Advanced in AI Security Management Free Sample Questions

The AI Steering Committee acts as a bridge between the Board of Directors and operational management. Its primary role regarding risk is to translate the high-level enterprise risk appetite set by the Board into actionable AI-specific risk tolerance levels that guide project approval and control implementation.

The 'Map' function in NIST AI RMF is focused on establishing context. Characterizing the intended purpose, users, and deployment environment is the foundational step to identifying risks. Without this context, downstream risks cannot be accurately mapped.

In Federated Learning, sending raw gradients to a server can still leak information about the training data (model inversion/inference attacks). To prevent this at the aggregation step, Secure Aggregation (often using SMPC) or adding Differential Privacy (noise) to the updates is required to ensure the server sees only the aggregate update, not individual contributions.

The most critical risk with public GenAI tools is the training of the provider's models on user data. An AUP must explicitly prohibit entering sensitive data (PII, IP, secrets) into public tools, while allowing use for non-sensitive tasks to balance security and productivity.

Under the EU AI Act, systems used in employment, workers management, and access to self-employment (e.g., CV screening software) are classified as High Risk due to their potential impact on fundamental rights.

AI systems evaluating creditworthiness of natural persons or establishing their credit score are High Risk as they determine access to essential private services.

This describes an Evasion Attack (often using adversarial examples). The attacker manipulates the input data (inference time) with perturbations (noise) to cause the model to make a prediction error. It does not tamper with the training data itself.

A Model Card (introduced by Google researchers and widely adopted) is a standardized document that details performance characteristics, intended use, limitations, and ethical considerations of an AI model.

Cryptographic signing of model artifacts (e.g., using Sigstore or similar tools) ensures integrity and authenticity. Validating the signature before loading prevents the execution of tampered models or models from untrusted sources, directly addressing the risk of supply chain injection attacks.

False. Explainability is critical for security forensics. If an AI system behaves maliciously (e.g., due to an adversarial attack or poisoning), security analysts need explainability tools (like SHAP or LIME) to understand why the model made that decision and identify the root cause of the breach.

Prompt Injection (specifically direct injection or jailbreaking) involves manipulating the input to override the model's original instructions or safety constraints. This allows the attacker to control the model's output.