Human in the loop
Level 1
Deliberately requiring a person to review or approve an AI action.
Term 7 of 7 in AI agents
In plain language
A human in the loop is a checkpoint. The AI proposes, a person confirms, and only then does anything real happen. It is the standard safeguard for anything irreversible.
Think of it like this
A second signature on a large payment.
Why it matters
This is the cheapest, most reliable safety measure available, and the first thing to add when an AI feature touches money, health or reputation.
Next terms in AI agents
- AI agentAn AI system that can take steps and use tools to complete a goal, not just answer.
- Tool useLetting a model call outside software to get facts or take action.
- Function callingThe mechanism that lets a model request a specific piece of software with specific inputs.
- ReasoningA model working through a problem in steps before giving an answer.