Australian musicians sound warning note after Nick Cave, Kylie and many more slurped into AI training tool
TL;DR
Australian musicians including Paul Dempsey, Bernard Fanning, Nick Cave, Kylie Minogue, Powderfinger and Jimmy Barnes have reportedly appeared in datasets tied to AI music training. The Atlantic’s dataset search tool shows millions of creative works scraped from the web. Dempsey says he found Something For Kate’s full catalogue and his solo songs. The disputed material sits in Sleeping-DISCO-9M, with 9.7 million YouTube tracks plus Genius lyrics, and LAION-DISCO-12M, with 12.3 million YouTube tracks.
Nauti's Take
For AI music builders, this is not a PR problem. It is a licensing landmine.
Treat training data like bycatch and you build products on burden of proof instead of trust. In music, provenance now matters as much as output quality.
Briefingshow
This case shows how little leverage creators have once training data is scraped at web scale from public platforms. For musicians, the issue is not only licensing revenue but bargaining power: once catalogues are already ingested, any later deal starts to look like cleanup rather than consent.
Sources