
Amer Alnajar
International Conflict Management×
Agentic AI Development
PhD Researcher · Kennesaw State University
Agentic AI for Conflict Analysis & Decision-Making
NVIDIA Certified Professional: Agentic AIAbout
I am a PhD researcher in International Conflict Management at Kennesaw State University. I design, build, and evaluate agentic AI systems (fully or semi-autonomous) that operationalize established frameworks in conflict analysis and decision-making, refining those frameworks through implementation.
My work bridges two domains that rarely speak to each other: the context-dependent world of international conflict management and the structured, parameterized architecture of AI agent systems.
“Where conflict theory meets computational architecture: AI systems designed for the complexity of human conflict.”
Research Domains
Structured Conflict Assessment
Examining the dynamics of interstate and intrastate conflict: actors, drivers, structures, escalation patterns, and pathways to resolution. The substantive foundation of all my work.
Structured Analytical Techniques (SATs) & Bias Mitigation
Applying structured analytical techniques to improve the quality and rigor of conflict assessment, reducing cognitive bias and strengthening the epistemic foundations of analysis.
AI-Assisted Conflict Analysis
Using AI as a methodological tool: building conflict frameworks and structured analytical techniques into AI agents, and testing whether they make conflict assessment faster, deeper, and more rigorous.
One integrated program: established frameworks and analytical techniques built into AI agents, then evaluated for speed, rigor, and bias.
AI Agents
Conflict Analysis Agents
AI agents that operationalize established conflict management frameworks (escalation dynamics, negotiation, stakeholder analysis) as structured analytical tools for researchers and analysts.
SAT-Integrated Reasoning
Structured analytical techniques built into AI-assisted workflows, so that AI-supported conflict analysis can be held to the same standards of rigor and bias control as expert assessment.
Crisis Simulation Design
Humanitarian crisis simulations designed with Prof. Volker Franke. We ran a cyber-humanitarian tabletop exercise for about 70 players at DEF CON 33, then adapted it into Attacking Maria, published by the Army Cyber Institute at West Point. I used AI to forecast the simulated attack's effects 72 hours out, which would otherwise take about a week of research. Next: AI agents that play roles alongside human participants.
Publications
Attacking Maria: A Cyber and Humanitarian Crisis Response Simulation
Army Cyber Institute at West Point, co-authored with Volker Franke, Nathan Case, and Cody Whitehouse
Epistemic Asymmetry: AI and the Cognitive Foundations of Strategic Signaling
International Studies Association (ISA) Annual Convention
Geopolitics of AI: Power, Conflict, and the Future of Global Order, edited by Hal Brands
Journal of Peace Research (PRIO)
The Remote Revolution: Drones and Modern Statecraft by Erik Lin-Greenberg
Journal of Peace Research (PRIO)
AI, Automation, and War: The Rise of a Military-Tech Complex by Anthony King
Journal of Peace Research (PRIO)
Radow College Researchers Exploring AI in Humanitarian Crisis Simulation Design
Kennesaw State University, featuring work with Prof. Volker Franke, presented at DEF CON 33
Contact
Interested in collaboration, research, or AI systems work?