Spotify's new tool doesn't distinguish deepfakes from metadata errors – the artist must assess it themselves

The launch of Artist Profile Protection comes eight months after the beta rollout, at a time when Sony Music has, according to Billboard, requested the removal of more than 260,000 deepfake songs.

File photo: File photo: Spotify huvudkontor i Stockholm
File photo: Goombah, CC0 1.0 · source

Spotify's new tool doesn't distinguish deepfakes from metadata errors – the artist must assess it themselves

The launch of Artist Profile Protection comes eight months after the beta rollout, at a time when Sony Music has, according to Billboard, requested the removal of more than 260,000 deepfake songs. The tool does not give artists AI detection — but control over what is attached to their name.

Spotify is now rolling out its "Artist Profile Protection" tool to all artists. The official launch was reported on 17 November 2026 by TechCrunch, and comes eight months after the company began beta testing with a select group of artists. The tool is not a detection technology for AI-generated music — it is an attribution tool: artists receive alerts when new music is submitted under their name, and can themselves decide whether the release is linked to their profile or not.

What the tool actually does

The feature is activated via Spotify for Artists. When new music is submitted under an artist's name, the artist receives an alert. If the release is not the artist's, they can remove the association, so that the release no longer appears on their profile.

The default behavior is deliberately low-friction: releases go live as planned unless the artist intervenes. For those who want tighter control, there is an advanced setting that requires the artist to manually approve or reject releases before they appear on the profile. That means artists who release a lot of music and don't want to risk delays can use the basic alerts feature, while those particularly vulnerable to impersonation can turn on full manual approval.

Why Spotify is launching this now

Spotify itself describes the tool as a response to a long-running problem across streaming services: music ends up on the wrong artist profile because of deliberate impersonation, metadata errors, or shared artist names — the last being a well-known phenomenon in which many different artists share the same name and thus the same profile. The company says it worked with thousands of artists during beta testing to make it easier to protect the profile without disrupting release schedules.

The timing coincides with a marked increase in AI-generated imitations on streaming platforms. Billboard reported in October 2026, as relayed by TechCrunch, that Sony Music has requested the removal of more than 260,000 deepfake songs from digital platforms — nearly double the number the label reported in March of the same year. The figures are cited from Billboard and have not been independently verified by AIMag.

It is worth being precise about what the figures describe: these are takedown requests from a single record label, not a total measure of AI-generated music on Spotify or other platforms. But the direction is unambiguous, and it explains why a platform like Spotify is now prioritizing attribution control.

Not detection — but control

The most important limitation is what Artist Profile Protection does not do. The tool does not distinguish between a deepfake and an ordinary metadata error. In both cases the mechanism is the same: music has been submitted under an artist's name, and it is the artist themselves — not an algorithm — who decides whether the release belongs on the profile. The sources describe no form of automatic identification of synthetic content.

That is an advantage in practice: an artist can remove both an AI-generated imitation and a mislabeled release with the same action, without the platform having to determine which is which. But it also means the burden of detection falls on the artist. Alerts only help if the artist sees them; manual approval only helps if the artist actually engages with it. For artists without active access to Spotify for Artists — small artists, amateurs, artists who share a name with others without knowing it — the tool provides no automatic protection.

What remains unclear

Several questions are not answered in the available source material. There is no primary documentation from Spotify — neither a press release, a product page, nor its own announcement in Spotify for Artists — underpinning this story; all details about the feature rest on TechCrunch's reporting and Spotify's claims as relayed there. The beta start date is given only in relative terms ("eight months"), with no exact date. How the alerts work in detail — how quickly they arrive, what happens if an artist does not respond before a release goes live — is not described. Nor are artists' own experiences from the beta period documented beyond Spotify's own description of the collaboration.

The big unanswered question is nonetheless a different one: where the deepfake music is coming from, and whether attribution control is enough. A tool that removes a deepfake from a real artist's profile does not prevent the song from remaining available on the platform under a different name. Whether Spotify — or the industry as a whole — will go further than attribution and more aggressively filter AI-generated imitations at submission remains unanswered. The Billboard figure Sony cannot yet claim a solution for suggests, at least, that the pressure is not letting up.

Sources: TechCrunch, 17 November 2026 [2b382730]

AIMag.no
AIMag.no
The AIMag.no editorial team covers artificial intelligence, tools, research, and regulation.

Get the best of AI MAG in your inbox

News, analysis, and ideas at the intersection of AI and society.