News & Events
Milestone 2 Achieved: LLM4CIP Threat Model Successfully Delivered
The LLM4CIP consortium is proud to announce the successful completion of Milestone 2: LLM4CIP Threat Model, marked by the submission of Deliverable D2.1 – “LLM4CIP Stakeholders Cyber-Defence Capability Mapping, Threat Model and Penetration Testing Definition.”
This milestone represents a key achievement in the early stages of the project, establishing the foundation for the development of the LLM4CIP platform and its advanced cybersecurity capabilities. According to the project plan, the threat model serves as the basis for automated penetration testing, risk assessment, cyber-range activities, and AI-driven cybersecurity services that will be developed throughout the project.
The threat model was developed following an extensive analysis of cyber-defence capabilities, threats, attacks, weaknesses, and vulnerabilities affecting critical infrastructures and essential services covered by the NIS2 Directive. The work examined a broad range of environments, including networks, applications, virtualisation environments, cloud infrastructures, Industrial Control Systems (ICS), Internet of Things (IoT) ecosystems. The findings provide a comprehensive understanding of the threat landscape and support the creation of realistic penetration testing scenarios and future cyber-range exercises.
The LLM4CIP Threat Model will directly support the development of several key project innovations, including AI-driven attack scenario generation, automated penetration testing services, dynamic risk assessment capabilities, cyber digital twin environments for critical infrastructures, and LLM-powered cybersecurity assistants and decision-support tools. These technologies will help organisations proactively identify vulnerabilities, assess cyber risks, and strengthen their preparedness against increasingly sophisticated cyber threats.
The successful completion of Milestone 2 paves the way for the next phase of the project, focusing on the definition of user and system requirements, technical specifications, KPIs, and the overall LLM4CIP platform architecture. We congratulate all partners involved in achieving this important milestone and look forward to delivering the next generation of AI-powered cybersecurity solutions for critical infrastructure protection.