Apple Music will tag AI-made songs

If you plug in your headphones and stream a new release today, there is a very real chance that the vocals, chords, and bassline were generated entirely by a server rack.

Now, Apple Music wants to make that impossible to miss. In a memo sent to industry partners, the platform announced that it will begin labeling tracks created with AI tools, rolling out a dedicated "Made With AI" tag later this year.

The policy requires record labels and distributors to flag any song where "a material portion of the content" was made using AI. While the backend infrastructure has been in motion since March under Apple's AI Transparency Tagging program, the company has not yet confirmed the exact date when these labels will become visible to regular listeners.

Over a third of uploads are purely synthetic

Apple is making it very clear that content providers must carry the compliance burden. In its partner note, Apple Music argued that distributors and labels are in the best position to know how their tracks were produced and which tags apply.

Yet the scale of the problem is massive. Apple Music vice president Oliver Schusser told Billboard earlier this year that more than a third of all songs uploaded to the platform are "100% AI."

Schusser noted that Apple already operates internal detection software capable of determining whether AI was used and identifying the underlying models. Still, he insisted that uploaders must take direct responsibility for disclosure.

He also delivered a blunt reality check to the industry, saying that every record label in the world is delivering AI tracks to streamers right now, even if they do not know it yet. Schusser warned that labels need to start building their own internal scanning tools.

Industry pressure meets competing playbooks

The move comes after three chaotic years dominated by the rise of prompt-to-song engines like Suno and Udio, which left traditional rights holders swinging between copyright lawsuits, licensing talks, and in-house remixing features.

Platforms have taken very different paths. Spotify previously introduced "AI Personas" tags on profiles that appear to be synthetic artists. Tidal went significantly further by flagging AI songs and outright banning them from earning royalties.

Meanwhile, music industry trade bodies have formed a unified front. Last month, a coalition including the RIAA, IFPI, SAG-AFTRA, and the Recording Academy behind the Grammys banded together to demand a standardized, industry-wide labeling framework for synthetic music.

Why it matters

For everyday listeners, this is primarily about transparency and trust. You might not mind listening to an algorithmically generated lo-fi beat while studying, but you probably care if an emotional acoustic ballad was mass-produced by software without a human artist behind it. Having a clear tag lets audiences decide what is worth their time and emotional investment.

For the music business, Apple Music is drawing a line in the sand. With over a third of uploads already completely synthetic, the economics of streaming are under immense pressure. If platforms cannot separate human musicians from low-effort AI flood tracks, real creators get starved of royalty payouts.

We think Apple's strategy is politically savvy but structurally leaky. By pushing the labeling duty onto labels and distributors, Apple positions itself as an orderly steward without taking on the legal liability of mislabeling tracks. However, defining a "material portion" is notoriously subjective. Unlike Tidal, which took the nuclear route by blocking monetization, Apple is relying mostly on voluntary self-reporting. Without tying labels directly to royalty caps or algorithmic penalties, we suspect bad actors will continue trying their luck until Apple actually enforces consequences.