Open main menu
Article
Quizzes
Tools
EN
Article
Quizzes
Tools
All quizzes
/
Understanding AI Agents
/
At which p...
At which point is it most appropriate to add a human-in-the-loop checkpoint in an agentic workflow?
Before the agent takes an irreversible or high-stakes action, such as deleting data or sending a payment
After every single tool call, to ensure no mistakes were made
Only at the very start of the task, when the user provides the initial instruction
Human checkpoints are unnecessary if the agent has been thoroughly tested
Submit answers