AI-Driven Analysis of Spanish-Language Crisis Discourse: Lessons from the 2025 Iberian Peninsula Blackout

Erokhin, D. ORCID: https://orcid.org/0000-0002-5191-0579 & Komendantova, N. ORCID: https://orcid.org/0000-0003-2568-6179 (2025). AI-Driven Analysis of Spanish-Language Crisis Discourse: Lessons from the 2025 Iberian Peninsula Blackout. In: EUI-ESM Conference On Crisis Communication, 28 November 2025, Florence, Italy.

Preview

Text AI-Driven Analysis of Spanish-Language Crisis Discourse.pdf - Presentation Available under License Creative Commons Attribution Non-commercial. Download (1MB) | Preview

Abstract

This presentation analyzes Spanish-language discourse on YouTube in the immediate aftermath of the unprecedented Iberian Peninsula blackout of April 28, 2025. Based on 76,398 comments from 360 relevant videos, we employed the OpenAI GPT-4o-mini model to explore emotional responses, misinformation, and civic engagement. The findings reveal a discourse dominated by anger (43% of comments) and a pervasive spread of unverified claims, present in 46% of comments. Blame was overwhelmingly directed at government and political failures, reflecting deep institutional distrust. Although one in five comments included a call to action, few offered constructive solutions beyond demands for infrastructure reform. These results underscore the critical need for multilingual, real-time crisis communication and highlight the distinct information challenges for Spanish-speaking communities during emergencies. The research contributes to our understanding of digital resilience and inclusive public policy in the face of systemic crises.

Actions (login required)

View Item