DistroKid, the distributor used by vast numbers of independent musicians, has confirmed that songs are being removed from streaming services in response to Universal Music Group's lawsuit against the company — and artists say the takedowns are catching work that has nothing to do with AI.
UMG sued the distributor in September, arguing in its complaint that DistroKid had effectively built a pipeline for mass-produced AI music, and asking the court to remove roughly 2,000 named recordings and close the accounts behind them. The Verge reports that DistroKid confirmed the recent removals are a direct response to UMG's claims.
The issue is how the removals are landing:
- Musician McGwire says six of his songs were pulled with no notification, before or after. Getting any explanation meant a "lengthy circular conversation" with a support bot before he reached a human. - One removed track was a cover of Stevie Wonder's "Isn't She Lovely" for which he says he held a proper license. Another used a beat he bought from a marketplace that had licensed it to others as well. - A third song was taken down over an uncleared sample — of a record released two years after his own track. - Nerdcore rapper Insane Ian posted that an original song was removed with no warning, and that the company would not say why. Rapper King Chase also struggled to reach a person at DistroKid.
The shared complaint is not the policy but the process: no warning, no stated reason, no obvious path to appeal, and a support front door that leads through bots first.
Why it matters: distributors sit between independent musicians and essentially every major streaming service. If automated AI-suspicion takedowns cannot reliably separate human work from machine output, enforcement falls hardest on the people with the least legal leverage — and trust in the platform carrying their careers erodes with every silent deletion.
What to watch: whether DistroKid publishes criteria and an appeals process; how the Delaware case proceeds; and whether other distributors copy the approach.
