Digital Twins Driven by Artificial Intelligence to Mitigate, Detect, and Simulate Virtual Space Cyber Threats
Abstract
Digital Twins (DTs) combined with Artificial Intelligence (AI) provide a powerful approach to mitigating, detecting, and simulating cyber threats in virtual spaces. By creating real-time virtual replicas of digital environments, AI-driven DTs enhance cybersecurity resilience and response strategies. However, existing methods struggle with limited adaptability to emerging threats, high computational overhead, and inefficient anomaly detection, making real-time mitigation challenging. These limitations compromise the accuracy and effectiveness of cybersecurity defenses. To address these challenges, we propose AI-Driven Digital Twin Cyber Threat Modeling (AI-DTCTM), which integrates machine learning, real-time monitoring, and predictive analytics. This framework continuously learns from cyber incidents, enhances threat detection accuracy, and optimizes decision-making for proactive mitigation. The proposed AI-DTCTM method is employed for early threat detection, automated risk assessment, and real-time security simulations, ensuring dynamic adaptation to evolving cyber threats in virtual environments. It also enhances cyber resilience by reducing false alarms and improving response times. The proposed method achieves the threat detection accuracy by 98.46%, false positives by 45.3% and response efficiency by 97.64%.