"Re-Tell the Fortune so I Can Believe It": How Chinese User Communities Engage with and Interpret GenAI-based Fortune-Telling

Why Read This from an HCI Perspective

This article shows how GenAI works not just as a predictive tool, but as a cultural interface for handling uncertainty. In particular, interaction patterns such as providing personal emotional reassurance, strengthening interpretation within a community, and repeating queries offer direct insights for UX design and trust-formation research. For HCI practitioners and researchers, it is meaningful to understand usage contexts that cannot be explained by ‘accuracy’ alone.

CIT's Commentary

What’s interesting is that this case explains AI trust not merely through information quality, but through a combination of rituality, social validation, and emotion regulation. In other words, users are looking for not only whether the answer is ‘right,’ but also a ‘trustworthy atmosphere’ and a ‘structure they can interpret together.’ This is a pattern that can be applied directly to many culturally oriented AI services. However, the study clearly reveals that repeated queries can strengthen confirmation bias, and that the AI may fail to refuse high-risk questions. Therefore, future design should address both expressive immersion and safety mechanisms. In particular, culturally aligned personas, meaningful delays, and context-aware refusal policies appear to be practical design directions.

Questions to Consider While Reading

• Q.In GenAI-based fortune-telling, what interaction elements most strongly shape the ‘rituality’ users feel?
• Q.What kind of design is needed to maintain emotional reassurance while preventing repeated queries and community validation from strengthening confirmation bias?
• Q.When implementing culturally aligned LLM responses alongside safe refusal policies, what criteria are appropriate for determining which questions are high-risk?