AAIA Timed Practice Test 1 — 10 Questions

Question 1

An organization is scaling its AI governance program. You need to automate the identification of 'Shadow AI' in the corporate network. Which tool would be the most effective to gain visibility into unauthorized AI model API usage?

10 minute time limit — choose an answer to begin.

Accepted Answer

Microsoft Defender for Cloud Apps. Cloud Access Security Brokers (CASBs) like Microsoft Defender for Cloud Apps provide visibility into unsanctioned SaaS and API usage, including AI platforms.

Answer

Azure AI Content Safety

Answer

GitHub Copilot Business logs

Answer

Azure Policy

Question 2

Which action is required when a model is found to be 'Non-compliant' with the internal AI Governance policy?

Accepted Answer

Immediate remediation or removal from the production environment. Non-compliant models must be removed from production or restricted until remediated to prevent risk.

Answer

Writing a blog post about the failure

Answer

Ignoring the finding if the model is profitable

Answer

Giving the model a new name

Question 3

What is the purpose of an 'AI Audit trail'?

Accepted Answer

Providing transparency and accountability. To provide a transparent, verifiable record of all actions taken on an AI system.

Answer

Reducing cloud latency

Answer

Increasing model performance

Answer

Changing model parameters

Question 4

A firm uses a third-party AI service. Which control is most critical to ensure compliance with the firm's own AI governance policy?

Accepted Answer

Including AI-specific compliance requirements in the vendor service-level agreement (SLA). Supply chain risk management requires mapping vendor controls to internal governance requirements.

Answer

Conducting weekly penetration tests on the vendor's office premises

Answer

Requesting the vendor's employee turnover rates for the last fiscal year

Answer

Requiring the vendor to submit copies of all their marketing brochures

Question 5

Which TWO of the following are examples of 'Data Leakage' that an auditor should look for in a pipeline?

Accepted Answer

Training on test data. Option B is correct because training on test data is a classic form of data leakage: information from the evaluation set contaminates the training process, so the model's reported performance is optimistically biased and does not reflect true generalization to unseen data. Option E is correct because including target-related features that are not available at inference time (e.g., using the label or a post-outcome variable as an input) leaks future/target information into training, producing a model that cannot be reproduced or used in production. Option A is not leakage per se; using cloud-based storage is an infrastructure choice that may raise other security or compliance concerns but does not itself contaminate training with evaluation or target information. Option C is not leakage; documenting model parameters is a governance and reproducibility practice. Option D is not leakage; using open-source libraries is a normal engineering decision that may introduce supply-chain risk but does not constitute data leakage.

Answer

Using cloud-based storage

Answer

Documenting model parameters

Answer

Using open-source libraries

Question 6

Which TWO are common 'AI Operational' challenges?

Accepted Answer

Data quality management. B (Data quality management) is a core AI operational challenge because production models depend on clean, consistent, and representative training and inference data; poor data quality causes drift, bias, and degraded accuracy that must be continuously monitored and remediated. E (Model maintenance) is also correct because deployed models require ongoing retraining, versioning, performance monitoring, and updates as data distributions and business requirements change over time. The remaining options are not AI operational challenges: A (Selecting a font) is a design/typography decision, C (Writing poetry) is a creative content task rather than an operational concern, and D (Buying new office chairs) is a facilities/procurement matter unrelated to running AI systems.

Answer

Selecting a font

Answer

Writing poetry

Answer

Buying new office chairs

Question 7

In the context of AI regulatory compliance, what is the 'Model Risk Management' (MRM) framework mainly concerned with?

Accepted Answer

Governance and validation controls. MRM manages the risks associated with model usage, including governance, validation, and control.

Answer

Reducing cloud expenses

Answer

Increasing developer productivity

Answer

Marketing model features

Question 8

You are auditing an organization's AI deployment in Microsoft Azure. The Chief Risk Officer requires a central dashboard to monitor AI model performance drift and data quality metrics across all deployed services. Which tool should you confirm is configured to provide these insights?

Accepted Answer

Azure Machine Learning Model Monitor. Azure Machine Learning Model Monitor is the service designed to continuously track model performance drift and data quality metrics for models deployed via Azure Machine Learning. It provides a central dashboard with metrics like feature attribution drift, data drift, and prediction drift, enabling the CRO to monitor all deployed services.

Answer

Microsoft Purview AI Compliance Manager

Answer

Azure Policy for AI

Answer

Azure AI Content Safety

Question 9

When auditing model versioning, what is the recommended practice for maintaining evidence?

Accepted Answer

Use immutable versioning. Immutable model repositories and versioning control (e.g., Git LFS, MLflow) ensure the exact state can be audited.

Answer

Delete old versions

Answer

Use mutable file names

Answer

Store models in cache

Question 10

A firm uses a 'Federated Learning' approach. Which governance benefit does this provide?

Accepted Answer

Enhanced data privacy by keeping raw data locally on devices. Federated learning allows training on decentralized data, keeping raw data on local devices, which enhances privacy.

Answer

Reduced usage of electricity

Answer

Faster model training times

Answer

Improved model interpretability