CPMAI AI Project Exam Prep practice questions and answers 2026. Tap an option to test yourself — you'll see the correct answer and a plain-English explanation for every question. Free, no login.
Q21A data scientist has developed a machine learning model with 99% accuracy on the training data but only 72% accuracy on the test data. What is the most likely issue?
✓ Correct answer: D. The model is overfitting to the training data
This significant drop in performance between training and test data is a classic sign of overfitting, where the model has memorized the training data rather than learning generalizable patterns.
Q22When deploying a machine learning model to production, which validation technique is most appropriate for estimating how the model will perform on unseen data?
✓ Correct answer: C. K-fold cross-validation
K-fold cross-validation provides a robust estimate of model performance by testing on multiple held-out data subsets, which better simulates real-world performance than a single train-test split.
Q23A company has deployed an AI recommendation system that initially showed good results but has seen declining performance over the past three months. What is the most likely cause?
✓ Correct answer: B. Concept drift in user preferences
Model drift occurs when the statistical properties of the target variable change over time, causing the model's predictions to become less accurate. This is common in recommendation systems as user preferences evolve.
Q24When aligning AI model results with business KPIs, which approach is most effective?
✓ Correct answer: A. Define business metrics before model development and design the model evaluation to reflect these metrics
Defining clear business metrics before model development ensures the AI solution is designed to address specific business goals rather than just technical performance metrics.
Q25What is the primary benefit of implementing a blue/green deployment strategy for AI models?
✓ Correct answer: D. Ability to quickly roll back to the previous model version if issues are detected
Blue/green deployment allows for immediate rollback to the previous version if issues are detected with the new model, minimizing downtime and user impact.
Q26A company is deciding between on-premise and cloud deployment for their new machine learning model. Which factor is MOST critical in this decision?
✓ Correct answer: C. Data privacy requirements and regulatory constraints
Data privacy and regulatory requirements often dictate where data can be stored and processed, making this a critical factor in deployment decisions, especially for sensitive information.
Q27What is the primary purpose of implementing a canary deployment for an AI model?
✓ Correct answer: B. To test the model with a small percentage of users before full deployment
Canary deployments limit risk by exposing the new model to a small percentage of users first, allowing for monitoring of performance and issues before full deployment.
Q28In the context of MLOps, what is the primary purpose of a feature store?
✓ Correct answer: A. To provide a centralized repository of features that can be reused across different models and applications
Feature stores serve as centralized repositories for preprocessed features, ensuring consistency across training and inference while reducing redundant computation.
Q29A company notices that their deployed fraud detection model has a high false positive rate. What is the most appropriate immediate action?
✓ Correct answer: D. Adjust the classification threshold to better balance precision and recall
Adjusting the classification threshold can help balance the trade-off between false positives and false negatives without requiring model retraining.
Q30What is the most effective strategy for monitoring an AI model in production?
✓ Correct answer: C. Implement monitoring for input data distributions, model predictions, and business metrics
A comprehensive monitoring strategy includes both input data and output predictions, along with business metrics, to detect issues like data drift, model drift, and misalignment with business goals.
Use the free CPMAI AI Project Exam Prep sample, download the PDF, then unlock web-based timed mock exams for a full exam rehearsal.
Theory