AAISM emphasizes that when introducing innovative AI systems, organizations reduce security and compliance risk by following a phased adoption approach. This allows incremental deployment, controlled testing, and gradual scaling while monitoring risks in real time. Re-evaluating risk appetite and evaluating compliance are important governance steps but do not directly mitigate risks during implementation. Seeking third-party advice can add expertise but does not provide the structured control that phased integration offers.
The most effective risk management approach for AI innovation is to adopt a phased rollout strategy.
References:
AAISM Exam Content Outline - AI Risk Management (Innovation and Risk Control) AI Security Management Study Guide - Phased Implementation Strategies

Prompt hacking manipulates large language models by injecting adversarial instructions into inputs to bypass or override safeguards. The AAISM framework identifies adversarial testing as the most effective way to simulate such manipulative attempts, expose vulnerabilities, and improve the resilience of controls. Load testing evaluates performance, input testing checks format validation, and regression testing validates functionality after changes. None of these directly address the manipulation of natural language inputs.
Adversarial testing is therefore the correct approach to mitigate prompt hacking risks.
References:
AAISM Exam Content Outline - AI Risk Management (Testing and Assurance Practices) AI Security Management Study Guide - Adversarial Testing Against Prompt Manipulation

AAISM technical coverage identifies recall as the metric that specifically measures a model's ability to capture all true positive cases out of the total actual positives. A high recall means the system minimizes false negatives, ensuring that relevant instances are not overlooked. Precision instead measures correctness among predicted positives, specificity focuses on true negatives, and the F1 score balances precision and recall but does not by itself indicate the completeness of capturing positives. The official study guide defines recall as the most direct metric for evaluating how well a model identifies all relevant positive cases, making it the correct answer.
References:
AAISM Study Guide - AI Technologies and Controls (Evaluation Metrics and Model Performance) ISACA AI Security Management - Model Accuracy and Completeness Assessments

When setting RPOs and RTOs for AI systems, especiallygenerative AI, thecritical factor is the restoration of training data and model artifacts within the recovery window. Without this, restored systems may function inaccurately or incompletely, undermining business continuity.
AAISM risk management principles emphasize:
* Recovery objectives must align withdata protection requirementsfor both training and inference data.
* The ability to restorelarge-scale training datasetsis primary, since downtime without them leads to operational and compliance risks.
* Computational efficiency and hardware consistency are secondary considerations, but not the primary drivers of RPO/RTO definitions.
Thus, ensuring backup and restore capabilities of training datasets directly within RTO is theprimary requirement.

The AAISM technical controls framework emphasizes data validation as the primary safeguard against data poisoning attacks. Poisoning occurs when attackers insert malicious or corrupted data into training sets.
Validation techniques verify the quality, authenticity, and consistency of input data before training, preventing compromised samples from corrupting the model. Restoration helps after compromise, watermarking protects ownership, and intrusion detection monitors networks rather than data quality. The most effective preventive measure is data validation.
References:
AAISM Study Guide - AI Technologies and Controls (Data Poisoning Mitigation) ISACA AI Security Management - Data Validation and Quality Controls