Agents & Agentic AI
Human-in-the-loop
HITL
A workflow where a person reviews, approves, corrects, or guides AI decisions.
Example
A recruiter reviews and approves an AI-generated candidate shortlist before anyone is contacted.
Why it matters
It adds human judgment where automated choices need correction or approval.
You'll hear it
“Hold this result until someone has checked it.”
Can you control it?
Set by the team
The people building or running the AI choose this setup. An everyday user generally needs their help to change how this part works.
Common questions
- Does adding a person guarantee mistakes are caught?
- No. A rushed or poorly informed reviewer may simply accept the suggestion.
- Must the person inspect every step?
- Not always. The process can reserve review for particular decisions or exceptions.
- Can unclear responsibility weaken the review?
- Yes. If nobody knows who must respond, work can pass unchecked or remain stuck.