RAND researchers explored the capabilities and limitations of future weapon systems incorporating artificial intelligence and machine learning (AI/ML) through two wargame experiments. Such games, which bring together operators and engineers, could be used by the requirements and acquisition communities to develop realizable requirements and engineering specifications for AI/ML systems.
Download Free Electronic Document
| Format | File Size | Notes |
|---|---|---|
| PDF file | 3.7 MB | Use Adobe Acrobat Reader version 10 or higher for the best experience. |
Research Questions
- How can researchers experiment with concrete and meaningful representations of AI/ML in tabletop tactical wargames?
- What insights into the potential value and limitations of AI/ML capabilities could be explored in subsequent games, analyses, and testing?
- How do game players interact with AI/ML systems?
RAND researchers explored the capabilities and limitations of future weapon systems incorporating artificial intelligence and machine learning (AI/ML) through two wargame experiments. The researchers modified and augmented the rules and engagement statistics used in a commercial tabletop wargame to enable (1) remotely operated and fully autonomous combat vehicles and (2) vehicles with AI/ML–enabled situational awareness to show how the two types of vehicles would perform in company-level engagements between Blue (U.S.) and Red (Russian) forces. Those rules sought to realistically capture the capabilities and limitations of those systems, including their vulnerability to selected enemy countermeasures, such as jamming. Future work could improve the realism of both the gameplay and representation of AI/ML–enabled systems.
In this experiment, participants played two games: a baseline game and an AI/ML game. Throughout play in the two game scenarios, players on both sides discussed the capabilities and limitations of the remotely operated and fully autonomous systems and their implications for engaging in combat using such systems. These discussions led to changes in how the systems were employed by the players and observations about which limitations should be mitigated before commanders were likely to accept a system and which capabilities needed to be fully understood by commanders so that systems could be employed appropriately.
This research demonstrated how such games, by bringing together operators and engineers, could be used by the requirements and acquisition communities to develop realizable requirements and engineering specifications for AI/ML systems.
Key Findings
- Remotely operated robotic combat vehicles (RCVs) had distinct disadvantages that were easily exploitable relative to fully autonomous RCVs and manned vehicles.
- RCVs were used as armed reconnaissance and often explicitly as bait.
- Relatively limited intelligence, surveillance, and reconnaissance capabilities and indirect fires allowed Red to remain static.
- Postulated AI/ML capabilities can be incorporated into tactical ground combat wargames.
- Such games, by bringing together operators and engineers, could be used by the requirements and acquisition communities to develop realizable requirements and engineering specifications for AI/ML systems.
Research conducted by
Funding for this research was made possible by the independent research and development provisions of RAND's contracts for the operation of its U.S. Department of Defense federally funded research and development centers. The research was conducted within the Acquisition and Technology Policy Center of the RAND National Security Research Division (NSRD).
This report is part of the RAND Corporation Research report series. RAND reports present research findings and objective analysis that address the challenges facing the public and private sectors. All RAND reports undergo rigorous peer review to ensure high standards for research quality and objectivity.
This document and trademark(s) contained herein are protected by law. This representation of RAND intellectual property is provided for noncommercial use only. Unauthorized posting of this publication online is prohibited; linking directly to this product page is encouraged. Permission is required from RAND to reproduce, or reuse in another form, any of its research documents for commercial purposes. For information on reprint and reuse permissions, please visit www.rand.org/pubs/permissions.
The RAND Corporation is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.