Development of a persuasive User Experience Research (UXR) Point of View for Explainable Artificial Intelligence (XAI)

Technical explanations alone don't build trust: The paper identifies that existing AI explanation methods (SHAP, LIME, and counterfactual explanations) often use complex technical language that non-expert users struggle to interpret, suggesting that simply making AI "explainable" in technical terms doesn't automatically make it understandable or trustworthy to the people actually using these systems.

UX designers need practical tools, not just technical knowledge: The research recognizes a critical gap where UX professionals lack adequate guidance to design AI explanations, highlighting that responsible AI requires collaboration between technical experts and user experience specialists rather than relying on either discipline alone.

Accessible design is essential for responsible AI adoption: The paper proposes a UX Research Playbook specifically designed to help professionals create "accessible, transparent, and trustworthy AI experiences," emphasizing that responsible AI depends on making explanations understandable to diverse users, not just technical audiences.

User-centered design should drive AI transparency: By advocating for a framework that bridges "technical explainability methods and user centered design," the paper argues that how AI systems are explained to users should be driven by research into what users actually need to understand and trust these systems, rather than what's easiest for developers to implement.