Agentic Digital Twins for Smart Cities

Modern cities are extremely complex environments where emergencies such as fires, crimes, and accidents are increasingly difficult to manage. Each situation requires coordinated actions involving multiple systems: early detection through sensors, rapid notification to firefighters, traffic management by the police, hospital preparation for potential victims, and real-time analysis of evolving risks. Ensuring safety and responsiveness in such interconnected urban ecosystems demands Agentic AI, AI capable of reasoning, deciding, and acting autonomously across distributed infrastructures to support proactive and efficient interventions.

This talk introduces CupCarbon KLAINES, a platform designed to model and simulate this new generation of intelligent urban agents. Built on the CupCarbon digital-twin ecosystem, KLAINES (Knowledge-based Learning Agentic AI for Networked Environments and Systems) enables the creation of autonomous agents that interpret sensor data, coordinate multi-agency responses, and manage safety-critical scenarios across networked city systems. By offering an integrated environment to design, test, and deploy agentic AI workflows, CupCarbon KLAINES provides researchers and practitioners with a practical and scalable tool to enhance real-time decision-making and improve the safety, resilience, and operational efficiency of future smart cities.