Human Oversight
Pengawasan Manusia
A person with the authority and the information to actually stop the system.
Human oversight means a person is genuinely able to intervene in a running system: they see what it is doing, they can stop it, and they understand enough to know when to.
The word is used loosely, and the looseness hides a real distinction. A person who receives a report after a thousand automated rejections has not overseen anything. Oversight needs three things at once: visibility into the current state, the authority to intervene, and the understanding to judge.
This is also where the regulatory requirements around automated decisions land. A system that scores applicants, flags content, or denies a claim generally needs a route for a person to review the outcome and reverse it, not merely a support email address.
Related terms
Human in the Loop
A person who reviews, approves, or corrects the agent's work at key points.
Approval
The moment a person says yes before the agent does something it cannot undo.
Compliance
Meeting the legal rules that apply to what you built, in the place you built it.
Agent
A model with tools, a system prompt, and a loop that lets it act and observe results.
AI Governance
The rules your organisation sets for how AI is used, and who is accountable for them.
Want this working inside your team?
It starts with one conversation about your workflow.