Position: AI Is Not Ready for Strategic Conflicts
Mark Riedl and Glenn Matlin (Georgia Tech) argue that no LLM-enabled wargame should inform planning, doctrine, policy or crisis response without an auditable safety case.
Ask this paper
Why wargames are risky: The model's text decides both what an actor attempts and what becomes simulated reality.
Five failure modes: Decision laundering, adjudication opacity, role collapse, escalation through adjudication and failure of strategic imagination.
Recommended use: Open-ended wargames should be used today to stress-test decision-influencing LLM agents, not as evidence for decisions.
Evaluation claim: Ordinary benchmarks cannot establish safety here, and a wargame is a stress test, not a safety case.
Abstract
Open-ended strategic wargames are high-stakes LM-based social simulations: they model adversaries, institutions, escalation, plan brittleness, doctrine, and crisis response. Language models (LMs) are attractive because they can play agents, generate scenario branches, adjudicate ambiguous actions, and summarize lessons, but the same affordances make open-ended roles dangerous: model language determines both what an actor attempts and what becomes simulated reality. This position paper argues that no LM-enabled wargame should inform planning, doctrine, policy, or crisis response without an auditable safety case, and that the proper use of open-ended wargames today is to stress-test decision-influencing LM agents. We identify five failure modes: decision laundering, adjudication opacity, role collapse, escalation-through-adjudication, and failure of strategic imagination. Ordinary benchmarks cannot establish safety for these settings. Wargames can expose failures as stress tests; they are not themselves safety cases for consequential use.