AI-generated music: creative assistant or threat to authors?

Models that draft melodies and voices force a new argument about authorship, labour and listening.

Generative AI can now propose chord loops, lyrics, vocal clones and full “songs” from text prompts in seconds. Labels, publishers and unions are already fighting over training data, likeness rights and credit. Fans meet the tech as novelty tracks, playlist filler and tools inside DAWs. The cultural question is sharper than the demo: help for makers, or replacement of makers?

The assistant camp says every era adopted machines — drum machines, samplers, Autotune, laptops. AI can speed sketching, help amateurs past blank-page fear, and free professionals from tedious MIDI chores. Used under human direction, it is another plugin. History, they argue, rewards people who absorb new tools without abandoning taste.

The threat camp says scale and opacity are different this time. Models trained on copyrighted catalogues without clear consent can flood platforms with cheap soundalikes, undercut session work, and blur who deserves royalties. Voice clones raise consent issues beyond melody. For this side, “assistant” is a soft word for labour displacement.

Law lags culture: some countries draft AI rules, platforms label synthetic media unevenly, listeners often cannot tell. Artists split between early adopters and vocal refusals. The debate will not wait for perfect regulation.

Do you see generative AI in music as a creative assistant, a threat to authors, or something in between you can still name? Draw your line below — and say what would change your mind.