The authors summarize research to date about using a game-theoretic model of competition and cooperation in outer space to examine the dynamics of conflict there, short of wars of total destruction. They focus specifically on the dynamics of space competition in the context of limited space war. They provide a summary of the aspects of the research that would be of interest to a general audience of space policymakers and potential warfighters.

Space Competition and the Dynamics of Conflict

Using Game Theory and Artificial Intelligence to Gain Strategic Insight

Bonnie L. Triezenberg, Krista Langeland, Bryce Downing

ResearchPublished Jul 5, 2022

Competition between great national powers has often played out in space. As it becomes increasingly militarized, understanding the long-term effect of nation-state investments in space security and subsequent use of products from those investments becomes strategically important. In 2014, the RAND Corporation began to develop a game-theoretic model to assess strategic implications of U.S. and a competitor nation's investments in space capabilities. In projects since, RAND researchers have built on traditional game theory to provide a context-rich assessment of how nation-state investments may play out over a range of possible futures. Although previous research using this model explored the effect of investments on deterring horizontal escalation of a terrestrial war into outer space, the authors focus here on the dynamics of space competition. They describe strategic interaction patterns, where possible; the conditions that give rise to them; and how investments shape those conditions. In many cases, they have yet to discover correlations between conditions in the game and resulting dynamics and strategic interaction patterns.

To create a context-rich assessment of space competition, they developed a complex model using sophisticated artificial intelligence (AI) methods. Although they found that this complexity adds context to their assessments of how investments may play out, it also hampered their ability to isolate the conditions that gave rise to different strategic interaction patterns. This report should be of interest to not only space policy decisionmakers but to anyone contemplating using AI models to perform exploratory research.

Key Takeaways

  • Improving defenses against one type of attack (in this case, kinetic) simply shifts the adversary's tactics, not their strategy. If it is in the adversary's long-term strategic interest to attack space assets, reducing the probability of success for their preferred attack option may simply mean they will switch to the next best weapon-target pairing.
  • Misperception was stabilizing when (1) the adversary perceives a peer opponent to be disadvantaged or (2) a disadvantaged opponent is perceived to be a peer. In other conditions, misperception was destabilizing or had no clear trend.
  • Decisions on whether to reveal or conceal investments in, or the actual extent of, offensive capabilities are not straightforward, especially when considering that not all potential opponents in space are peer competitors.

There were three distinct patterns of strategic interaction in games in which deterrence was lost but that stopped short of wars of total destruction

  • In early shaping of the battlefield games, attacks occurred two or more years before ground conflict.
  • In wars of weapons attrition games, both sides reserve their attacks until just before or simultaneous with the start of a ground war and the attacks from both sides are focused primarily on reducing their opponent’s offensive capability.
  • In horizontal escalation games, attacks are made well after the start of the terrestrial conflict and focus on creating exploitable shortfalls in the adversary's ability to project power from space.

Recommendations

  • Developing a better sense of the causal dynamics at play in a game-theoretic model could help build a more informative narrative about the forces at play in a conflict. Future work using a similar game-theoretic approach could benefit from a targeted examination of more bounded questions to elucidate some of these cause-and-effect conditions.
  • More targeted studies of strategic interactions in space could help identify these favorable tactics even as long-term adversary strategies remain unchanged.
  • Additional work could leverage a similar but simpler game model, along with existing work that RAND and others have done on strategic messaging, to determine how uncertainty about perceptions could impact conflict outcomes and strategic interaction patterns.

Topics

Document Details

Citation

Chicago Manual of Style

Triezenberg, Bonnie L., Krista Langeland, and Bryce Downing, Space Competition and the Dynamics of Conflict: Using Game Theory and Artificial Intelligence to Gain Strategic Insight. Santa Monica, CA: RAND Corporation, 2022. https://www.rand.org/pubs/research_reports/RRA751-1.html.
BibTeX RIS

Research conducted by

This research is sponsored by the National Reconnaissance Office (NRO) and conducted within the Cyber and Intelligence Policy Center of the RAND National Security Research Division (NSRD).

This publication is part of the RAND research report series. Research reports present research findings and objective analysis that address the challenges facing the public and private sectors. All RAND research 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.

RAND 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.