Turning Student–AI Interactions into Accreditation-Ready Evidence

Session Abstract: Accreditation and quality assurance (QA) often rely on outcome data such as grades and surveys. While useful, these measures give little insight into how students reason and decide. This poster presents a practical way to capture process evidence through AI-supported digital twin interactions in a Business Ethics course.

Students co-constructed ethical decision trees with AI learning companions, while a comparison group received instructor-designed trees. Interaction data—dialogues and decision tree artifacts—were reviewed with three simple criteria: stakeholder inclusion, decision path clarity, and reasoning evidence. These became process indicators that can be integrated into Assurance of Learning (AoL) reports, accreditation documentation, and course improvement.

The poster shows examples and a simple flow of how interaction data become accreditation-ready evidence. With generative AI rapidly expanding, this approach offers IR professionals timely new options for QA.

Keywords: Digital Twin Interactions, Accreditation, Quality Assurance, Assurance of Learning, Business Ethics