Towards a Praxis for Intercultural Ethics in Explainable AI

Explainability benefits are currently concentrated among technical experts: Most XAI tools are designed for machine learning developers rather than general users, meaning the promised transparency benefits of explainable AI remain inaccessible to the broader population who actually use these systems.

Cultural context fundamentally shapes how people understand AI: The paper argues that cultural nuances significantly impact technology adoption and use, suggesting that one-size-fits-all explanations of AI systems fail to account for diverse social and cultural backgrounds, particularly in the Global South.

Current XAI approaches neglect the "majority world": Despite AI's global reach, limited research addresses the needs of users in low-resource regions and culturally diverse areas, leaving a significant gap between where XAI is developed (primarily in wealthy nations) and where it's increasingly deployed.

An intercultural ethics framework can improve AI understanding: By integrating cultural awareness into how AI systems are explained, developers can create explanations that resonate with diverse users' values and knowledge systems, leading to better comprehension and more effective use of these technologies.

Democratizing AI explainability is essential for responsible AI: Making XAI accessible across cultural and socioeconomic boundaries is not just an equity issue—it's necessary for ensuring that AI systems are actually understood and appropriately used by the communities they affect most.