Human-in-the-Loop
A design in which an agent waits for approval or hands control to a person at defined points. OpenAI's agent guide names two triggers: exceeding failure thresholds (for example, the agent failing to understand customer intent after several attempts) and sensitive, irreversible or high-stakes actions (cancelling orders, large refunds, payments).
AI and GrowthTerm 70 / 81
In practice
For retention, a human in the loop is a condition of trust before it is a cost line: one agent that cancels a customer's subscription by mistake can wipe out the trust built by many good answers (our reading). OpenAI describes this intervention as the way to find failures and edge cases early and build a solid evaluation cycle; Anthropic also notes that agents can pause for human feedback at checkpoints. It belongs in unit economics too: every task handed to a person adds cost. Report the agent's success rate together with its handoff rate (our reading).