Modelling Trust and Trusted Systems: A Category Theoretic Approach

Abstract

We introduces a category-theoretic framework for modelling trust as applied to trusted computation systems and remote attestation. By formalizing elements, claims, results, and decisions as objects within a category, and the processes of attestation, verification, and decision-making as morphisms, the framework provides a rigorous approach to understanding trust establishment and provides a well-defined semantics for terms such as `trustworthiness' and 'justification'/forensics. The trust decision space is formalized using a Heyting Algebra, allowing nuanced trust levels that extend beyond binary trusted/untrusted states. We then present additional structures and in particular utilise exponentiation in a category theoretic sense to define compositions of attestation operations and provide the basis of a measurement for the expressibility of an attestation environment. We present a number of worked examples including boot-run-shutdown sequences, Evil Maid attacks and the specification of an attestation environment based upon this model. We then address challenges in modelling dynamic and larger systems made of multiple compositions.