Which Salesforce capability helps ensure AI recommendations follow data access permissions and privacy rules?
A) Object-Level Security
B) Einstein Trust Layer
C) Lightning Experience
D) Field History Tracking
Answer: B
This framework ensures AI uses data safely and follows privacy, compliance, and security rules.
What is the primary purpose of an AI agent in Salesforce?
A) Replace all human users
B) Assist users by generating suggestions based on CRM data
C) Create new Salesforce objects automatically
D) Encrypt Salesforce data
Answer: B
AI agents help users by suggesting next steps or summarizing information from Salesforce.
What does “grounding” an AI agent’s response mean?
A) Restricting responses to only what’s in Salesforce CRM data
B) Lowering the priority of AI responses
C) Caching responses in memory
D) Generating responses without any data context
Answer: A
Grounding ensures AI only references verified Salesforce records, so it doesn’t give incorrect or random answers.
Which of the following is a use case for Agentforce AI in a Service Cloud context?
A) Automatically generating leads
B) Summarizing a case’s recent activities for an agent
C) Importing contacts from Excel
D) Exporting reports to PDF
Answer: B
Agentforce helps by summarizing case notes and interactions.
An admin must control which users can interact with AI agents. Which Salesforce feature is most directly used for this?
A) Permission Sets
B) Validation Rules
C) Page Layouts
D) Workflow Rules
Answer: A
Permission sets let you give AI access to certain users while restricting others.
An AI agent should only use CRM data the user can already access. This requirement relates to:
A) Data Lifecycle Management
B) Sharing Model & Permissions
C) Salesforce Releases
D) AppExchange Packages
Answer: B
AI must respect the same sharing rules and permissions as a human user so it only sees data the user is allowed to see.
Which of these helps protect customer data when AI processes text?
A) Encrypted Custom Metadata
B) Einstein Trust Layer
C) Auto-Number Fields
D) Static Resources
Answer: B
This layer is supposed to make sure AI doesn’t expose sensitive data and follows Salesforce compliance and security standards.
A user reports the AI agent gave incorrect insights because it referenced unrelated data. The admin should first check:
A) Whether grounding is correctly configured
B) The login history of the user
C) The Salesforce release version
D) Chatter moderation settings
Answer: A
Incorrect AI insights usually happen if the AI isn’t “grounded” to your CRM data, so it pulls unrelated info.
Which principle ensures an AI agent only uses data a logged-in user is authorized to view?
A) Data normalization
B) Role hierarchy
C) Trust and permission enforcement
D) Data import wizard
Answer: C
AI follows Salesforce security rules, meaning it only uses data the user has permission to access.
An organization wants AI recommendations to reflect real-time Salesforce records instead of generic responses. What should be used?
A) Static prompts
B) Grounded CRM data
C) External spreadsheets
D) Manual text inputs
Answer: B
Grounded data keeps AI responses accurate and up to date with current Salesforce records, rather than generic answers.
