Evaluating Multilingual Prompting with Consumer-Facing Generative World Models for Crisis Response Preparedness
Year: 2026
Type: preprint
Abstract: Generative world models are emerging as interactive systems capable of producing and modifying simulated environments in response to user instructions. Their potential applications include disaster response preparedness, where interactive scenarios may support training and decision-making exercises for, for example, emergency medical teams. Building on previous work on AI-generated disaster simulation scenarios and multilingual synthetic facilitator companions, this study proposes a preliminary evaluation of consumer-facing generative world models. The findings reveal some promising capabilities, but also highlight persistent limitations in multilingual consistency across platforms and scenarios.
Author: Samo, Giuseppe
Institution: Beijing Language and Culture University
Language: English
Topic: Speech and dialogue systems
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7225136790