Artificial intelligence methods for enhancing cybersecurity in Oman: a comprehensive review

Abstract The adoption of artificial intelligence (AI) in cybersecurity has become critical as cyber threats become more frequent and sophisticated. However, translating global AI-based cybersecurity approaches into effective national practices remains challenging, particularly in emerging digital economies. This study systematically examines artificial intelligence–based methods for enhancing cybersecurity capabilities in Oman, while critically identifying the technical, organizational, and regulatory challenges constraining their effective adoption. A convergent mixed-methods approach was employed, integrating survey data from 138 cybersecurity practitioners with semi-structured interviews involving 15 experts from government, banking, energy, healthcare, and education sectors. Grounded in the Technology–Organization–Environment framework and Diffusion of Innovations theory, the analysis examined adoption drivers, barriers, and policy implications. Quantitative results show a strong relationship between AI adoption factors and cybersecurity effectiveness (R 2 = 0.78). While institutional awareness of AI is high (mean = 4.22 on a five-point scale), realized benefits remain lower (mean = 3.85), indicating implementation gaps. Anomaly detection (73%), Arabic natural language processing (67%), and explainable AI (60%) emerged as the most contextually relevant techniques. Major barriers identified include skills shortages (87%), legacy infrastructure (73%), cost constraints (73%), and regulatory ambiguity (67%). Based on these findings, the study proposes the Oman Cybersecurity AI Framework (OCAIF), a policy-oriented framework that supports phased, context-sensitive AI adoption. The framework integrates technological readiness, workforce development, ethical governance, and sector-specific priorities, offering an empirically grounded roadmap for strengthening cybersecurity resilience in Oman and comparable Gulf Cooperation Council contexts.