Judgment Kernel is a structural architecture for accountable decision-making in AI systems.
A Judgment Kernel is a first-class architectural component that authorizes decisions under explicit normative constraints and binds responsibility at the moment of judgment.
Inference–Judgment Separation
Inference produces recommendations. Judgment authorizes decisions. Decision executes outcomes.
Governance Primitive → Judgment Kernel → Judgment Layer → Responsibility Anchor → Accountability Drift
Nakayama, S. (2026). Structural Governance Terminology Definitions.
This repository serves as the canonical terminology source for structural AI governance architecture.
Active research framework. Terminology definitions are stable.