SRI Session: Integration of AI in Early Warnings for All: Addressing Equity, Trust, and Local Knowledge Systems

As climate extremes intensify, early warning systems are evolving from single-hazard to multi-hazard platforms. The UN/WMO Early Warnings for All initiative aims to protect every person on Earth by 2027—and artificial intelligence is reshaping how this happens.

Yet AI integration raises critical questions: Who accesses warnings? How are they interpreted? What knowledge counts in decision-making?

This webinar, organized by the World Weather Research Programme for the Sustainability, Resilience and Innovation Congress addresses two interconnected challenges:

1. Ethical AI Integration: Ensuring algorithmic transparency, mitigating dataset bias, and maintaining accountability when AI outputs fail. With AI-generated misinformation proliferating, how do we build public trust in AI-enhanced warnings while guaranteeing equitable access?
2. Preventing Marginalization: How do we bridge AI forecasting with Indigenous knowledge and community observations? Accelerated AI adoption risks digital divides that exclude vulnerable populations, particularly where technology ignores local context, language, and capacity.

Join us for research and case studies on collaborative approaches that balance technological sophistication with social equity—advancing equitable early warning capabilities for a multi-hazard future.

The line up of speakers for this session is as follows, 

• Katharine Vincent, Kulima Consulting, South Africa
• Ricardo Branco, Brazilian National Secretariat for Protection and Civil Defense, Brazil
• Peter Tuju, Intergovernmental Panel on Climate Change (IPCC)
• Gilber Siame, University of Zambia, Zambia 
• Abubakr Salih Babiker, WMO Region I (Africa) Office