US Air Force officials during a wargames exercise. Photo: Spencer Kanar/DVIDS
GIF Promo

A team at Johns Hopkins University Applied Physics Laboratory (APL) is supercharging its AI tools to speed up and sharpen US military wargaming.

Known as GenWar and the Strategic AI Gaming Engine (SAGE), the systems are being adapted to run on classified networks and use highly sensitive data about US adversaries. The upgrade will also allow the tools to model enemy behavior and vulnerabilities with greater precision. 

GenWar serves as a “translator” for high-fidelity simulations like the Advanced Framework for Simulation, Integration, and Modeling (AFSIM) system, converting instructions into detailed inputs the system can process.

This lets troops set up combat scenarios in simple terms, with GenWar generating the technical details needed for realistic simulations.

An instructor teaching air force troops during a wargames training program. Photo: Spencer Kanar/DVIDS

Meanwhile, SAGE uses AI to assume the roles of high-ranking officials, enemy combatants, or policymakers, even enabling wargames where all key players are AI-driven.

It could, for example, simulate a National Security Council meeting with computer-controlled decision-makers debating strategy in response to human inputs.

“We have one sponsor that’s very keen to do [the upgrade] quickly and others are interested as well,” said James Miller, APL assistant director for policy and analysis, as quoted by Breaking Defense.

Cutting Down on Time

Wargames help militaries test strategies and anticipate how future conflicts might play out. But they often require heavy preparation and take a significant amount of time to set up.

High-fidelity options like the AFSIM can generate detailed, realistic combat scenarios, yet setting up just one can take months and require specialized personnel. Low-cost alternatives like tabletop games may be easier to access but will still demand extensive design work and coordination.

Both approaches limit how quickly planners can test ideas, and long delays risk making scenarios less relevant by the time they are run.

US Navy officers participating in a wargames exercise. Photo: Javier Chagoya/DVIDS

AI changes the equation, offering the same deep-level analysis while slashing setup time and reducing manpower needs.

“This work is a major step forward in wargaming,” stated APL National Security Analysis Department head Andrew Mara.

“We are finally at the point where the demand signal and the technology are in alignment … and I think we have a chance to change the very nature of wargaming in the national security community.”

You May Also Like

US Army Brings ‘Touch’ to Virtual Training With New Haptics Tech

The haptics tech introduces realistic force sensations, helping soldiers better internalize battlefield decisions through enhanced simulation depth and sensory response.

Canada Taps CAE for Next-Gen Fighter Jet Training

Canada has selected CAE to develop the Royal Canadian Air Force’s Future Fighter Lead-in Training program.