XAI for All: Can Large Language Models Simplify Explainable AI?

XAI remains inaccessible to non-experts: The paper identifies that current Explainable AI methods are primarily designed for users with strong technical backgrounds, creating a barrier for business professionals and other non-technical stakeholders who need to understand and trust AI systems.

LLMs can translate complex AI concepts: The study demonstrates that custom Large Language Models like "x-[plAIn]" can effectively generate clear, audience-appropriate summaries of various XAI methods, suggesting that AI itself can help democratize understanding of AI.

Audience-tailored explanations improve comprehension: A key finding is that the model's ability to adapt explanations to match different knowledge levels and interests—whether for business professionals or academics—significantly improves accessibility and practical usability of XAI insights.

Bridging the gap between theory and practice: The research shows that making XAI more understandable facilitates better decision-making by end users, helping organizations move from complex technical documentation to actionable insights that non-experts can actually use.

Promising scalability for diverse users: Use-case study results indicate this approach works across different XAI methods and audience groups, suggesting a scalable solution for bringing explainable AI to a wider range of stakeholders beyond just technical teams.