This post is also available in:
עברית (Hebrew)
At Johns Hopkins University Applied Physics Laboratory (APL), researchers are refining a pair of artificial intelligence tools that could significantly change how the U.S. military runs its wargames. The systems—called GenWar and the Strategic AI Gaming Engine (SAGE)—are being adapted to operate on classified networks and ingest sensitive data, enabling more accurate simulations of adversary behavior and strategic vulnerabilities.
Wargames have long been a staple of military planning, offering a way to test strategies and explore how conflicts might unfold. But traditional methods are time-consuming and resource-intensive. Building a high-fidelity scenario with tools like AFSIM (Advanced Framework for Simulation, Integration, and Modeling) can take months and require technical specialists to configure every detail.
According to Breaking Defense, GenWar is designed to solve that bottleneck. It takes plain-language instructions from users—such as commanders or analysts—and converts them into the detailed inputs that AFSIM needs. This allows planners to build out complex, data-rich scenarios without needing deep technical expertise, significantly speeding up the process.
SAGE takes the concept a step further by using AI to simulate the decision-making of key actors in a scenario. It can play the roles of military leaders, enemy forces, or even national policymakers—responding to human actions with AI-driven strategy. One application could simulate a National Security Council meeting, with AI-generated officials debating policy responses in real time.
By combining these tools, APL aims to cut down not just on setup time, but also on the number of people needed to run meaningful exercises. The goal is to make high-level wargaming faster, more accessible, and more reflective of real-world complexity.
While the technology is still being integrated into secure environments, the potential is already clear: planners could soon run more frequent, more nuanced simulations—without waiting weeks or months to get started. For the defense community, this shift could mark a turning point in how strategic foresight is developed and tested.
