An AI-supported learning simulation for Climate, Peace and Security learning

Same problem – different analytical starting points  

When a drought happens, different practitioners may see different dimensions of the same crisis. A climate specialist may examine rainfall trends, a peacebuilding practitioner may look at tensions over water access, and a programme officer may consider how to protect livelihoods. In Climate, Peace and Security (CPS) analysis, all these perspectives need to come together: understanding a climate hazard from one point of view is not enough to explain how it might affect peace and security. 

This creates a challenge for learning design. How can practitioners, who come from different backgrounds and bring in different frameworks and levels of understanding, learn to connect different types of evidence and expertise as they analyze an event?  

At UNSSC’s Peace and Security Hub, this question has led to the development of an AI-supported CPS learning simulation, currently in the prototype stage. Consultations with 17 climate practitioners highlighted recurring difficulties: integrating fragmented information, explaining how climate pressures interact with governance and social conditions, and identifying opportunities for prevention and peacebuilding. Professionals need learning opportunities to practice connecting these dots and receive support where they encounter difficulties. 

A shared case for practicing integrated analysis  

Drawing on the Climate Security Mechanism Toolbox, the team structured the learning journey around climate risk identification, climate-security risk analysis, policy implications and response design. The current prototype focuses on the first two stages, using a fictional case for UN practitioners to explore how climate pressures interact with livelihoods, governance and security. 

Through a series of connected screens, learners encounter new information as the scenario unfolds. They explore different sources in the “Data Discovery Dashboards,” record supporting evidence and develop their analysis in the “Climate–Security Pathway Builder,” and use the “AI Analysis Coach” to reflect on and refine their reasoning. 

For example, declining water availability may threaten agricultural livelihoods, but this does not automatically imply conflict. Learners must consider who depends on the affected resources, how access is managed, and whether existing community or institutional capacities could reduce tensions. They then explain how these factors could contribute to insecurity or create opportunities for cooperation, drawing on evidence from the case. As new information emerges, learners revisit their conclusions and identify what remains unknown. 

The exercise gives practitioners a structured way to connect climate information with social and governance conditions, while making clear which conclusions are supported by evidence and which require further investigation. 

AI as a learning companion 

Within the same exercise, learners may need different kinds of support. Some may need a definition of “vulnerability”; others may need help connecting climate pressures with governance conditions. The “AI Analysis Coach” is designed to respond to their questions and entries with explanations, prompts and feedback. It has two connected roles: on the one hand, it helps learners develop their analysis; on the other hand, it gives them an opportunity to critically evaluate AI-generated advice. 

For example, a learner might suggest that water scarcity could increase tensions between farming communities. As a learning aid, the AI Analysis Coach could ask whether local arrangements for sharing water might reduce those tensions, helping the learner consider a governance perspective. As a subject of critical reflection, the AI’s advice would then be checked against the case materials to see whether it fits the situation. This way, learners can practice both broadening their analysis and checking AI advice against evidence. 

Bringing perspectives together  

Each perspective brings a valuable piece of the puzzle. The aim of this initiative is to offer a shared space to connect those perspectives, using AI to support learners to develop and question their analyses. The next step will be piloting the prototype with practitioners to assess how effectively this approach meets their learning needs.