Tailoring Generative AI Chatbots for Multiethnic Communities in Disaster Preparedness Communication: Extending the CASA Paradigm

2024-06-12

Journal of Computer and Communications

AI Summary

Multicultural communication barriers impede disaster preparedness among racially diverse communities during hurricanes. Method: We developed three culturally adapted, GPT-4–powered chatbots—targeting Black, Hispanic, and White populations—extending the Computers Are Social Actors (CASA) paradigm to multiethnic disaster preparedness for the first time. Leveraging systematic cultural adaptation—including value embedding and contextualized exemplars—and calibrated formality in tone, we conducted a randomized controlled trial (N = 441) and tested causal mechanisms via structural equation modeling. Contribution/Results: Culturally tailored generative AI significantly enhanced users’ perceived trustworthiness and likeability of the chatbots (p < 0.001), which in turn positively predicted disaster preparedness behavioral intentions. This work establishes a replicable methodological framework and empirically grounded foundation for equity-oriented public emergency communication using generative AI.

Technology Category

• Humans and AI: AI for Accessibility
• Multiagent Systems: Agent Communication
• Natural Language Processing: Safety and Robustness

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search

Abstract

This study is among the first to develop different prototypes of generative artificial intelligence (GenAI) chatbots powered by GPT-4 to communicate hurricane preparedness information to diverse residents. Drawing from the Computers Are Social Actors paradigm and the literature on disaster vulnerability and cultural tailoring, we conducted a between-subjects experiment with 441 Black, Hispanic, and Caucasian residents of Florida. Our results suggest that GenAI chatbots varying in tone formality and cultural tailoring significantly influence perceptions of their friendliness and credibility, which, in turn, relate to hurricane preparedness outcomes. These results highlight the potential of using GenAI chatbots to improve diverse communities’ disaster preparedness.

Problem

• Cross-cultural Communication
• Disaster Preparedness
• Ethnic Communities

Innovation

• GPT-4 Technology
• Cultural Sensitivity
• Disaster Preparedness