Proceedings of the International scientific and practical conference ―Science, Technology and Culture in the Era of Globalization‖ (December 24-26, 2025) / Publisher website: www.naukainfo.com. – Geneva, Switzerland, 2026. – 234 p.

200 The implementation of blockchain-based technologies adds transparency to energy transactions, increasing trust and the accuracy of anomaly detection. Federated learning plays a central role in this context, as it allows devices to exchange model updates rather than raw data. This approach significantly improves learning efficiency and protects privacy, which is especially important in urban environments with stringent information security requirements. Keeping local data on devices reduces the risk of personal information leaks while simultaneously contributing to the continuous refinement of global algorithms. In order to ensure reliable and fast exchange of information between devices, specialized communication protocols are used, in particular MQTT [5]. They are developed to operate under resource-constrained conditions and support effective interaction in real time. This allows energy management systems to quickly respond to changes in the grid, optimizing energy distribution and maintaining operational stability. Thus, the integration of Collaborative Edge AI into decentralized energy grids creates the preconditions for improving the efficiency, reliability, and security of energy supply. It facilitates better utilization of local energy sources, reduces losses, enhances system resilience, and maintains high data protection standards. This approach meets modern requirements for sustainable development and energy security, opening new opportunities for innovation in the energy sector. After examining the basic mechanisms and communication protocols of Collaborative Edge AI (CEAI) in decentralized energy systems, it is important to highlight diversity of its practical applications, which significantly improve the efficiency and reliability of energy networks. CEAI not only facilitates load balancing and privacy protection but also becomes the foundation for intelligent process optimization in various energy sectors—from production and storage to distribution and consumption. In renewable energy systems, CEAI uses machine learning algorithms to forecast both energy production and consumption. This enables flexible regulation of the supply-demand balance, ensuring grid stability even during significant