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

TL;DR

This paper investigates how GenAI chatbots can tailor hurricane preparedness information for multiethnic communities using the CASA paradigm. It reports a between-subjects online experiment with 441 Florida residents, manipulating chatbot tone (formal vs informal) and cultural tailoring (tailored vs generic) via GPT-4; outcomes include perceived friendliness and credibility, plus information seeking, sharing, and disaster preparedness. Key findings show that formal tone boosts credibility while informal tone increases friendliness, with credibility driving all preparedness outcomes and friendliness driving information sharing; cultural tailoring raises perceived tailoring and credibility, yielding indirect effects on outcomes, especially for Hispanic and Black participants. The work extends CASA to disaster communication, demonstrates practical potential for inclusive risk communication, and provides open access to the chatbot code for reproducibility and further research.

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.

Figures (2)

• Figure 1: Full Sample Results from SEM.
• Figure 2: The Hispanic and Black Sample Results from SEM.