What is human in the loop (HITL)?
Human in the loop (HITL) is an approach where people review, approve or correct an AI system's work at key points, for example approving a refund an AI agent proposes or reviewing low-confidence answers, combining AI speed with human judgment and accountability.
Patterns
| Pattern | Example |
|---|---|
| Approval | Agent requests sign-off before acting |
| Review | Sample of answers checked |
| Escalation | Low confidence → human |
| Feedback | Corrections improve the system |
When to use HITL
High-risk actions, regulated decisions, new deployments and low-confidence cases.
Human on the loop
Humans monitor and can intervene rather than approving every step.
