Which action should an admin take if an Agentforce agent consistently fails on edge cases?

The correct answer is C. Add fallback and escalation paths. Edge cases are scenarios outside the agent's confident operating range. The correct engineering response is graceful degradation: configure the agent to recognize when it cannot handle a situation and escalate to a human agent or trigger a fallback action (such as creating a case

AI Features for Service (e.g., Einstein Bots, Next Best Action)

Question

Options

AIncrease autonomy
BRemove human oversight
CAdd fallback and escalation paths
DDisable logging

How the community answered

(51 responses)

A
4% (2)
B
2% (1)
C
82% (42)
D
12% (6)

Explanation

Edge cases are scenarios outside the agent's confident operating range. The correct engineering response is graceful degradation: configure the agent to recognize when it cannot handle a situation and escalate to a human agent or trigger a fallback action (such as creating a case). This preserves trust and ensures customers are still served. Increasing autonomy (A) would make failures more frequent and severe. Removing human oversight (B) is the opposite of what's needed-it eliminates the safety net. Disabling logging (D) would make the problem invisible and harder to diagnose. Escalation paths are the industry-standard pattern for handling LLM uncertainty.

Topics

#AI agent configuration#Fallback strategies#Escalation paths#Einstein Bots

Community Discussion

No community discussion yet for this question.

Full AI-201 Practice