AI Associate Ethical AI and Data Privacy Practice Test 1 — 15 Questions

Question 1

A sales operations manager notices that the AI-driven lead scoring model assigns lower scores to leads from a particular region, even though those leads historically convert at a higher rate. Which Salesforce Trusted AI principle is most directly violated?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Accuracy. The AI-driven lead scoring model is producing outputs that do not match the ground truth (historical conversion rates), which is a direct failure of the Accuracy principle. Accuracy requires that AI systems perform as intended and produce reliable, correct predictions; here, the model's scores are systematically wrong for a specific region, violating that requirement.

Answer

Empathy

Answer

Safety

Answer

Transparency

Question 2

A company deploys an AI-powered email composer that drafts responses to customer inquiries. To comply with GDPR, which control should they implement regarding automated decisions?

Accepted Answer

Require human review before any AI-generated email is sent. Under GDPR, Article 22 grants individuals the right not to be subject to a decision based solely on automated processing, including profiling, which produces legal effects or similarly significant effects. For an AI email composer that drafts responses to customer inquiries, requiring human review before any AI-generated email is sent ensures that the final decision to communicate is not fully automated, thereby complying with GDPR's requirement for meaningful human intervention in automated decision-making.

Answer

Allow the AI to send emails automatically if confidence is high

Answer

Disable the AI composer entirely to avoid GDPR risk

Answer

Anonymize customer data before drafting

Question 3

What is the purpose of ‘toxicity detection’ in the Einstein Trust Layer?

Accepted Answer

To detect and prevent harmful or abusive language in AI outputs. Toxicity detection in the Einstein Trust Layer is designed to identify and filter harmful or abusive language in AI-generated outputs before they reach the user. It uses natural language processing (NLP) models to score content against categories such as hate speech, profanity, and harassment, ensuring outputs remain safe and appropriate. This prevents the AI from inadvertently disseminating offensive material, which is critical for maintaining trust and compliance in enterprise environments.

Answer

To increase the speed of AI responses

Answer

To monitor user interactions for compliance with data protection laws

Answer

To identify and mask PII in user prompts

Question 4

A data scientist is building a churn prediction model for a subscription service. The dataset includes highly correlated features: ‘number of support tickets’ and ‘average response time’. Which action is BEST to ensure model accuracy and interpretability?

Accepted Answer

Remove one of the correlated features. Removing one of the highly correlated features (D) is the best action because multicollinearity between 'number of support tickets' and 'average response time' can inflate the variance of coefficient estimates, making the model unstable and harder to interpret. By dropping one feature, you reduce redundancy without significant information loss, preserving both accuracy and interpretability in a linear or tree-based model.

Answer

Keep both features to maximize information

Answer

Apply principal component analysis to combine them

Answer

Increase the regularization strength

Question 5

When using Einstein Copilot to generate email content, what mechanism ensures that the AI does not use customer data to improve the underlying large language model?

Accepted Answer

Zero data retention. Einstein Copilot employs a zero data retention policy specifically for the underlying large language model (LLM). This means that any customer data processed during email generation is not stored, logged, or used for model training or fine-tuning, ensuring compliance with data privacy standards. The mechanism explicitly prevents the LLM from learning from or being improved by customer interactions, isolating the AI's behavior from proprietary data.

Answer

PII masking

Answer

Grounding

Answer

Toxicity detection

Question 6

A healthcare provider is using Einstein to predict patient readmission risks. They must ensure the model is both accurate and fair. Which THREE actions should they take? (Choose 3)

Accepted Answer

Validate model accuracy separately for different patient subgroups. Validating model accuracy separately for different patient subgroups is a key fairness practice. It ensures the model performs consistently across diverse populations, preventing hidden disparities that could lead to biased clinical decisions. This aligns with responsible AI principles that require subgroup performance analysis to detect and mitigate algorithmic bias.

Answer

Use only historical data from the past six months

Answer

Include as many features as possible to maximize predictive power

Question 7

A sales rep receives an AI-generated lead score of 95, but the rep notices the lead's email domain is 'example.com' and the phone number is invalid. The rep suspects the AI model is overvaluing certain features. Which TWO actions should the rep take to investigate and address the issue?

Accepted Answer

Report the suspicious lead and request a model audit for bias. Reporting suspicious leads and requesting a model audit is the proper governance procedure when an AI model produces outputs that contradict observable data. This triggers a review of the model's feature weights and training data to identify bias or overvaluation of specific fields, ensuring the model remains reliable and compliant with ethical AI standards.

Answer

Delete the lead from the CRM to avoid influencing future predictions

Answer

Manually override the score to 50 and move on

Answer

Contact Salesforce support to disable the AI model

Question 8

Which of the following is a key feature of Salesforce Einstein Trust Layer that protects customer data when using AI?

Accepted Answer

Zero data retention. The Salesforce Einstein Trust Layer is designed with a zero data retention policy, meaning that once an AI model processes a request and returns a response, Salesforce does not store the customer's data. This ensures that sensitive information is not retained on Salesforce servers, protecting customer privacy and complying with data protection regulations. The zero data retention feature is a core architectural principle that prevents data leakage from AI interactions.

Answer

Automatic model retraining

Answer

Data masking for all PII

Answer

Open-source model deployment

Question 9

A company uses Einstein Discovery to predict customer churn. They want to ensure the predictions are explainable to non-technical stakeholders. What is the best way to provide explanation?

Accepted Answer

Use the score factors feature to show top predictors. Einstein Discovery's score factors feature explicitly lists the top predictors and their contributions to each prediction, making the model's reasoning transparent and accessible to non-technical stakeholders. This aligns with the need for explainable AI (XAI) without requiring deep data science knowledge.

Answer

Share the full training dataset for transparency

Answer

Allow stakeholders to retrain the model themselves

Answer

Provide the raw model weights and coefficients

Question 10

A lead scoring model trained on historical sales data is found to assign lower scores to leads from certain postal codes. What is the MOST likely cause?

Accepted Answer

The training data contains biased outcomes from past human decisions. The model assigns lower scores to leads from certain postal codes because the training data reflects historical human biases, such as sales representatives prioritizing leads from affluent areas. Machine learning models learn patterns from the data they are trained on; if past sales decisions were biased against certain regions, the model will replicate those biases. This is a classic case of bias in training data leading to biased model outcomes, not an inherent flaw in the algorithm itself.

Answer

The model's algorithm is inherently biased against certain regions

Answer

The model overfits to the training data

Answer

The model was not trained long enough

Question 11

A company deploying Einstein Bots for customer service wants to ensure compliance with GDPR's right to explanation. Which TWO measures should they implement?

Accepted Answer

Implement an audit trail that logs which AI decisions were made and why. GDPR's right to explanation requires organizations to provide meaningful information about the logic involved in automated decision-making. An audit trail that logs which AI decisions were made and why directly satisfies this by recording the decision path, input features, and model version used, enabling retrospective explanation. Option C is correct because score factors (e.g., feature importance weights in Einstein Bot's predictive models) allow the bot to show why a specific action was recommended, fulfilling the transparency requirement under Article 22 of GDPR.

Answer

Store all customer chat transcripts permanently

Answer

Use only rule-based bots to avoid AI complications

Answer

Allow customers to opt out of all AI interactions

Question 12

Which Salesforce feature allows administrators to mask personally identifiable information (PII) in prompts sent to large language models?

Accepted Answer

Einstein Trust Layer. The Einstein Trust Layer is a security architecture within Salesforce that intercepts prompts sent to large language models (LLMs) and automatically masks personally identifiable information (PII) before the data reaches the external AI service. After the LLM returns a response, the Trust Layer reinserts the masked data, ensuring that sensitive information never leaves the Salesforce ecosystem in plain text. This feature is specifically designed to address data privacy and compliance requirements when using generative AI features like Einstein GPT.

Answer

Sharing Rules

Answer

Permission Sets

Answer

Data Export Service

Question 13

A marketing manager wants to use AI to generate personalized email content for customers. According to Salesforce's Trusted AI principles, what should the manager ensure before sending?

Accepted Answer

A human reviews the AI-generated content before sending. Salesforce's Trusted AI principles emphasize human oversight to ensure AI-generated content is accurate, appropriate, and aligned with brand values. The manager must have a human review the content before sending to mitigate risks like bias, inaccuracy, or inappropriate messaging, which aligns with the principle of accountability in AI deployment.

Answer

The emails are sent automatically to maximize efficiency

Answer

The emails are sent only to customers who opted in to AI interactions

Answer

The AI model is retrained daily on new customer data

Question 14

What is the purpose of grounding in the Einstein Trust Layer?

Accepted Answer

To connect the AI to relevant CRM records so responses are accurate and context-aware. Grounding in the Einstein Trust Layer connects the AI model to relevant CRM records (e.g., Accounts, Opportunities, Cases) so that generated responses are based on accurate, up-to-date customer data rather than the model's general training data. This ensures context-aware and factual outputs while maintaining data privacy by not exposing raw records to the model.

Answer

To limit the AI's responses to pre-approved templates

Answer

To encrypt all data sent to the AI model

Answer

To ensure the AI only uses data from Salesforce, not third-party sources

Question 15

A company is concerned about the data minimization principle when using AI to predict customer lifetime value. Which approach aligns with this principle?

Accepted Answer

Use only fields directly relevant to purchase history and engagement, excluding PII like social security numbers. Aligns with the data minimization principle by restricting data collection to only fields directly relevant to the prediction task (purchase history and engagement) and explicitly excluding personally identifiable information (PII) like social security numbers. This reduces privacy risk and complies with regulations such as GDPR and CCPA, which require that data collected be adequate, relevant, and limited to what is necessary for the processing purpose.

Answer

Include all available fields to maximize model accuracy

Answer

Obtain explicit consent for every field used

Answer

Anonymize all data before training, including purchase amounts