CRatE researches the role of artificial intelligence and machine learning to design a self-adapting system for predictive cyber risk analytics that can form an automatic anomaly detection system. The team aims to develop a higher Technology Readiness Level (TRL) for dynamic analytics of cyber-attack threat event frequencies, that enable predicting the cyber risk loss magnitude.
Lack of probabilistic data leads to qualitative cyber risk assessment approaches. This leads to speculative assumption. Quantitative risk impact estimation based on real-time data is needed for making decisions on cybersecurity, cyber risk and cyber insurance. Without dynamic real-time risk data and cyber risk analytics enhanced with artificial intelligence and machine learning cognition and data collection mechanisms, these estimations can be outdated and imprecise.
CRatE is using a series of red teaming events to understand how users and autonomous systems interact to ensure system resilience and how to stress-test such systems for complex attacks. The team is developing a demonstration project that would exhibit how such integration for dynamic real-time cyber risk analytics would work in practice.