The AI Labeling Act Would Make AI Audio Labels a Federal Rule
A bipartisan Senate bill would require visible and machine-readable labels on AI-generated audio, video, and images. It has not passed, but it points in the same direction as the DSP disclosure fields artists already need to track.
Short answer
On June 25, 2026, Senators Brian Schatz, John Curtis, and Mark Warner introduced the AI Labeling Act of 2026. The bill would require providers of generative AI systems to label covered AI-generated content and embed machine-readable provenance data. Large online platforms would have to identify labeled AI content and avoid stripping the disclosure. The bill covers audio, video, and images, but it is not law yet and it does not create a new music royalty. For independent artists, the practical move is to document AI use in recordings, artwork, videos, ads, and promotional assets so metadata and platform disclosures can be handled cleanly if the legal and DSP rules tighten.
The AI Labeling Act is not law yet. If it passed, it would require labels and provenance data on covered AI-generated audio, video, and images. For artists, the useful move now is to document AI use before the platform or legal paperwork asks for it.
Key takeaways
- On June 25, 2026, Senators Brian Schatz, John Curtis, and Mark Warner introduced the AI Labeling Act of 2026.
- The bill would require clear labels and machine-readable disclosures for covered AI-generated audio, video, and images.
- Large online platforms would have to identify labeled AI content and avoid removing the disclosure.
- It does not create a new music royalty, and it has not passed. Treat it as a signal that AI disclosure is becoming infrastructure, not a side note.
What happened?
A bipartisan group of US senators introduced the AI Labeling Act of 2026 on June 25. The bill would require providers of generative AI systems to add a clear disclosure when they create covered AI-generated content. It would also require a machine-readable disclosure that identifies the content, the system and version used, and the date and time the content was created or modified.
The bill applies to covered AI-generated audio, video, and images. It also creates duties for covered online platforms, including a requirement to identify labeled AI content and not tamper with or remove the disclosure when the content is shared.
Why this matters for music
The bill is not music-specific, but audio is in scope. That matters because music platforms are already building their own disclosure systems. Apple Music has Transparency Tags. Spotify has tested AI credits. Tidal now labels wholly AI-generated tracks and says those tracks will not earn royalties on Tidal. Deezer says it tags AI tracks and removes them from recommendations.
| What it would do | What it would not do | |
|---|---|---|
| Disclosure | Require visible and machine-readable labels for covered AI-generated audio | Set one music-platform metadata standard by itself |
| Rights | Make provenance harder to strip when content moves online | Create a new royalty for human artists or songwriters |
| Status | Give the FTC and a standards process a role if passed | Change your current release-delivery rules today |
The direction is clear: if AI touched the recording, artwork, video, or campaign asset, you need a record of what happened.
What independent artists should do now
Keep an AI-use log for every release
Track whether AI was used for the sound recording, composition, artwork, music video, lyric video, ad creative, voice, cover image, or social clips. Write down the tool, the version if you know it, the date, who approved it, and whether the final asset contains a material AI-generated element.
Separate help from authorship
A spellcheck pass, caption cleanup, or brainstorming prompt is different from a synthetic vocal, AI-generated instrumental, or AI-made album cover. Platform forms may not draw that line perfectly, so keep your internal notes specific.
This is also a distributor question. If your distributor asks for AI metadata at upload, answer from records, not memory. If it does not ask yet, keep the notes anyway. Retroactive metadata cleanup is much harder once the release is already live.
What is still unclear?
This is only a bill
The AI Labeling Act has not passed, and the final technical standard could change. The bill also does not say exactly how a music distributor, DSP, short-form platform, or artist upload form would map the federal disclosure into release metadata. For now, the useful artist action is preparation: document AI use, keep clean rights records, and avoid unauthorized voice or likeness use.
Sources
- U.S. Senator Brian Schatz: Schatz, Curtis, Warner Introduce Bipartisan Legislation To Provide More Transparency On AI-Generated Content
- U.S. Senator Brian Schatz: AI Labeling Act of 2026 bill text
- Music Business Worldwide: US senators revive bill that would force AI-generated audio, video and images to carry labels
Related Velveteen guides
Get music industry updates without the noise
Short notes on platform changes, royalty issues, and release marketing moves that actually affect independent artists.
Was this useful? Send a signal or flag a correction.