Human in the loop (HITL)

What human in the loop (HITL) means for AI agents, how approvals fit in, and where people review an agent's work.

A way of working where people review or approve an agent’s work at set points, such as before a change ships or a message goes to a customer. A human handoff is one moment in that loop.

The agent does the work it can, and a person makes the calls that carry risk. People set the goal, grant access, and review the steps that matter. The most common form is an approval: a person reviews what an agent wants to do and allows it, changes it, or stops it.

The loop works best where the work already is. When the request, the agent's work, and the approval sit in one thread, the reviewer sees the whole story and the next person starts from it. A human handoff is the agent asking for that review.

Frequently asked questions

Is human in the loop the same as an approval?

An approval is the most common form. Human in the loop is the wider practice of people reviewing or approving an agent’s work at set points.