AI-Enabled Wargaming at the U.S. Army Command and General Staff College: Its Implications for PME and Operational Planning

Timothy J. Williams, Anthony A. Joyce, Seth Lavenski, Jody Colton, Regina Ebell, Tyree Meadows

Source Link

Wargaming remains a cornerstone of military planning, enabling commanders to test courses of action (COAs), anticipate adversary responses, and refine operational designs under compressed timelines. As articulated in Joint Publication (JP) 5-0, Joint Planning, wargaming synchronizes warfighting functions through action-reaction-counteraction cycles, exposing vulnerabilities and optimizing resource allocation. Yet, traditional analog methods—reliant on manual adjudication and static maps—constrain iteration and depth, particularly in multi-domain scenarios against peer threats. The advent of artificial intelligence (AI) offers a transformative solution. Hybrid pipelined ontological-augmented generation (OAG) with retrieval-augmented generation (RAG) models, such as those integrated into the Army’s Vantage platform, can adjudicate outcomes probabilistically while adhering to doctrinal constraints, accelerating decision cycles without sacrificing rigor.

Recent U.S. military experiments underscore this potential. The Air Force’s Decision Advantage Sprints exercises have employed AI to simulate human-machine teaming in wargames, reducing adjudication time from hours to minutes. Similarly, the Johns Hopkins Applied Physics Laboratory’s (JH APL) GenWar initiative uses large language models (LLMs) to automate scenario generation and replay, addressing the labor-intensive nature of traditional exercises. At the U.S. Army Command and General Staff College (CGSC), where faculty have led the Army’s integration of AI into military education, similar innovations culminated in a wargame exercise in November of 2025 where AI not only amplified throughput but also fostered deeper doctrinal application among novice planners. This paper analyzes the CGSC event’s execution, outcomes, and enabling factors, drawing parallels to broader Army and Joint Force initiatives. It concludes with recommendations for scaling AI integration, emphasizing the ethical and operational imperatives in an era of accelerating great-power competition.