Salesforce AI Associate AI Associate practice test

A Salesforce admin implements Einstein Bots for customer service. To ensure the bot does not use biased language, what should the admin do?

Trap 1: Use only predefined responses

Predefined responses may still be biased if not carefully designed.

Trap 2: Disable the bot for sensitive topics

Avoidance is not a comprehensive solution; addressing bias is better.

Trap 3: Monitor conversations regularly

Monitoring detects issues but does not prevent bias in the bot initially.

A
Use only predefined responses
Why wrong: Predefined responses may still be biased if not carefully designed.
B
Disable the bot for sensitive topics
Why wrong: Avoidance is not a comprehensive solution; addressing bias is better.
C
Review training data for representativeness
Correct. Diverse and representative training data reduces bias.
D
Monitor conversations regularly
Why wrong: Monitoring detects issues but does not prevent bias in the bot initially.

Question 2hardmultiple choice

A data architect is designing a data model for Einstein Discovery. The data includes categorical variables with high cardinality (e.g., postal codes). What is the best practice to handle such features?

Trap 1: Encode them as one-hot vectors.

One-hot encoding with high cardinality creates too many features.

Trap 2: Exclude them from the model.

Excluding may lose valuable information.

Trap 3: Use the raw values without transformation.

Raw values with many levels can cause overfitting.

A
Encode them as one-hot vectors.
Why wrong: One-hot encoding with high cardinality creates too many features.
B
Exclude them from the model.
Why wrong: Excluding may lose valuable information.
C
Use the raw values without transformation.
Why wrong: Raw values with many levels can cause overfitting.
D
Group them into higher-level categories (e.g., region).
Reduces cardinality while preserving signal.

Question 3mediummulti select

A data analyst is evaluating data quality for an Einstein model. Which TWO dimensions are most critical for model accuracy?

Trap 1: Uniqueness

Important for deduplication but not as critical for model accuracy.

Trap 2: Consistency

Important but less critical than completeness and accuracy.

Trap 3: Timeliness

Timeliness matters for relevance, but not directly for accuracy.

A
Uniqueness
Why wrong: Important for deduplication but not as critical for model accuracy.
B
Accuracy
Incorrect values directly degrade model predictions.
C
Consistency
Why wrong: Important but less critical than completeness and accuracy.
D
Completeness
Missing data can bias the model.
E
Timeliness
Why wrong: Timeliness matters for relevance, but not directly for accuracy.

Question 4mediummulti select

Which TWO actions are required to prepare data for an Einstein Discovery model?

Trap 1: Remove all records with missing values in any field.

Einstein Discovery can handle missing values.

Trap 2: Select exactly 10 predictor fields manually.

Einstein can automatically select features.

Trap 3: Create a separate dataset for training and validation.

Einstein automatically splits data.

A
Remove all records with missing values in any field.
Why wrong: Einstein Discovery can handle missing values.
B
Select exactly 10 predictor fields manually.
Why wrong: Einstein can automatically select features.
C
Ensure the data is stored in a Salesforce object or a connected data source.
Einstein Discovery requires data accessible via Salesforce.
D
Create a separate dataset for training and validation.
Why wrong: Einstein automatically splits data.
E
Define the outcome field that the model will predict.
The outcome is essential for supervised learning.

Question 5easymultiple choice

A sales manager wants to automatically prioritize leads based on their likelihood to convert. Which Einstein feature should the admin enable?

Trap 1: Einstein Activity Capture

Activity Capture logs emails and meetings but does not score leads.

Trap 2: Einstein Prediction Builder

Prediction Builder creates custom models but requires more setup and is not the standard lead scoring tool.

Trap 3: Einstein Bot

Einstein Bot is for conversational AI, not lead scoring.

A
Einstein Lead Scoring
Einstein Lead Scoring directly scores leads based on conversion likelihood.
B
Einstein Activity Capture
Why wrong: Activity Capture logs emails and meetings but does not score leads.
C
Einstein Prediction Builder
Why wrong: Prediction Builder creates custom models but requires more setup and is not the standard lead scoring tool.
D
Einstein Bot
Why wrong: Einstein Bot is for conversational AI, not lead scoring.

Question 6easymultiple choice

A marketing team wants to use Einstein Engagement Scoring to prioritize leads. What is the primary input for this AI feature?

Trap 1: Historical conversion data from closed opportunities.

That is used for lead scoring, not engagement.

Trap 2: Lead demographic information like industry and company size.

Demographics are not the primary input.

Trap 3: Social media posts and mentions of the company.

Social media is not part of Einstein Engagement Scoring.

A
Lead interaction history with emails and web activity.
Engagement is measured by interactions.
B
Historical conversion data from closed opportunities.
Why wrong: That is used for lead scoring, not engagement.
C
Lead demographic information like industry and company size.
Why wrong: Demographics are not the primary input.
D
Social media posts and mentions of the company.
Why wrong: Social media is not part of Einstein Engagement Scoring.