AI0-001 Practice Test 37 — 15 Questions

Question 1

A developer is building an LLM-powered code assistant. They want to prevent the model from generating insecure code. Which OWASP LLM Top 10 category is most relevant to this risk?

Accepted Answer

Insecure output handling. Insecure output handling (B) is the most relevant OWASP LLM Top 10 category because the risk is that the LLM generates insecure code, which is a direct output from the model. This category specifically addresses failures to validate, sanitize, or restrict the model's output before it is used in downstream applications, such as a code assistant. By not properly handling the generated code, the assistant could introduce vulnerabilities like SQL injection or command injection into the user's codebase.

Answer

Sensitive information disclosure

Answer

Model denial of service

Answer

Prompt injection

Question 2

A data scientist is preparing a dataset for a binary classification model to detect fraudulent transactions. The dataset contains 1% fraudulent and 99% legitimate transactions. The goal is to maximize recall for the fraud class while maintaining a precision above 0.5. Which data preparation strategy is MOST effective?

Accepted Answer

Apply SMOTE (Synthetic Minority Oversampling Technique) to generate synthetic fraud examples. Handling imbalanced data typically requires resampling. For recall maximization with moderate precision constraint, oversampling the minority class (e.g., SMOTE) is effective. Undersampling loses too many majority samples, and using raw data leads the model to predict the majority class.

Answer

Apply random undersampling of the majority class until the dataset is balanced

Answer

Remove all duplicate transactions from the dataset

Answer

Use the raw dataset without any resampling, relying on class weights during training

Question 3

Which TWO of the following are common methods for mitigating bias in AI models?

Accepted Answer

Reweighting training samples based on sensitive attributes. Reweighting training samples based on sensitive attributes is a common pre-processing bias mitigation technique. It assigns higher weights to underrepresented groups or lower weights to overrepresented groups to balance the dataset, thereby reducing the model's reliance on biased correlations. This method directly addresses data-level bias before model training begins.

Answer

Using adversarial training

Answer

Applying L1 regularization

Answer

Performing k-fold cross-validation

Question 4

An AI security team is conducting a threat model for a new document summarization service. They want to identify threats related to spoofing of the AI's identity. Which STRIDE category should they consider?

Accepted Answer

Spoofing. Spoofing in STRIDE refers to impersonating something or someone else. In the context of AI, an attacker could spoof the AI service to provide false summaries.

Answer

Repudiation

Answer

Tampering

Answer

Information disclosure

Question 5

Which technique adds controlled noise to query results or training data to prevent an attacker from inferring whether a specific individual's data was included in the dataset?

Accepted Answer

Differential privacy. Differential privacy adds calibrated noise to ensure the output does not reveal individual participation. Anonymisation removes identifiers. Pseudonymisation replaces identifiers. Federated learning decentralises data but does not necessarily add noise.

Answer

Anonymisation

Answer

Federated learning

Answer

Pseudonymisation

Question 6

A financial institution uses a machine learning model to approve loans. They want to protect against membership inference attacks. Which THREE techniques are effective?

Accepted Answer

Applying model truncation or output perturbation. Differential privacy adds noise to training, model truncation reduces overfitting (which helps prevent inference), and limiting output granularity reduces the information leaked. Shadow models are used to train attack models, not defend. Federated learning alone does not prevent inference.

Answer

Implementing federated learning

Answer

Using shadow models to distract attackers

Question 7

A credit risk model is being developed to predict loan defaults. The dataset has 95% non-default and 5% default instances. The data scientist trains a logistic regression model and obtains 95% accuracy, but the recall for defaults is only 10%. Which action is most appropriate to improve the model's ability to identify defaults?

Accepted Answer

Use SMOTE to oversample the default class. SMOTE (Synthetic Minority Oversampling Technique) is the most appropriate action because it generates synthetic samples for the minority class (defaults) rather than simply duplicating existing ones. This directly addresses the severe class imbalance (95% non-default vs. 5% default) that causes the logistic regression model to achieve high accuracy by predicting nearly all instances as non-default, while failing to identify actual defaults (recall of only 10%). By creating realistic synthetic default instances, SMOTE balances the training data and forces the model to learn decision boundaries that better capture the minority class.

Answer

Apply principal component analysis (PCA) to reduce dimensionality

Answer

Collect more data from loan applicants to increase dataset size

Answer

Undersample the non-default class to match the number of defaults

Question 8

A large e-commerce company has deployed a real-time product recommendation system using a neural collaborative filtering model. The model was trained on six months of user click and purchase data. For the first three months after deployment, the click-through rate (CTR) improved by 15%. However, starting in the fourth month, CTR began decreasing steadily despite no changes to the system infrastructure or data pipeline. The product manager suspects model decay but the engineering team insists the model is static and should not degrade. The data science lead suggests investigating further. They have access to production logs, A/B testing framework, and historical model versions. What is the BEST course of action to diagnose and address the issue?

Accepted Answer

Set up an A/B test comparing the current model against the original baseline model using recent traffic.. Setting up an A/B test comparing the current model against the original baseline model using recent traffic directly isolates whether the model's predictive performance has degraded due to concept drift (changes in user behavior over time). Since the model is static but the data distribution has shifted, the A/B test provides empirical evidence of decay by measuring CTR differences under identical conditions, which is the standard diagnostic step before any retraining or feature engineering.

Answer

Re-deploy the model with additional features such as time of day and user device.

Answer

Increase the frequency of batch inference from hourly to every 10 minutes to improve responsiveness.

Answer

Retrain the model on only the most recent 30 days of data and replace the current model.

Question 9

An AI ethics board is reviewing a model that recommends criminal sentencing lengths. They want to ensure that the model's false positive rates for different demographic groups are equal. Which fairness metric should they use?

Accepted Answer

Equalized odds. Equalized odds requires that the model's true positive rates and false positive rates are equal across groups. Demographic parity only requires equal selection rates. Individual fairness ensures similar individuals are treated similarly but does not define group rates. Calibration ensures predicted probabilities match actual outcomes for each group but does not enforce equal error rates.

Answer

Calibration

Answer

Individual fairness

Answer

Demographic parity

Question 10

A company wants to build a system that automatically tags uploaded images with objects they contain (e.g., 'car', 'tree', 'person'). Which AI application type is this?

Accepted Answer

Image classification/object detection. The task of identifying and labeling objects (e.g., 'car', 'tree', 'person') within an image is a classic use case for image classification combined with object detection. Image classification assigns a single label to the entire image, while object detection localizes and classifies multiple objects within the image, which is exactly what the system requires.

Answer

Recommendation system

Answer

Anomaly detection

Answer

Document intelligence

Question 11

After deploying a model for fraud detection, the data scientist observes a steady decline in precision over two months. Which issue is most likely occurring?

Accepted Answer

Concept drift. Concept drift occurs when the statistical properties of the target variable change over time, causing the model's decision boundary to become outdated. In fraud detection, fraudsters continuously adapt their methods, so the relationship between input features and the fraud label shifts, leading to a steady decline in precision as false positives increase.

Answer

Data drift

Answer

Model overfitting

Answer

Adversarial attack

Question 12

A company wants to use machine learning to recommend products to customers based on their purchase history. Which TWO techniques are appropriate for this task? (Select TWO)

Accepted Answer

Collaborative filtering. Collaborative filtering recommends based on user similarities. K-Nearest Neighbors can find similar users or items. Both are suitable for recommendation.

Answer

Principal Component Analysis (PCA)

Answer

Naive Bayes

Answer

Linear regression

Question 13

Which of the following best describes the difference between narrow AI and general AI?

Accepted Answer

Narrow AI is designed for a specific task; general AI aims to perform any cognitive task a human can.. Narrow AI specializes in one task, while general AI would possess human-like cognitive abilities across domains.

Answer

Narrow AI relies on supervised learning; general AI uses unsupervised learning exclusively.

Answer

Narrow AI requires large datasets; general AI can learn from few examples.

Answer

Narrow AI can perform any intellectual task; general AI is limited to specific tasks.

Question 14

A company wants to build a real-time anomaly detection system for IoT sensor data using edge AI. The model must run on resource-constrained devices with minimal power consumption. Which model optimization technique is MOST important?

Accepted Answer

Model quantization (INT8). Quantization reduces model precision (e.g., FP32 to INT8), decreasing model size and computation, which is critical for resource-constrained edge devices.

Answer

Use FP32 precision

Answer

Increase the number of layers

Answer

Use a larger batch size

Question 15

A team is training a deep learning model for image classification. The training loss decreases steadily but the validation loss plateaus after 20 epochs and then starts to increase. Which action is MOST likely to improve generalization?

Accepted Answer

Implement early stopping. Early stopping halts training when validation loss stops improving, preventing overfitting. Increasing learning rate would worsen divergence; adding more layers increases capacity and overfitting; reducing batch size may help optimization but not directly address overfitting.

Answer

Add more convolutional layers

Answer

Increase the learning rate

Answer

Reduce the batch size