Good Job, Spotify. Now Check the Music Too, Like Beatport Just Did.
Spotify checked who. Beatport checked what. Who wins? Illustration by Dhyana Rao-Middleton
Written by Dhyana Rao-Middleton
I was ready to be done talking about AI for a while. The plan, genuinely, was to spend this space on the thing that actually warms my heart, intentional listening and real music community, the reasons I built Harmonize in the first place. Then two days ago Spotify announced something I could not let go by without addressing, so this is the last stop on this particular road for now, and I want to give it a proper look before I move on.
What Spotify actually announced
On August 11, Spotify introduced what it is calling an AI Persona badge. It marks artist profiles that represent an AI generated identity rather than a real person, and it will show up on the artist's profile, in search results, and on track rows. There are two ways a profile gets flagged. An artist can self disclose through Spotify for Artists, or Spotify's own team can apply the badge after reviewing profiles where the name and imagery appear to be a photorealistic AI generated identity, starting with the artists who already have a large audience. Tap the badge and you can see which of the two happened. Artists who think they were flagged wrongly can appeal, and Spotify says listeners will eventually be able to report suspected AI personas themselves. Once a profile carries the badge, it is excluded from personalized recommendations and editorial playlists by default, unless a listener has actually chosen to follow that artist. It rolls out in mid September.
Where it helps, and where it stops short
Credit where it is due, this is a real response to a real problem, not a symbolic gesture. Fabricated AI artists with fabricated backstories have already gone properly viral, racking up millions of monthly listeners with nobody quite able to say who, or what, was behind the voice. A badge that surfaces that right on the profile, and pulls undisclosed personas out of algorithmic and editorial recommendation by default, is a meaningfully better, structural outcome, not just a label sitting on a page nobody reads.
But here is the sentence in Spotify's own announcement I keep coming back to: the badge is about the artist's public identity, not about how the music was made. That means a persona can be labeled, disclosed, and still live fully on the platform, earning streams, without anyone ever checking whether the music underneath it was made from stolen assets or from something genuinely new. Disclosure is not the same as verification. Spotify can now say, correctly, this profile is an AI persona, while never asking the question that actually determines whether that persona's catalogue should exist at all, did the people whose music trained it agree to that. The question I asked in my last post is still exactly as open as it was before Tuesday, is there any proof that AI music is not theft when it was trained on the work of people who never agreed to it. A label on a profile does not answer that. It just means we finally know who to ask.
What checking the music itself actually looks like
One day after Spotify's announcement, on August 12, Beatport announced something with a very different shape. Working with Beatdapp, the fraud detection partner it already used, Beatport now scans every file at the point of upload to determine whether the actual audio is human made, AI assisted, or fully synthetic. Tracks that are wholly or substantially AI generated are withheld automatically, before they ever reach the platform, and the rightsholder is notified. Tracks made by a human with AI assistance are still accepted, but tagged for the curation team. Beatport's CEO, Matt Gralen, put the distinction the way I would put it myself, there is a difference between a tool that assists human creation and a system that replaces it entirely.
That is the missing half of Spotify's announcement. Proof that checking the content itself, not just the identity behind it, is not some far off technical fantasy. It is happening, one day apart, on a different platform. Spotify chose to verify who someone claims to be. Beatport chose to verify what was actually made. Both are real steps. Only one of them checks the thing I think most people actually care about. Great job Beatport!
Where I am taking this next
So, this is where I leave the AI thread for now. Not because the questions are answered, they are not, but because I started Harmonize for a different reason than defending against AI, and I want this blog to spend more time there. I want to write about intentional listening, about the small, deliberate ways people are already choosing to find music on purpose again instead of letting a feed choose for them. I want to write about what real community around music actually looks like, because that is the thing I actually get out of bed excited about. The AI conversation is not going away, and I am sure I will come back to it when there is something worth saying. But for now, consider this the last word on the topic here for a while, and consider the next post an invitation into the part of this I actually built Harmonize to protect.
