Agents & Agentic AI
Human-on-the-loop
HOTL
A setup where AI acts independently while a person supervises and can intervene.
Example
A fraud-monitoring agent runs independently while a person watches alerts and can intervene.
Why it matters
It allows routine actions to proceed while keeping a person available for exceptions.
You'll hear it
“The system is running, but the supervisor can pause 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
- Is supervision the same as approving every action first?
- No. Actions can happen before the person reviews or intervenes.
- Can intervention come too late?
- Yes. Fast or irreversible actions may finish before a supervisor can respond.
- Can too many alerts make supervision worse?
- Yes. Repeated low-value warnings can distract from the events that need attention.