Human Approval
75%
Lesson 6 of 8

Human Approval

Human approval should sit immediately before actions where mistakes would be costly, irreversible or sensitive.

Active reading
Why this matters

Human approval should sit immediately before actions where mistakes would be costly, irreversible or sensitive.

Prepare

Agent drafts the proposed action.

Review

Person sees evidence and impact.

Execute

Only approved actions run.

Understand the working model

Approval is more useful when the reviewer sees exactly what will happen: recipient, amount, record changes, evidence used and any uncertainty. Avoid generic “Approve?” prompts with no context.

Keep preparation separate from execution. The agent may draft an email or refund proposal automatically, but the external side effect happens only after approval.

Example
Agent proposes: refund $120 to order 4821. Evidence: policy section 3 + verified order. Reviewer sees proposal → Approve → payment API executes.
Practice

Apply the pattern above to one real task you already do. Keep the first version small enough that you can inspect every input and output.

Common mistake

Do not ask for approval after the irreversible action already happened.

Key takeaway

Place human approval at the last safe point before high-impact actions execute.

Next lesson

Evaluate an Agent

Continue to the next lesson.

Continue →