For the Risk-informed Early Action Partnership, the discussions were particularly significant. Across multiple sessions, partners demonstrated how AI can strengthen the entire early warning, early action value chain: from risk monitoring and forecasting to anticipatory action, risk communication and decision-making. Their contributions highlighted a shared commitment to ensuring that technological innovation translates into practical impact for vulnerable communities.
Below are some of the key highlights from sessions and partner contributions during AI for Good 2026.
REAP Partners championing AI for the greater good
Global Dialogue on AI Governance: Ensuring AI Serves Humanity
Ahead of the Summit, UN Secretary-General António Guterres opened the first Global Dialogue on Artificial Intelligence Governance in Geneva. His remarks underscored both the immense opportunities and significant risks associated with the rapid expansion of AI. While warning against growing inequalities, concentration of power and threats to public trust, he also pointed to AI's potential to democratize access to expertise, including for farmers, health workers and communities facing climate risks. Importantly, he linked the governance agenda directly to the AI for Good Summit, positioning it as part of a broader UN effort to ensure that AI development remains inclusive, safe and aligned with sustainable development priorities.
AI for Early Warnings for All: From Innovation to Impact
One of the most relevant sessions for REAP partners was the workshop AI for Early Warnings for All: From Innovation to Impact, hosted by the AI Group of the Early Warnings for All initiative, led by pillar leads UNDRR, WMO, ITU, and IFRC. The workshop brought together policymakers, technologists, humanitarian actors and climate experts to showcase how operational AI applications are already strengthening multi-hazard early warning systems around the world. Discussions focused on AI-enhanced forecasting, decision-support systems for anticipatory action, risk communication and end-to-end integration across the warning value cycle. Participants also launched two important resources: the AI Solutions for Early Warning Systems Catalogue (REAP partners are welcome to submit additional solutions) and the report Leveraging AI to Enhance Multi-Hazard Early Warning Systems, providing practical guidance for countries seeking to integrate AI into operational early warning systems.
Resurgence, a long standing REAP partner, summarized its key messages during the AI for EW4All: From Innovation to Impact workshop. Resurgence is working on Pillars 2 and 3 with the AI for Early Warnings for All Group through its Daraja+AI solution.
- AI generated forecasting and early warning remain largely and legitimately in internal verification mode, but are showing growing improvements over incumbent Numerical Weather Prediction (NWP) models.
- The use of generative AI for warning dissemination and communication is also at a similar piloting phase.
- The use of generative AI through text to speech and LLMs to address the structural challenges of illiteracy and secondary / minority language exclusion hold great promise.
These applications could help make warnings more understandable, inclusive and actionable for communities that are often underserved by existing communication systems.
Malawi is one particularly compelling example also showcased during the AI for EW4All: From Innovation to Impact workshop came from Malawi, where partners including the Norwegian Meteorological Institute, ECMWF and Malawi's Department of Climate Change and Meteorological Services collaborated to develop an AI-powered weather prediction capability. The initiative demonstrated how tailored AI tools can strengthen national forecasting systems while building local capacity and ownership. Rather than creating parallel systems, the project focused on enhancing existing national services. This is an approach aligned with broader principles of sustainability and country-led resilience. Malawi’s efforts are funded by among others, CREWS and SOFF, laying the foundations for their vision to ensure anticipatory action further down the value chain.
WMO: Bridging Science, Standards and Innovation
In the session AI in Disaster Resilience: Bridging Science, Standards and Innovation, WMO Secretary-General Celeste Saulo emphasized that AI offers a major opportunity to accelerate progress under Early Warnings for All by making climate intelligence more timely, accurate and accessible. She highlighted how emerging AI forecasting systems can deliver high-quality forecasts with lower computational requirements, potentially democratizing forecasting capabilities for countries with limited resources. At the same time, WMO stressed that AI must be built on strong observational data, scientific infrastructure and trusted national institutions. AI should strengthen National Meteorological and Hydrological Services, not replace them, It should help ensure that no country is left behind in benefiting from lifesaving early warning services.
Partnerships as the Key Enabler
A recurring message throughout the Summit was that successful AI deployment depends on partnerships. Speakers emphasized that governance, infrastructure, financing, technical expertise and local capacity must advance together. The discussions reinforced the importance of collaboration between UN agencies, national governments, research institutions, humanitarian organizations and the private sector to move from promising pilots to sustainable national capabilities. This partnership-based approach is at the heart of both the Early Warnings for All initiative and REAP's work.
Looking Ahead: AI in Service of REAP's Mission
The conversations at AI for Good 2026 made one thing clear: AI has enormous potential to strengthen early warning systems, improve risk-informed decision-making and enable faster, more effective early action. But technology alone will not deliver resilience. Lasting impact requires trusted institutions, country ownership, inclusive partnerships and a relentless focus on turning information into action.
This vision aligns closely with REAP's Strategic Vision, which seeks to ensure that better risk information leads to earlier, more effective action that protects lives, livelihoods and development gains. By helping improve the accuracy, accessibility and usability of risk information, AI can become a powerful enabler of anticipatory action and climate resilience. The contributions of REAP partners at the AI for Good Summit demonstrate that the global community is not only innovating, but it is working together to ensure those innovations reach the people who need them most.
Congratulations to all organizations, countries and REAP partners who contributed to these discussions and showcased practical solutions. Their leadership is helping transform the promise of AI into real-world action, bringing us closer to a future where everyone, everywhere, is protected by effective early warning systems and empowered to act before disaster strikes.
