AI0-001 AI Security, Ethics and Governance Practice Test 6 — 15 Questions

Question 1

Which TWO are common types of adversarial attacks on AI models?

Accepted Answer

Evasion attack. Evasion attacks (Option C) are a common type of adversarial attack where an attacker crafts malicious input data that is intentionally designed to cause a trained AI model to make incorrect predictions or classifications, often by adding imperceptible perturbations to legitimate inputs. This exploits the model's sensitivity to small changes in feature space, leading to misclassification without altering the model itself.

Answer

Hyperparameter tuning

Answer

Transfer learning

Answer

Backdoor attack

Question 2

Which THREE are key principles of trustworthy AI according to the OECD?

Accepted Answer

Robustness. Robustness is a key principle of trustworthy AI according to the OECD, ensuring that AI systems operate reliably and securely under a wide range of conditions, including handling errors, adversarial inputs, and unexpected scenarios. This principle directly supports the goal of maintaining system integrity and preventing harm, which is fundamental to trustworthiness.

Answer

Profitability

Answer

Scalability

Question 3

Which TWO techniques are specifically designed to protect individual privacy when training AI models?

Accepted Answer

Differential privacy. Differential privacy (C) is a technique that adds calibrated noise to training data or model updates, ensuring that the output of the model does not reveal whether any specific individual's data was included. This provides a formal mathematical guarantee of privacy, quantified by the epsilon parameter, making it a direct privacy-preserving method for AI training.

Answer

Dropout

Answer

Model pruning

Answer

Regularization

Question 4

Refer to the exhibit. Which model is NOT in full compliance with the policy?

Accepted Answer

CreditScorer v2. CreditScorer v2 is not in full compliance because it uses a black-box neural network that cannot provide explainability for its credit decisions, violating the policy's requirement for model interpretability and transparency. The policy mandates that all models must support post-hoc explanation methods such as SHAP or LIME, which CreditScorer v2 lacks due to its opaque architecture.

Answer

ChurnPredict v1

Answer

FraudDetect v4

Answer

LoanApproval v3

Question 5

A large hospital system deploys an AI triage system for emergency rooms. The system uses patient vitals and symptoms to recommend treatment priority. Six months after deployment, complaints arise that the system frequently underestimates the severity of symptoms for patients from certain ethnic backgrounds. A data scientist runs a bias audit and finds that the model's false negative rate is 20% higher for the minority group. The hospital's AI governance board requires immediate corrective action. The data science team has limited resources and cannot retrain the entire model from scratch. They have access to the training data, which is imbalanced. The model is a gradient boosted tree. Which course of action best addresses the bias while minimizing operational impact?

Accepted Answer

Post-process the model's predictions by adjusting thresholds for the minority group. Post-processing by adjusting decision thresholds for the minority group directly compensates for the higher false negative rate without requiring retraining. Since the team has limited resources and cannot retrain the entire gradient boosted tree model, this approach minimizes operational impact while addressing the bias. The threshold adjustment effectively lowers the probability cutoff for the minority group, making the model more sensitive to their symptoms and reducing underestimation of severity.

Answer

Rebalance the training data using SMOTE and retrain the model

Answer

Use adversarial debiasing during training to remove protected attribute correlations

Answer

Replace the model with a simpler logistic regression model to improve interpretability

Question 6

A security analyst is reviewing logs from an AI-powered recommendation system and notices an unusually high number of requests for products from a specific vendor. The analyst suspects data poisoning. Which mitigation strategy should be implemented first?

Accepted Answer

Implement input validation and sanitization for training data. Input validation and sanitization directly prevent malicious or anomalous data from entering the training pipeline, which is the root cause of data poisoning. In an AI-powered recommendation system, poisoned training data can cause the model to learn biased associations, such as favoring a specific vendor. By validating and sanitizing inputs before they are used for training, the attack vector is blocked at the earliest stage, making it the most effective first mitigation step.

Answer

Encrypt all training data at rest

Answer

Deploy an anomaly detection system on model outputs

Answer

Retrain the model with a smaller, curated dataset

Question 7

A healthcare startup deploys an AI model to predict patient readmission rates. An internal audit reveals that the model consistently underestimates readmission risk for non-native English speakers. According to AI ethics principles, what is the most appropriate course of action?

Accepted Answer

Retrain the model with a more representative dataset that includes diverse language backgrounds. It directly addresses the root cause of the bias: the training data lacks sufficient representation from non-native English speakers, leading to systematic underestimation of readmission risk for that group. Retraining with a more representative dataset aligns with the AI ethics principle of fairness by ensuring the model learns patterns across all demographic groups equally, rather than masking the issue with disclaimers or manipulating sample sizes.

Answer

Add a confidence score disclaimer to model outputs

Answer

Reduce the sample size of non-native English speakers to balance the dataset

Answer

Continue using the model as is, since overall accuracy is acceptable

Question 8

During a penetration test, a security engineer discovers that an AI-powered chatbot can be tricked into revealing sensitive customer data by using specially crafted prompts. What type of attack is this, and what is the best mitigation?

Accepted Answer

Prompt injection attack; implement input validation and context sanitization. This is a prompt injection attack, where an attacker crafts inputs that cause the AI model to override its original instructions or constraints, leading to unintended behavior such as revealing sensitive data. The best mitigation is input validation and context sanitization, which filters or neutralizes malicious prompt content before it reaches the model, preventing the injection from succeeding.

Answer

Model inversion attack; apply differential privacy during training

Answer

Data poisoning attack; implement strict access controls

Answer

Membership inference attack; add noise to model outputs

Question 9

An organization deploys an AI system that processes personal data of EU citizens. Which regulatory framework imposes strict requirements on automated decision-making and profiling?

Accepted Answer

General Data Protection Regulation (GDPR). The General Data Protection Regulation (GDPR) is the correct regulatory framework because it specifically governs the processing of personal data of EU citizens and imposes strict requirements on automated decision-making and profiling under Article 22. This article 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. The GDPR also mandates data protection impact assessments and transparency obligations for such AI-driven processing.

Answer

Payment Card Industry Data Security Standard (PCI DSS)

Answer

Health Insurance Portability and Accountability Act (HIPAA)

Answer

Sarbanes-Oxley Act (SOX)

Question 10

Which TWO of the following are common techniques to improve the transparency and interpretability of an AI model?

Accepted Answer

Generate SHAP (SHapley Additive exPlanations) values. SHAP values are correct because they provide a unified measure of feature importance based on cooperative game theory, specifically Shapley values, which quantify the marginal contribution of each feature to a model's prediction. This makes the model's decision-making process transparent by showing how each input feature influences the output, which is a core technique for interpretability in AI governance.

Answer

Use differential privacy to add noise to training data

Answer

Implement a random forest algorithm

Answer

Use deep neural networks to increase model complexity

Question 11

Which TWO of the following are effective defenses against adversarial examples in AI systems?

Accepted Answer

Train the model with adversarial examples (adversarial training). Adversarial training (A) is effective because it exposes the model to perturbed inputs during training, forcing it to learn robust decision boundaries that are less sensitive to small, malicious perturbations. Input sanitization and feature squeezing (D) reduce the attack surface by compressing input features (e.g., reducing bit depth or spatial smoothing) to eliminate adversarial noise before inference, making it harder for an attacker to craft a successful perturbation.

Answer

Use an ensemble of models and majority voting

Answer

Increase the model's sensitivity to input changes

Answer

Reduce model complexity through pruning

Question 12

Which THREE of the following are key principles of trustworthy AI as defined by major regulatory bodies?

Accepted Answer

Fairness and non-discrimination. Fairness and non-discrimination (A) is a core principle of trustworthy AI because regulatory bodies like the European Commission's High-Level Expert Group on AI and the OECD require that AI systems do not perpetuate or amplify biases against protected groups. This involves implementing bias detection and mitigation techniques during model training and validation, such as using fairness metrics like demographic parity or equalized odds to ensure equitable outcomes across different demographic segments.

Answer

Maximum profitability

Answer

Proprietary secrecy

Question 13

A large e-commerce company uses a recommendation engine trained on millions of user interactions. Recently, the marketing team noticed a sharp increase in click-through rates for a particular product category. Upon investigation, an engineer found that a competitor had injected fake user profiles that consistently clicked on their products, skewing the training data. The company needs to remediate the attack and prevent future occurrences. The team has limited time and budget. Which course of action should the company take first?

Accepted Answer

Identify and remove the fake user profiles from the training dataset, then retrain the model. The immediate priority is to remove the poisoned data from the training set and retrain the model, as the fake profiles are actively skewing predictions and causing incorrect click-through rate spikes. This direct remediation addresses the root cause with minimal time and budget, aligning with the team's constraints. Without cleaning the data, any further training or defensive measures would still operate on corrupted inputs.

Answer

Implement adversarial training to make the model robust to future poisoning attempts

Answer

Decrease the frequency of model retraining to limit exposure to new data

Answer

Add differential privacy noise to the training data to mask the injected profiles

Question 14

A hospital deploys an AI diagnostic assistant that analyzes medical images. The system has been in use for six months, and radiologists have reported that the AI is increasingly confident in its predictions, but sometimes misses rare conditions. The AI ethics board is concerned about overreliance and potential harm from false negatives. They want to implement a governance framework that ensures appropriate human oversight. The hospital has a limited IT budget. What is the best approach?

Accepted Answer

Implement a human-in-the-loop process where the AI flags low-confidence or rare condition predictions for mandatory radiologist review. A human-in-the-loop process that triggers mandatory radiologist review only for low-confidence or rare-condition predictions directly addresses the risk of overreliance and false negatives without overwhelming the limited IT budget. This targeted oversight ensures that the AI's increasing confidence does not lead to missed rare conditions, while still allowing routine high-confidence predictions to proceed efficiently. The approach balances safety and resource constraints by focusing human attention where the AI is most likely to err.

Answer

Add a warning to the AI interface that says 'This tool may miss rare conditions'

Answer

Require all AI predictions to be reviewed by a radiologist before final diagnosis

Answer

Increase the AI's false positive threshold to reduce missed cases

Question 15

A credit union uses an AI model to approve personal loans. The model was trained on historical data from the past five years. A recent internal review shows that the model approves loans predominantly for white applicants compared to other ethnicities, even when income and credit scores are similar. The credit union wants to comply with fair lending laws without significantly reducing overall approval rates. The data science team has access to the training data. What is the most appropriate remediation step?

Accepted Answer

Resample the training data to ensure balanced representation of ethnicities. Resampling the training data to ensure balanced representation of ethnicities directly addresses the root cause of the bias—skewed historical data—without altering the model's decision logic or approval thresholds. By rebalancing the dataset (e.g., oversampling underrepresented groups or undersampling the majority), the model learns from a more equitable distribution of features, reducing disparate impact while preserving overall approval rates. This approach aligns with fair lending laws by mitigating bias at the data level, which is the most fundamental and effective remediation step.

Answer

Apply a fairness constraint that penalizes the model for disparate impact

Answer

Discontinue the AI model and use manual approval for all loans

Answer

Adjust the approval threshold so that approval rates are equal across ethnic groups