2026/01/01 by Shengjie Xu, Yi Qian
Computer Science · Social Sciences · #Access Control and Trust #Adversarial Robustness in Machine Learning #Security and Verification in Computing
paper · doi:10.1109/mcom.001.2500749
openalex publication_date 2026/01/01 · openalex created_date 2026/04/04 · openalex updated_date 2026/06/11
Edge AI brings intelligence closer to data sources, enabling real-time decision-making in critical applications. However, the distributed and resource-constrained nature of edge environments creates unique security challenges that traditional perimeter-based defenses cannot adequately address. This paper presents a Zero Trust security framework specifically designed for protecting distributed intelligence in Edge AI deployments. We first analyze concrete attack scenarios that demonstrate the limitations of traditional security approaches in distributed AI. We then examine the Edge AI security workflow to establish the operational context and requirements. Next, we propose a three-layer security framework (Cloud, Edge, Device) featuring scalable identity management, context-aware policy enforcement, and secure deployment lifecycle components tailored explicitly for protecting distributed intelligence. Our framework addresses the complete Edge AI security workflow. We also present a comprehensive policy framework enabling dynamic access control and real-time threat response for distributed AI. Finally, we discuss open challenges and future directions in securing distributed intelligence systems.