Enhancing cybersecurity with Explainable Artificial Intelligence: technical framework and applications in training labs
This paper discusses how explanation outputs can be operationalized in cybersecurity training labs through auditable “rationale artifacts,” while clarifying that any observed reduction in false positives should be interpreted as the outcome of explanation-guided interventions (e.g., threshold tuning and triage rule adj...