From Cloud to Edge: The Rise of Intelligent, Distributed AI Systems

M. Zmuda (Intel, Poland)

The rapid evolution of Edge computing is redefining how intelligent systems are designed, deployed, and secured across industrial, automotive, healthcare, and smart infrastructure domains. This presentation explores the latest trends in Edge-class architectures, with a particular focus on the convergence of Artificial Intelligence and Edge computing (Edge AI) as a key enabler of low-latency, privacy-aware, and energy-efficient systems. The talk will examine how modern AI workloads are increasingly moving from centralized cloud environments to the Edge, driven by real-time constraints, data sovereignty requirements, and total cost of ownership considerations. Practical Edge AI application examples will be discussed, including computer vision, predictive maintenance, and intelligent control systems, highlighting architectural patterns and design trade-offs observed in real deployments. Special attention will be given to the opportunities and challenges of Edge AI, such as scalability, model lifecycle management, heterogeneous hardware integration, and power-performance optimization. The presentation will also showcase the role of Intel’s Open Source software ecosystem—including frameworks, toolkits, and reference architectures—in accelerating Edge AI development and enabling interoperability. Finally, the session will address security aspects of Edge and AI systems, covering secure boot, hardware-based trust, data protection, and AI model integrity, emphasizing the need for security-by-design in distributed intelligent systems.