Human-over-the-Loop

Human-over-the-loop is a governance model for lower-risk AI systems in which humans define policies and performance thresholds, monitor outcomes through periodic reviews, and intervene when the system operates outside those established parameters.