An Agentforce agent is designed to recommend next steps but NOT execute them automatically. What is this an example of?

The correct answer is C. Advisory (copilot-style) agent. An advisory or copilot-style agent surfaces insights, suggestions, and recommended actions to a human user who then decides whether to act. This is a deliberate design choice that keeps a human in the decision loop-appropriate for high-stakes scenarios where autonomous execution

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

Question

Options

AFully autonomous agent
BRead-only agent
CAdvisory (copilot-style) agent
DBackground system agent

How the community answered

(62 responses)

A
5% (3)
B
2% (1)
C
87% (54)
D
6% (4)

Explanation

An advisory or copilot-style agent surfaces insights, suggestions, and recommended actions to a human user who then decides whether to act. This is a deliberate design choice that keeps a human in the decision loop-appropriate for high-stakes scenarios where autonomous execution would carry too much risk. A fully autonomous agent (A) acts without human confirmation. A read-only agent (B) only retrieves and displays data without making recommendations. A background system agent (D) operates without any user-facing interaction. The copilot pattern is common in early-stage deployments where organizations want AI assistance without full AI autonomy.

Topics

#AI Agents#Advisory Agents#Copilot#Next Best Action

Community Discussion

No community discussion yet for this question.

Full AI-201 Practice