Real-Time AI-Driven Cybersecurity Analytics Dashboards for Critical-Infrastructure Protection under Saudi Vision 2030: An NCA- and SDAIA-Aligned Governance-to-Execution Framework

Abstract

Background. Saudi Arabia’s Vision 2030 has moved from strategic planning into execution, where artificial intelligence, data governance and cybersecurity are interdependent national capabilities. The paper addresses how real-time cybersecurity analytics dashboards can connect to telemetry, AI detection, compliance evidence and executive decision-making without collapsing security operations, governance and policy oversight into one visual layer. Objectives. The objective is to develop a PRISMA-informed design-science framework for real-time AI-driven cybersecurity analytics dashboards aligned with Saudi Arabia’s NCA, SDAIA, NDMO, PDPL and Vision 2030 priorities. Methods. A narrative evidence synthesis was conducted using academic databases, official policy/regulatory artefacts and selected industry threat reports. The article is positioned as design-science framework development, not as empirical performance evaluation. Search strings, search dates, inclusion/exclusion rules, evidence classes and quality appraisal are documented. Screening and coding were conducted by one reviewer; inter-rater reliability is therefore not claimed and is treated as a limitation. Results. The paper contributes to a dashboard typology, five-layer architecture, sectoral applicability matrix, operational threat-to-control mapping, KPI dictionary, Responsible-AI matrix and indicative 180-day pilot pathway. A final reference-base strengthening added thirteen additional sources on SOC maturity, SIEM/security analytics, cyber-resilience, critical infrastructure, AI assurance and cybersecurity governance. Conclusion. Real-time dashboards may support cyber-resilience only where telemetry coverage, model governance, human oversight, compliance evidence, and response workflows are implemented and validated. The Saudi control mapping is context-specific; the layered architecture is transferable if localized.