You’re implementing the escalation logic for when the agent should call escalate_to_human. Your team…

Question

You’re implementing the escalation logic for when the agent should call escalate_to_human. Your team proposes four different approaches for triggering escalation. Which approach will most reliably identify cases that genuinely require human intervention?

Accepted Answer

C. Instruct the agent to escalate when the customer requests a human, when the issue requires policy exceptions, or when the agent cannot make meaningful progress.

Answer

A. Implement sentiment analysis that monitors for frustration indicators (negative language, repeated questions, exclamation marks) and trigger escalation when the frustration score exceeds a configured threshold.

Answer

B. Configure the agent to escalate after three consecutive tool calls that fail to resolve the customer’s stated issue, ensuring a reasonable attempt before involving a human.

Answer

D. Build a rules engine that maps specific issue types, customer segments, and product categories to escalation decisions, removing the need for model judgment calls.

Question 16