is the focus or a bed.
• If it is the focus, answer Yes under AI use at upload.
• Keep your prompt and the generated file together, so you can show how the track was made.
• Re-read the YouTube page when you publish; the list is the platform's to change.