Opening Remarks - Workshop on Resilience to Natural Hazards through AI Solutions

Dear delegates, partners and colleagues,

I am very grateful to Hans Das, the Deputy Director-General for European Civil Protection and Humanitarian Aid Operations, for the words of welcome and for hosting this event today. We are truly thankful for all your support and leadership.  

We gather here today as floods wash away livelihoods; heatwaves threaten health and food security; storms leave entire communities in the dark.

Hundreds of millions of people are affected worldwide. We have seen this most recently – and tragically – with the deadly flooding in Indonesia and Sri Lanka which were hit by rare and devastating tropical cyclones. Early warnings were indeed issued, but communities were still caught unaware and unprepared.

Rising global temperatures are turbo-charging more extreme weather. The frequency and severity of natural hazards are increasing.  

It is our duty to ensure that natural hazards do not turn into devastating outcomes.  

And Artificial Intelligence is a powerful ally in helping us achieve this and building global resilience.  

AI is transforming how we translate science into action. It has the potential to revolutionize early warning systems so they reach every person, everywhere.

But this transformation is not only about technology — it is about trust, inclusion, and solidarity.  

We are here today at this workshop on resilience to natural hazards through AI solutions because we share the conviction that AI, used responsibly, can support us in emergency and crisis management.  

We must harness its power to ensure that the intelligence we create — both human and artificial — becomes a bridge between science and society, between data and decisions, and ultimately, between knowledge and action.

At the Extraordinary World Meteorological Congress this year, 115 Member States endorsed the Call to All Stakeholders on AI and Machine Learning. This call anchors how we approach every AI partnership — grounded in ethics, inclusion, and the single authoritative voice of national meteorological and hydrological services.

We are already witnessing the unlocked potential of AI to accelerate Early Warnings for All.

Allow me to share several examples of how WMO and its partners are using AI to turn scientific progress into tangible resilience.

Through our mandates, we are laying the groundwork for the systematic, safe and interoperable integration of AI and machine-learning models into global, regional and national forecasting chains.

In Nigeria, Viet Nam, Uruguay, and the Czech Republic, AI-powered flood forecasting pilots are helping identify early signals that might otherwise have gone unnoticed. This gives local authorities and communities critical time to act.  

In Malawi, a pilot developed with MET Norway and built on ECMWF’s “Forecast-in-a-Box” concept allows local meteorologists to run advanced AI models using shared cloud resources. This is a major step forward for small island and least developed states that lack supercomputing capacity.

Across Africa, the Caribbean, and the Pacific, our Regional Climate Centres are using AI for sub-seasonal prediction — improving early warning lead times and helping communities prepare before hazards strike.  

Under the Agricultural Innovation Mechanism (AIM for Scale) programme, WMO is contributing to delivering tailored AI-powered services to smallholder farmers in 30 countries by 2027. We are also working with national partners to utilize AI for renewable energy forecasts in Latin America.  

WMO is proud to be a co-leader of the Global Initiative on Resilience to Natural Hazards through AI Solutions. We are committed to working with ITU, UNEP, UNESCO, UPU, and UNFCCC.  

Through this collaboration, we are shaping international standards and governance frameworks for trustworthy, safe, and interoperable AI in weather, climate and water services.  

In this respect WMO has a particular interest in the impressive progress made at EU level through the Destination Earth initiative which aims at developing digital twins of the Earth, building on the outstanding expertise of its entrusted entities, including ECMWF and EUMETSAT.  

Together with UNDRR, IFRC and ITU, we co-lead the AI for Early Warnings 4 All Sub-Group, which brings together a wide community - including major technology partners, academia, NGOs, and international organizations. These efforts are identifying practical, scalable AI applications that strengthen early warnings across all four EW4All pillars.  

A flagship report on AI for Early Warnings will be launched at the AI for Good Global Summit in July 2026. It will showcase how AI can be a tool to translate country needs into actionable use cases, and catalyze collaboration across governments, UN agencies, industry, academia, and civil society.  

And through our collaboration with ITU, UNCCD (United Nations Convention to Combat Desertification), UNDP, and CERN under the Working Group on Digital Transformation for Hydrology and Water Resources, we are expanding our horizon beyond AI. We are exploring digital twins, quantum computing, and other emerging technologies with the potential to transform hydrological and water resource services worldwide.  

In this context, today we have a workshop organized by my colleagues showcasing some of the AI applications around flood management and I strongly encourage you to join and contribute to these discussions.

As we work to achieve Early Warnings for All by 2027 we must act together to ensure that AI reduces rather than widen global inequalities.

Ladies and gentlemen, colleagues, friends

Today, I call for three commitments:

• First, Science is our lifeline. Let us ensure that the standards we apply in AI integration are rooted in science, transparency, and equity, to accelerate co-innovation.  
• Second, let us accelerate support to the most vulnerable, ensuring that they can adopt, adapt and sustain early warning systems  
• Third, let us strengthen a united approach, as resilience is built best in a system working as one. Let us make this era of artificial intelligence an era of collective intelligence — where foresight becomes action, and action becomes resilience.

Thank you.