Trustworthy Agentic AI in Salesforce SaaS: Product Governance and Risk Controls

Abstract

The integration of agentic Artificial Intelligence (AI) into enterprise Software-as-a-Service (SaaS) platforms such as Salesforce is transforming organizational governance and risk management practices. This study examines how trustworthy agentic AI, Salesforce SaaS capabilities, and product governance collectively influence risk control effectiveness. Employing a mixed-methods design, data were collected from 212 survey respondents and 15 expert interviews across multiple industries. Reliability and validity analyses confirmed the robustness of the measurement constructs, while regression and structural equation modeling revealed that governance maturity is the strongest predictor of effective risk controls. Trustworthy agentic AI significantly contributes to governance quality through features such as transparency, fairness, and robustness, while Salesforce SaaS functions as a strategic enabler by providing secure and scalable infrastructures. Mediation and moderation analyses further highlighted governance as a mediator and compliance adherence as a moderator in enhancing AI-driven risk controls. The findings emphasize that governance should be treated as a strategic enabler rather than a compliance obligation, ensuring ethical, resilient, and sustainable AI adoption in SaaS environments.