
Artificial Intelligence
PeaceRep’s approach to artificial intelligence (AI) in peace research
At PeaceRep, we use Artificial Intelligence (AI) to enhance our peace and conflict research while maintaining rigorous academic standards and ethical practices. Our approach combines technological innovation with deep domain expertise in peace processes.
We recognise both the potential and limitations of AI in peace research and practice. Our experience has taught us that AI tools are most effective when:
- Combined with human expertise and verification
- Designed with cultural and contextual awareness
- Focused on augmenting rather than replacing human judgement
- Built on transparent and explainable methodologies
Our Principles
We believe in leveraging AI responsibly and ethically in peace research and practice. We focus on proven, explainable techniques; rigorous verification processes; and open-source sharing of methods. Our approach is guided by these core principles:
Human-centred design: while we embrace AI’s capabilities as a support tool, we maintain human oversight and expertise at every stage. While AI succeeds at processing large volumes of information, human insight remains crucial for understanding nuance, context and implications. Our ‘human-in-the-loop’ approach ensures that the tools enhance, rather than replace human judgement.
Transparency and reproducibility: we are committed to making our methodologies, code and experiments open-source, enabling others to re-use, build upon and verify our work.
Model sensitivities: we recognise and actively try to address potential bias in AI systems, particularly regarding Global South perspectives and local contexts.
Interdisciplinary collaboration: we bring together peace researchers, technology experts and local stakeholders to ensure our AI initiatives are both technically sound and contextually appropriate.
Our Projects
Advanced Text Analysis: Using named entity recognition and natural language processing to identify key actors and relationships in peace-related documents.
Enhanced Coding Efficiency: Leveraging semantic similarity analysis and language models to accelerate our research coding processes, specifically for identifying key categories and issues in peace agreements, as well as extracting information on mediation efforts and third party involvement.
Thematic Analysis: Employing AI-powered topic classification and concept mapping to track evolving themes in peace processes.
Agreement Translation: Using AI-assisted translation capabilities to help discover peace agreements across multiple languages.
For more information on our PeaceTech work using data and peace analytics to support peace and transition processes, visit the resources below.