How should Hybrid AI music be classified? Beyond “AI-generated” vs “Human”
The current discussion around AI music often seems to be reduced to a simple question: human or AI-generated.
But hybrid workflows are becoming more common:
- human vocals + AI-generated instruments
- human production + AI-assisted elements
- AI-generated parts combined with traditional recording workflows
Authio recently published AOCS v1.0, a proposed technical framework for discussing these cases.
The idea is to separate technical audio analysis from questions like authorship, ownership or artistic value.
AOCS proposes three categories for eligible Hybrid recordings:
- Human-Dominant Hybrid
- Balanced Hybrid
- AI-Dominant Hybrid
The classification is based on the detected AI contribution within the Full-Mix recording, rather than trying to determine who “created” the work.
From a production perspective:
Do you think this type of distinction is useful?
Would producers prefer a final master classification, or more granular information about where AI elements appear (vocals, instruments, processing, etc.)?
Interested in hearing opinions from people working with audio.