The Hamecohming Framework — Institutional Design for Source Transparency and Non-Training Consent in the Generative AI Society

Abstract:
This paper proposes the Hamecohming / Umecohming Framework—a dual-layer institutional and technical design for ensuring source transparency and non-training consent in generative AI. The institutional layer (Hamecohming) aligns copyright, non-training rights, and fair distribution through a public trust fund. The technical layer (Umecohming) embeds traceable metadata and non-training tags into AI outputs, allowing verifiable provenance without reducing model performance. Together, they establish a foundation for accountable, energy-efficient, and ethically synchronized AI ecosystems.