AI0-001 Practice Test 28 — 15 Questions

Question 1

A data scientist is preparing a dataset for a binary classification model. The dataset has 95% majority class and 5% minority class. Which data preparation technique is BEST to address the class imbalance?

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

Accepted Answer

SMOTE oversampling of the minority class. SMOTE (Synthetic Minority Oversampling TEchnique) generates synthetic samples for the minority class, balancing the dataset without simply duplicating existing minority instances.

Answer

Min-max normalization of all features

Answer

Random undersampling of the majority class

Answer

Removing all minority class samples

Question 2

An AI team is developing a model that approves loan applications. The dataset contains historical loan decisions where a protected group was disproportionately denied loans. The team wants to ensure the model does not perpetuate this bias. Which fairness metric should be used during validation to directly measure whether the model's positive prediction rate is equal across groups?

Accepted Answer

Demographic parity. Demographic parity requires the probability of a positive prediction (loan approval) to be equal across groups. This directly addresses the concern of perpetuating historical denial rates. Equalised odds measures error rates, not positive prediction rates.

Answer

Calibration

Answer

Individual fairness

Answer

Equalised odds

Question 3

A logistics company uses a machine learning model to predict delivery times based on historical data including distance, traffic, weather, and driver performance. The model is deployed as a REST API using Flask and run on a single server. Recently, the model has been returning predictions with high latency (over 2 seconds) during peak hours when the API receives 500 requests per second. The server has 8 CPU cores and 32 GB RAM. The model is a gradient boosting model (XGBoost) with 500 trees. The engineer wants to reduce inference latency to under 500ms without retraining the model. Which action is most effective?

Accepted Answer

Prune the model by reducing the number of trees to 100 and limit tree depth. Pruning the model by reducing the number of trees (e.g., from 500 to 100) directly reduces the number of decision paths evaluated during inference, which decreases latency. This does not require retraining; pruning can be applied post-training. Option B (linear regression) would require retraining and likely lose accuracy. Option C (horizontal scaling) improves throughput but does not reduce per-request latency; it may help if the server is overloaded, but pruning is more effective for latency. Option D (increasing RAM) does not address the CPU-bound tree traversal bottleneck.

Answer

Replace XGBoost with a linear regression model

Answer

Scale horizontally by deploying additional servers behind a load balancer

Answer

Increase server RAM to 128 GB

Question 4

A marketing team wants to segment customers into groups based on purchasing behavior without predefined categories. Which algorithm should they use?

Accepted Answer

K-means clustering. K-means clustering is an unsupervised learning algorithm that groups data points into clusters based on similarity without requiring predefined labels. Since the marketing team wants to segment customers based on purchasing behavior without predefined categories, K-means is the correct choice as it discovers natural groupings in the data.

Answer

Naive Bayes classifier

Answer

Logistic regression

Answer

Support vector machine

Question 5

A medical imaging team is developing an AI model to detect tumors from CT scans. They have 10,000 labeled scans, but the labels were created by a semi-automated process with an estimated 20% error rate (mislabeled tumor vs. no tumor). The team trains a convolutional neural network (CNN) and achieves 90% accuracy on a held-out test set that was carefully validated by an expert radiologist. However, when deployed to a new hospital's patient population, the accuracy drops to 70%. The team suspects domain shift and label noise. Which strategy is most likely to improve model robustness for the new hospital?

Accepted Answer

Use active learning to select the most uncertain predictions from the new hospital's data, then have an expert radiologist correct those labels. Active learning selects the most uncertain predictions from the new hospital's data, allowing an expert radiologist to efficiently correct the most informative labels. This directly addresses both label noise (by correcting mislabeled examples) and domain shift (by focusing on samples where the model is uncertain in the new domain). Option B is wrong because random selection may not target the most impactful errors, wasting expert effort. Option C is wrong because adding more noisy labels from the same flawed process will amplify label noise without correcting the domain shift. Option D is wrong because reducing model complexity and dropout are regularization techniques that do not fix label noise or domain shift.

Answer

Randomly select 1,000 scans from the new hospital and have them re-labeled by the radiologist

Answer

Collect 20,000 more scans with the same semi-automated labeling process

Answer

Reduce the CNN's number of layers and apply dropout to combat overfitting

Question 6

A data scientist is training a model to classify customer support tickets into categories. The dataset has 10,000 labeled examples, but the 'billing' category contains 8,000 examples while the 'technical' category contains 2,000. Which technique is most appropriate to address this imbalance before training?

Accepted Answer

Apply random oversampling on the 'technical' category.. Random oversampling duplicates examples from the minority class ('technical') to balance the class distribution, preventing the model from becoming biased toward the majority class ('billing'). This technique directly addresses the class imbalance before training, which is critical for classification tasks where the minority class is underrepresented.

Answer

Remove all examples except 'billing' and use a one-class classifier.

Answer

Use accuracy as the only evaluation metric.

Answer

Train the model as is, then adjust thresholds post-training.

Question 7

During a penetration test, a security analyst crafts a prompt that tricks an LLM into revealing its system prompt. Which OWASP LLM Top 10 vulnerability does this exploit?

Accepted Answer

Sensitive information disclosure. Tricking an LLM into revealing its system prompt directly exposes sensitive configuration details, which is a classic case of sensitive information disclosure. This vulnerability occurs when an attacker uses crafted inputs to bypass the model's intended boundaries and extract internal data that should remain confidential.

Answer

Model denial of service

Answer

Insecure output handling

Answer

Prompt injection

Question 8

A security team is threat modeling an AI-powered recommendation system. Using STRIDE, which THREE threats are MOST relevant to the model's training data pipeline?

Accepted Answer

Tampering with training data (data poisoning). Data poisoning directly targets the integrity of the training data pipeline. In STRIDE, 'Tampering' refers to unauthorized modification of data, and an attacker injecting malicious samples into the training set can cause the recommendation model to learn biased or harmful behaviors. This is a primary threat to AI systems, as the model's outputs are only as trustworthy as its training data.

Answer

Spoofing of data sources

Answer

Repudiation of data provenance

Question 9

Which TWO techniques are commonly used to handle missing data in a machine learning dataset? (Choose TWO.)

Accepted Answer

Imputation with mean or median. Imputation with mean or median is a standard technique for handling missing numerical data because it preserves the dataset size and avoids introducing bias from simply discarding rows. By replacing missing values with the central tendency of the observed data, the model can still learn patterns without losing information, though it may reduce variance slightly.

Answer

Normalization

Answer

One-hot encoding

Answer

Dimensionality reduction

Question 10

A financial services firm deploys an AI system to screen loan applications. The model was trained on historical data that reflected biased lending practices. After deployment, a regulatory body investigates and finds that the model denies loans at a disproportionately higher rate to a protected demographic group. The firm must address this issue while maintaining compliance with fair lending laws. The Chief AI Officer proposes four possible actions. Which action is the most appropriate first step?

Accepted Answer

Retrain the model using a debiased dataset and implement fairness-aware algorithms, then validate with fairness metrics. The most appropriate first step because it directly addresses the root cause of biased predictions by retraining the model on a debiased dataset and implementing fairness-aware algorithms. This proactive approach aligns with regulatory expectations to correct systemic bias. Option B (documenting) is reactive and does not fix the underlying issue. Option C (rule-based manual review) is inefficient, costly, and may still produce inconsistent decisions. Option D (disclosing to applicants) could be part of a remediation plan but is premature without first correcting the model.

Answer

Document the model's predictions and submit a report to the regulator explaining the historical dataset bias

Answer

Immediately deploy a rule-based system to manually review all denial decisions from the AI system

Answer

Disclose the bias findings to all rejected applicants and offer them priority reconsideration

Question 11

A hospital wants to deploy a machine learning model to predict patient readmission risk within 30 days. They have a dataset with 10,000 records, 70 features including demographics, lab results, and past admissions. The target variable is binary (readmitted or not). The data scientist trains a logistic regression model and achieves an AUC of 0.85 on the test set. However, the hospital's clinicians require interpretability of predictions to trust the model. Which action should the data scientist take to ensure the model meets the interpretability requirement while maintaining performance?

Accepted Answer

Use the logistic regression model as is, since it is inherently interpretable with coefficients. (PCA + logistic regression) reduces dimensionality but loses interpretability and may degrade performance. Option B (random forest with feature importance) is less interpretable than logistic regression. Option C (deep neural network with LIME/SHAP) adds complexity and may reduce transparency. Option D (keep logistic regression) provides inherent interpretability through coefficients, meeting the requirement without sacrificing performance.

Answer

Reduce the number of features to 10 using PCA and retrain the logistic regression

Answer

Replace logistic regression with a random forest model and use feature importance plots

Answer

Train a deep neural network and apply LIME or SHAP for explanations

Question 12

A company deploys an LLM-based application that retrieves external web content to answer user queries. An attacker crafts a webpage that, when retrieved, injects a hidden instruction telling the LLM to ignore its system prompt and output sensitive internal data. What type of attack is this?

Accepted Answer

Indirect prompt injection. Indirect prompt injection occurs when malicious instructions are embedded in content retrieved by the LLM, as opposed to direct injection where the user themselves provides the malicious prompt.

Answer

Direct prompt injection

Answer

Jailbreaking

Answer

Model inversion attack

Question 13

A company is preparing a dataset for training a supervised machine learning model. The dataset contains missing values, outliers, and categorical features. Which two preprocessing steps are typically performed to prepare the data? (Choose two.)

Accepted Answer

Impute missing values with the mean. Imputing missing values with the mean is a standard technique to handle incomplete data, ensuring the model can process all records without discarding potentially valuable information. Option C is correct because one-hot encoding converts categorical features into a binary vector representation, which is required by most machine learning algorithms that expect numerical input.

Answer

Normalize numerical features to a standard range

Answer

Remove all features with low variance

Answer

Increase the number of features using PCA

Question 14

During an AI model deployment, the operations team notices that inference requests are taking longer than expected. Which component is most likely causing the bottleneck?

Accepted Answer

The machine learning model's size and architecture. The machine learning model's size and architecture directly determine the computational complexity of inference. Larger models with more parameters or deeper architectures require more matrix multiplications and memory bandwidth, which increases latency per request. This is the most common bottleneck in AI deployment because the model itself is the core computation unit, and its inference time scales with its complexity.

Answer

Input data preprocessing pipeline

Answer

API gateway rate limiting

Answer

Database connection pool size

Question 15

A natural language processing (NLP) team is building a sentiment analysis model. The raw text data contains punctuation, stop words, and URLs. Which TWO preprocessing steps are most appropriate to improve model performance? (Choose two.)

Accepted Answer

Remove all punctuation and URLs. Removing punctuation and URLs eliminates noise that does not contribute to sentiment (e.g., 'http://...' or '!!!'), allowing the model to focus on meaningful words. This step reduces vocabulary size and prevents the model from learning spurious correlations tied to formatting artifacts.

Answer

Apply stemming to reduce words to root forms

Answer

Convert all text to lowercase

Answer

Tokenize the text into individual words