ExplainitAI: When do we trust artificial intelligence? The influence of content and explainability in a cross-cultural comparison

Cultural context significantly shapes AI trust: Korean participants demonstrated substantially higher trust in AI chatbots and more positive user experiences compared to German participants, suggesting that AI systems cannot be designed with a "one-size-fits-all" approach and must account for cultural differences in attitudes toward technology.

Topic matters as much as the technology itself: Participants reported notably lower trust when AI addressed socially contested issues like migration compared to neutral domains like health or entertainment, indicating that the same AI system will be perceived differently depending on what it's discussing—a critical consideration for deployment in sensitive policy areas.

Explainability alone isn't the solution: The study found that explainability's effectiveness was not uniform but rather depended on interactions between cultural context, topic, and other factors, suggesting that simply adding explanations to AI systems won't automatically build trust across all users and contexts.

Design must be culturally adaptive: The research emphasizes that responsible AI development requires integrating "cultural and contextual nuances" into system design rather than treating all users as having identical needs and expectations, with implications for how companies and governments should approach global AI deployment.