Relationship-based Distributed and Variable Trust Metrics for Secure Computing

In modern distributed computing environments, a diverse array of actors—ranging from users to intelligent edge devices—interact through heterogeneous communication channels with one another and with centralized data centers. A critical aspect of securing such systems lies in the accurate and adaptive evaluation of trust across these actors. While several trust models have been proposed in literature, many adopt static or symmetric assumptions that fail to capture the nuanced dynamics of real-world systems. This paper introduces a relationship-based, distributed trust model that emphasizes four foundational principles: (1) trust between two entities is inherently non-mutual; (2) trust evaluations are actor-specific and shaped by the nature of their roles and interdependencies; (3) trust is contextual and influenced by the content and sensitivity of individual transactions; and (4) trust is not static but evolves over time with ongoing interactions. Through a real-world case study, we demonstrate how these principles manifest and influence trust-based decisions.