AAISM Practice Test 2 — 10 Questions

Question 1

Which THREE activities help manage 'AI Lifecycle' risk?

Accepted Answer

Pre-deployment design and threat modeling.. Option A is correct because pre-deployment design and threat modeling identify AI-specific risks (e.g., data poisoning, model evasion, prompt injection, bias) before the system is fielded, allowing controls to be built in rather than bolted on. Option B is correct because formal decommissioning and sunsetting processes manage end-of-life risk by retiring models, revoking credentials, archiving or destroying training data, and preventing stale or unsupported AI components from remaining exploitable. Option E is correct because continuous operational monitoring detects model drift, data drift, performance degradation, and anomalous inference behavior during runtime, which is essential since AI risk evolves after deployment. Option C is incorrect because ignoring software updates leaves known vulnerabilities unpatched and increases lifecycle risk. Option D is incorrect because bypassing security reviews for speed skips necessary risk assessment and control validation, directly undermining AI lifecycle risk management.

Answer

Ignoring all software updates.

Answer

Bypassing security reviews for speed.

Question 2

Which THREE security controls are essential for protecting the integrity of the AI model training supply chain?

Accepted Answer

Data provenance and lineage tracking. Data provenance and lineage tracking (A) is essential because it records the origin, transformations, and custody of every training dataset, allowing integrity violations such as tampered or poisoned data to be detected and traced through the pipeline. SBOM analysis for dependencies (B) is required because AI training stacks pull in numerous third-party libraries and frameworks, and an SBOM lets you identify and remediate vulnerable or malicious components before they affect the model. Cryptographic signing of model artifacts (D) is essential because digital signatures (e.g., using asymmetric keys over hashes of weights and checkpoints) let consumers verify that a model was produced by a trusted source and has not been altered in transit or storage. Hardening the physical security of the server room (C) protects physical access to hardware but does not address the logical integrity of data, dependencies, or model artifacts across the supply chain, and increasing the number of neurons (E) is a model-capacity design choice that has no bearing on supply chain security.

Answer

Hardening the physical security of the server room

Answer

Increasing the number of neurons in the model

Question 3

A researcher is using 'Adversarial Training' to defend against evasion. How exactly does this work?

Accepted Answer

It trains the model on adversarial examples alongside clean data.. Adversarial training is a defense technique where the model is trained on a dataset that includes both clean examples and adversarial examples—inputs intentionally perturbed to fool the model. By learning from these adversarial examples, the model becomes more robust and can correctly classify them, thereby improving its resistance to evasion attacks.

Answer

It uses a secondary model to detect malicious input.

Answer

It restricts input to a whitelist of authorized users.

Answer

It adds noise to the weights of the model.

Question 4

What is the purpose of 'Logging' in an AI system?

Accepted Answer

To provide an audit trail for incident investigation. Logging in an AI system captures events, inputs, outputs, model versions, and decisions over time so that incidents can be reconstructed and responsibility assigned. Its primary governance purpose is to provide an audit trail for incident investigation (B) — determining what the model did, why, and under what conditions when something goes wrong. This is foundational to AI incident response, regulatory reporting, and post-mortem analysis.

Answer

To prevent unauthorized access to the server

Answer

To encrypt data at rest

Answer

To compress the model for faster deployment

Question 5

Which THREE elements are essential for 'AI Governance'?

Accepted Answer

Defined AI ethics and safety policies.. AI governance requires a formal policy foundation, so option A (defined AI ethics and safety policies) is correct because written principles for fairness, transparency, accountability, and safety give the organization enforceable rules and decision criteria for AI use. Option C (a standard AI risk management framework) is correct because a recognized framework such as the NIST AI RMF or ISO/IEC 42001 provides a structured, repeatable process for identifying, assessing, treating, and monitoring AI risks across the lifecycle. Option E (a cross-functional oversight committee) is correct because governance needs an accountable body drawing on legal, compliance, security, data science, and business stakeholders to approve, review, and enforce AI decisions. Option B is not essential and would violate privacy and least-privilege principles, since governance never requires blanket access to employees' private emails. Option D is not essential because governance is about policy, risk, and accountability, not about mandating the most expensive GPU hardware, which is a cost and performance decision rather than a governance element.

Answer

Access to all employees' private emails.

Answer

A mandate to use the most expensive GPUs available.

Question 6

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. Output perturbation deliberately adds noise or rounds confidence scores returned by the API, degrading the fidelity of the information an adversary can use to train a clone. Model extraction relies on precise output signals, so perturbing them reduces clone accuracy while preserving acceptable utility for legitimate users. This is a direct, low-cost mitigation at the inference boundary.

Answer

Enable data augmentation

Answer

Use a larger model architecture

Answer

Rotate API tokens every 30 days

Question 7

Which THREE of the following are potential indicators of a 'Model Poisoning' attack occurring during the training phase?

Accepted Answer

Inconsistent model convergence during training. Option B is correct because model poisoning corrupts the training data or update process, so the loss curve and parameter updates no longer converge consistently, producing erratic or stalled convergence during training. Option C is correct because poisoned training samples can cause the model to memorize manipulated patterns, yielding high training-set accuracy while validation accuracy degrades due to the corrupted decision boundary. Option E is correct because attackers often implant backdoors or trigger-based behaviors, so the model behaves normally on ordinary inputs but produces attacker-chosen outputs when a specific trigger pattern is present. Option A is not a training-phase indicator, since increased API request latency reflects inference or service performance rather than poisoned training data. Option D is also not indicative, because larger model-weight storage is a resource/cost artifact and does not by itself demonstrate poisoned training behavior.

Answer

Increased latency for API requests

Answer

Increase in storage costs for the model weights

Question 8

A healthcare organization uses an AI system to prioritize patient follow-up calls. The system was trained on historical data that underrepresented certain demographic groups. The governance team wants to implement a control that detects and mitigates bias in the model's outputs on an ongoing basis. Which approach is most effective?

Accepted Answer

Implement continuous monitoring of model outputs with fairness metrics and a defined mitigation workflow.. Continuous monitoring with fairness metrics and a mitigation workflow is the most effective approach because it provides ongoing detection and correction of bias in production. It acknowledges that bias is not a one-time issue and requires sustained vigilance. This control enables the governance team to fulfill its responsibility to ensure equitable AI outcomes and comply with regulations.

Answer

Rely on the vendor's claims that the model is fair and unbiased.

Answer

Remove all demographic data from the training set to prevent bias.

Answer

Conduct a one-time bias audit before deployment and document the results.

Question 9

What is the primary risk of using 'Pre-trained models' from public hubs without verification?

Accepted Answer

The model may contain backdoors or malicious code.. Pre-trained models downloaded from public hubs (e.g., Hugging Face, TensorFlow Hub) are serialized artifacts that may execute arbitrary code on load — Python pickle files, custom layers, or Lambda layers can embed malicious payloads. Without verification (checksums, signatures, model cards, or scanning), an attacker who uploaded a poisoned model can achieve code execution or implant a backdoor that triggers on specific inputs. This is the primary supply-chain risk of consuming untrusted model artifacts.

Answer

The model will not support Python.

Answer

The model will be too small.

Answer

The model will not have a web interface.

Question 10

A retail company's AI governance committee is drafting a policy for employee use of public generative AI tools. The Chief Information Security Officer wants to ensure the policy addresses the most significant risk of data leakage. Which policy element is MOST directly aimed at preventing sensitive corporate data from being exposed through these tools?

Accepted Answer

Prohibiting the input of confidential, personal, or proprietary data into public generative AI tools.. The most direct way to prevent sensitive data leakage through public generative AI tools is to prohibit entering such data into them. This policy element removes the risky behaviour itself, rather than relying on training, access controls, or endpoint monitoring that do not specifically address the content shared with external AI services. It is clear, enforceable, and targeted at the root cause.

Answer

Requiring employees to complete annual security awareness training.

Answer

Mandating multi-factor authentication for all corporate accounts.

Answer

Deploying endpoint detection and response agents on all employee laptops.