AI0-001 Practice Test 39 — 15 Questions

Question 1

A company is training a large language model from scratch and wants to minimise its environmental impact. Which practice aligns with green AI principles?

Accepted Answer

Use model pruning and train on a smaller, representative dataset. Green AI advocates for resource-efficient AI, including using smaller models, pruning, and efficient architectures to reduce carbon footprint. Training larger models with more data increases environmental impact, not reduces it.

Answer

Use more GPUs to parallelise training and reduce wall-clock time

Answer

Deploy the model on a cloud provider with renewable energy certificates

Answer

Train the model on a larger dataset to improve accuracy

Question 2

Which NIST AI RMF function involves identifying the context, risks, and potential impacts of an AI system, including mapping the AI lifecycle and stakeholders?

Accepted Answer

Map. The AI RMF's four functions are: Govern, Map, Measure, Manage. Map focuses on context and risk identification. Govern sets policies. Measure evaluates metrics. Manage addresses risks through controls.

Answer

Manage

Answer

Measure

Answer

Govern

Question 3

Which privacy-preserving technique allows a model to be trained across decentralized data sources without the raw data ever leaving each source?

Accepted Answer

Federated learning. Federated learning trains models locally on each device or server and only shares model updates, preserving data locality.

Answer

Homomorphic encryption

Answer

Secure multi-party computation

Answer

Differential privacy

Question 4

A security engineer is conducting threat modeling for an AI system that uses a pre-trained image classifier. Applying STRIDE, which threat category most directly addresses an attacker manipulating the model's behavior by providing carefully crafted inputs that the model was not trained to handle robustly?

Accepted Answer

Tampering. Tampering involves unauthorized modification of data or systems. In this context, adversarial examples tamper with the input data to alter the model's behavior. Spoofing is about impersonation, Repudiation is about denying actions, and Information disclosure is about exposing sensitive data.

Answer

Repudiation

Answer

Information disclosure

Answer

Spoofing

Question 5

A developer is building an AI-powered code completion tool. They want to ensure that the tool does not inadvertently suggest insecure code patterns. Which practice is MOST effective for reducing this risk?

Accepted Answer

Red teaming the AI system. Red teaming involves adversarial testing to find vulnerabilities. Output filtering can catch some insecure suggestions but may not cover all patterns. Secure data pipelines focus on training data security, not output. Rate limiting is unrelated.

Answer

Rate limiting

Answer

Secure data pipelines

Answer

Output filtering of insecure patterns

Question 6

A team is developing a threat model for an AI system that processes user uploads. Using STRIDE, which threat involves an attacker modifying the model's training data to cause misclassification?

Accepted Answer

Tampering. Tampering is the STRIDE category for unauthorized modification of data. Data poisoning is a form of tampering with training data.

Answer

Spoofing

Answer

Repudiation

Answer

Information disclosure

Question 7

During a security audit of an AI system, the auditor applies the STRIDE threat model. Which threat category is MOST relevant to an attacker manipulating the training data to cause the model to misbehave on specific inputs?

Accepted Answer

Tampering. Tampering refers to unauthorized modification of data or code. Data poisoning is a form of tampering with the training dataset.

Answer

Spoofing

Answer

Repudiation

Answer

Information disclosure

Question 8

A company is deploying a pre-trained image classification model for facial recognition in a security system. They are concerned about adversarial examples. Which TWO of the following are effective defenses against adversarial examples?

Accepted Answer

Adversarial training during model development. Adversarial training (including the model with adversarial examples during training) and input sanitization (e.g., JPEG compression, denoising) are proven defenses against adversarial perturbations. Gradient masking is a weak defense. Homomorphic encryption and federated learning are unrelated to adversarial robustness.

Answer

Gradient masking to hide model gradients

Answer

Homomorphic encryption of input images

Answer

Federated learning to train on distributed data

Question 9

A team is designing a secure API for an AI model. They want to prevent data leakage through overly detailed error messages. Which principle should they follow?

Accepted Answer

Use generic error messages. Least-privilege API access and minimal error information reduce the attack surface. Specifically, returning generic error messages prevents leaking internal details.

Answer

Return detailed error codes for debugging

Answer

Log errors to the client side

Answer

Disable all error messages

Question 10

An organization wants to classify support tickets into categories (billing, technical, etc.). Which type of machine learning is most suitable?

Accepted Answer

Supervised learning. Supervised learning is the correct choice because the organization has labeled historical support tickets (e.g., 'billing' or 'technical') and wants to train a model to map new tickets to these predefined categories. This is a classic classification task, where the algorithm learns from input-output pairs to predict the correct label for unseen data.

Answer

Unsupervised learning

Answer

Reinforcement learning

Answer

Regression

Question 11

A security team is reviewing an LLM-powered application that can execute SQL queries based on user requests. They want to implement defenses against prompt injection that could lead to unauthorized database access. Which TWO controls are MOST effective? (Select two.)

Accepted Answer

Input validation and sanitization. Input validation and sanitization can filter malicious prompt content, and output filtering can block responses containing unauthorized data. Rate limiting does not prevent injection, access controls on the database are important but not a direct defense against injection (they limit impact), and robust training does not prevent injection at inference time.

Answer

Robust training techniques

Answer

Access controls on the database

Answer

Rate limiting

Question 12

A large financial services company deploys multiple AI models on a shared Kubernetes cluster with GPU nodes. The models serve real-time fraud detection and credit scoring. Recently, the operations team observed frequent out-of-memory (OOM) errors during peak hours, causing inference failures. The monitoring dashboards show GPU memory utilization averaging 90% during peak times, and pods are being evicted. The team has allocated 8GB per pod and the total cluster GPU memory is 32GB. The models require at least 4GB each, but the fraud detection model occasionally spikes to 7GB. Which course of action best resolves the OOM errors while maintaining high availability?

Accepted Answer

Set resource limits and requests per model based on observed usage, and implement pod priority classes. It uses Kubernetes resource management features—setting precise resource requests and limits based on observed GPU memory usage—combined with pod priority classes to ensure critical fraud detection pods are scheduled and retained during contention. This prevents OOM errors by capping memory per pod while allowing the spike-prone fraud model to be prioritized over less critical workloads, maintaining high availability without overprovisioning.

Answer

Reduce the batch size and model complexity for all models to lower memory footprint

Answer

Provision larger GPU nodes with 48GB memory each

Answer

Increase the memory request for all pods to 8GB to ensure they have enough

Question 13

A healthcare startup needs to deploy an AI model for real-time patient monitoring on IoT devices with limited battery and compute. The model must run locally with minimal latency. Which TWO strategies are most appropriate?

Accepted Answer

Use TensorFlow Lite to convert and run the model on the device. TensorFlow Lite is specifically designed to run TensorFlow models on resource-constrained edge devices like IoT sensors. It optimizes the model for low latency inference by using a specialized interpreter and hardware acceleration delegates (e.g., NNAPI, GPU), enabling real-time patient monitoring without cloud dependency.

Answer

Apply model distillation to create a smaller student model

Answer

Deploy the model on a cloud server and stream data

Answer

Use ONNX Runtime with a GPU backend

Question 14

A company wants to use AI to automatically detect anomalies in server log data. The data is time-series and labeled with 'normal' and 'anomaly' for the past year. Which TWO techniques are appropriate for this use case?

Accepted Answer

Use a time-series anomaly detection model (e.g., Isolation Forest with sliding windows). Isolation Forest with sliding windows is a well-suited unsupervised technique for detecting anomalies in time-series data by isolating outliers in feature windows extracted from the log stream. Option C is correct because the company has labeled data ('normal' and 'anomaly'), enabling a supervised classification model like XGBoost to learn patterns from engineered features and predict anomalies accurately.

Answer

Train an image classification model (CNN) on screenshots of log graphs

Answer

Use a code generation model to fix the anomalies automatically

Answer

Build a recommendation system based on user activity logs

Question 15

Which TWO actions should be taken to ensure an AI model complies with GDPR requirements when processing personal data?

Accepted Answer

Limit data collection to only what is necessary for the model. GDPR's data minimization principle (Article 5(1)(c)) requires that personal data collected be adequate, relevant, and limited to what is necessary for the purpose for which it is processed. In AI model training, this means collecting only the features essential for the model's objective, reducing the risk of processing excessive or irrelevant personal data.

Answer

Provide a full explanation of model predictions

Answer

Store all user data for a minimum of 10 years

Answer

Anonymize all personal data before use