AAISM Practice Test 15 — 15 Questions

Question 1

Which of the following is an example of an 'Evasion' attack on a computer vision model?

Accepted Answer

Adding adversarial noise to input images. Adding adversarial noise to an image (e.g., a sticker) causes the model to misclassify it, which is the definition of evasion.

Answer

Intercepting training data packets

Answer

Changing the model's learning rate

Answer

Brute forcing the administrator password

Question 2

You are auditing an AI system for compliance with the EU AI Act. Which control is mandatory for 'High-Risk' AI systems to ensure transparency for human oversight?

Accepted Answer

Human-in-the-loop design and logging. The EU AI Act requires that high-risk systems be designed with human-in-the-loop (HITL) capabilities and adequate logging to ensure transparency and accountability.

Answer

Automated decision-making without oversight

Answer

Mandatory anonymization of all input data

Answer

Exclusive use of proprietary datasets

Question 3

When utilizing the 'Adversarial Robustness Toolbox' (ART) to defend against evasion attacks, which attack type is specifically mitigated by using 'Adversarial Training'?

Accepted Answer

Evasion attacks. Adversarial training involves augmenting the training set with adversarial examples, hardening the model against evasion (inference-time) attacks.

Answer

Model Inversion attacks

Answer

Model Theft attacks

Answer

Supply Chain attacks

Question 4

You are configuring Microsoft Defender for Cloud to monitor your Azure OpenAI service. Which specific resource diagnostic setting must be enabled to ensure that prompt injection attempts are logged in the Log Analytics workspace?

Accepted Answer

Enable 'RequestResponse' diagnostic logs for the AI resource. To detect security threats like prompt injection, you must enable diagnostic logging for the Azure OpenAI service specifically for the 'RequestResponse' category, which captures the interaction payload.

Answer

Enable 'AccessControl' logging in Entra ID

Answer

Enable 'MetricAlerts' at the subscription level

Answer

Enable 'AuditLogs' in the Storage Account settings

Question 5

Which THREE of the following are examples of AI system 'Access Controls'?

Accepted Answer

API Key authentication. Access controls include authentication (IAM), network restrictions (VPC/Security Groups), and fine-grained permissions for API operations.

Answer

Disabling logging services

Answer

Increasing model training epochs

Question 6

When an AI model is suspected of being involved in a security incident, what is the first step in the incident response process?

Accepted Answer

Isolate the model endpoint from the network. Isolation is the first step to prevent further damage or exfiltration of sensitive data.

Answer

Retrain the model on new data

Answer

Delete the model weights

Answer

Update the documentation

Question 7

What is the purpose of 'Model Signing' in an MLOps pipeline?

Accepted Answer

To ensure model integrity and non-repudiation. Model signing uses digital signatures to verify that the model artifact has not been tampered with between the training environment and the production environment.

Answer

To encrypt the model weights at rest

Answer

To restrict access to the model registry

Answer

To increase inference speed

Question 8

Which security principle is violated when an AI model is allowed to access external data sources without sufficient authorization checks on the data itself?

Accepted Answer

Broken Access Control. Unauthorized data access via AI agents often results from 'Broken Access Control' where the AI inherits privileges it should not possess.

Answer

Model Poisoning

Answer

Model Inversion

Answer

Evasion Attack

Question 9

What is the primary risk associated with 'Shadow AI' in an organization?

Accepted Answer

Unmonitored data leakage and compliance risk. Shadow AI refers to the use of AI tools without IT or security oversight, leading to data leaks and compliance violations.

Answer

Lower cloud costs

Answer

Increased model accuracy

Answer

Faster code development

Question 10

To prevent unauthorized access to an Amazon Bedrock model, which AWS service is the primary point of control for enforcing 'Least Privilege' at the API layer?

Accepted Answer

AWS IAM. AWS IAM (Identity and Access Management) is the primary tool for defining permissions for Bedrock API operations.

Answer

AWS Glue

Answer

AWS CloudFront

Answer

Amazon S3

Question 11

Which TWO of the following are risks associated with using public pre-trained models without further security validation?

Accepted Answer

Hidden backdoors injected into the model weights. Public models may contain hidden backdoors (poisoning) or have been trained on sensitive/copyrighted data that could cause legal or compliance issues.

Answer

Improved model interpretability

Answer

Reduced inference latency compared to custom models

Answer

Automatic hardware acceleration

Question 12

Why is it important to perform 'Model Validation' before production deployment?

Accepted Answer

To verify performance and safety against benchmarks. Model validation ensures that the model meets performance and safety requirements, minimizing the risk of unexpected behavior.

Answer

To automatically remove users from the system

Answer

To encrypt the model weights

Answer

To reduce the cost of GPU hardware

Question 13

Which technique is best suited to defend against 'Model Extraction' where an adversary queries a public API to build a clone of your proprietary model?

Accepted Answer

Implement output perturbation. Adding noise (differential privacy) to the outputs or rate-limiting makes the model's decision boundaries difficult to map precisely.

Answer

Enable data augmentation

Answer

Use a larger model architecture

Answer

Rotate API tokens every 30 days

Question 14

To mitigate adversarial evasion attacks against a vision model, you decide to implement defensive distillation on your TensorFlow model. What is the primary operational trade-off you must evaluate?

Accepted Answer

Reduced model sensitivity to noise and potential drop in clean accuracy. Defensive distillation increases the model's robustness against small adversarial perturbations but can lead to reduced accuracy on clean data or increased training complexity.

Answer

Requirement for GPU-based training clusters only

Answer

Incompatibility with gradient-based optimizers

Answer

Increased latency in inference requests

Question 15

When implementing Google Cloud's 'Vertex AI' model guardrails, you need to prevent the leakage of PII in model outputs. Which configuration tool is used to enforce these constraints at the request/response level?

Accepted Answer

Vertex AI 'Safety Settings' filters. The Vertex AI 'Safety Settings' and 'Content filtering' allow practitioners to define thresholds for harassment, hate speech, and PII leakage.

Answer

Vertex AI 'Feature Store' access control

Answer

Vertex AI 'Model Registry' IAM roles

Answer

Vertex AI 'Pipeline' execution logs