My colleagues Katherine O’Lone, Michael Gantley, Justin Lane, and I have just had an article published in the New England Journal of Public Policy titled “Employing Multi-Agent AI to Model Conflict and Cooperation in Northern Ireland.”

Here is a link to the website, and here is a direct link to a PDF copy of the article.

Abstract: In this article, we outline the development of a multi-agent artificial intelligence (MAAI) model for post-conflict Northern Ireland. We discuss the insights it provides into the primary drivers of conflict and cooperation in the post-Agreement era. Analyses reveal that leading drivers of cooperation in the model are fairness and sadness, while the main drivers of conflict are related to anxiety and perceived moral authority. We examine these findings in the context of previous computational modeling efforts in Northern Ireland, the social psychological literature on intergroup conflict, and the current geopolitical landscape. We conclude by advocating for the application of this technology as a tool to inform policymaking and address the ethical considerations raised by its use in peacebuilding and reconciliation efforts.