Hank Green identified a significant gap in YouTube’s disclosure rules for artificial intelligence. The platform mandates labels only when creators use AI to meaningfully alter or generate photorealistic content. This policy creates an arbitrary boundary where AI-generated music requires a warning but a photorealistic unicorn does not. The official summary states that realistic content needs disclosure while non-realistic or minor edits do not. However, the rules permit undisclosed use of AI for idea generation, production assistance, and creating video outlines or scripts. Creators can also clone their own voices without warning and use AI to animate missiles in fully animated videos. These specific allowances mean substantial AI involvement often remains invisible to the audience.
This discrepancy matters because it undermines the core intent of transparency. Viewers cannot distinguish between human-made and machine-assisted work when the assistance is significant yet unlabelled. The current framework treats the visual realism of the output as the sole metric for disclosure rather than the degree of artificial intervention. This approach leaves a large section of modern video production unregulated by the platform’s own standards. It also fails to provide clear guidance to creators who operate in the grey areas between realistic and non-realistic styles.
* Voice cloning is explicitly exempt from disclosure requirements
* AI assistance in scripting and outlining is not flagged
* Animated content involving AI tools faces no labeling obligation
