What is Human in the Loop (HITL)?

A design in which a person reviews or approves an AI system's output before it takes effect, rather than the system acting unsupervised. The useful version is selective: approval is placed at the steps that are irreversible or expensive to undo, not at every step, because a system that asks permission for everything is quickly ignored. Placing the checkpoints well is a design decision, not a safety afterthought.