The Integration of Trustworthy AI Values: A Comprehensive Model for Governance, Risk, and Compliance in Audit Architecture Framework context

Abstract

The Trustworthy AI-compliant Governance, Risk, Compliance Architecture (GRC Architecture) as outlined in this paper has the objective of analyzing the performance of a firm vis-àvis all the three facets. In other words, the system, analyzing the data from multiple sources monitor the GRC aspects of an organization in real time and report it to the stakeholders as to where the firm lacks performance/ risk/ compliance measures thereby enabling them to take corrective measures in time. The system is a supervised ML system which has sufficient human intervention and oversight. It is built on features like robustness and safety; the AI system has an in-built model control faculty that would take care of wrong actions by not letting them manifest, by classifying AI activity into multiple classes in which super-critical functions wherein a potential misstep can cause business disruptions; human approval is necessary there. Complying with all the laws and regulations governing the respective domains, the GRC Architecture is ISOcertified and adheres to all the privacy and data governance laws. It is transparent: all its actions are well accounted for. To ensure diversity, non-discrimination, and fairness, a special model is employed. It considers societal and environmental wellbeing in that paperless electronic digital system ensures a reduced carbon footprint contributing to infinite intangible benefits hidden prima facie, as the GRC
aspects of many a firm are taken care of. In this research paper, a Governance Risk and Control system architecture is shown to adhere to these parameters of Trustworthy AI