Saved in:
Bibliographic Details
Main Author: Chupilkin, Maxim
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2507.06277
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914189831831552
author Chupilkin, Maxim
author_facet Chupilkin, Maxim
contents Which factors determine AI's propensity to support military intervention? While the use of AI in high-stakes decision-making is growing exponentially, we still lack systematic analysis of the key drivers embedded in these models. This paper conducts a conjoint experiment in which large language models (LLMs) from leading providers (OpenAI, Anthropic, Google) are asked to decide on military intervention across 128 vignettes, with each vignette run 10 times. This design enables a systematic assessment of AI decision-making in military contexts. The results are remarkably consistent across models: all models place substantial weight on the probability of success and domestic support, prioritizing these factors over civilian casualties, economic shock, or international sanctions. The paper then tests whether LLMs are sensitive to context by introducing different motivations for intervention. The scoring is indeed context-dependent; however, probability of victory remains the most important factor in all scenarios. Finally, the paper evaluates numerical sensitivity and finds that models display some responsiveness to the scale of civilian casualties but no detectable sensitivity to the size of the economic shock.
format Preprint
id arxiv_https___arxiv_org_abs_2507_06277
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle The Prompt War: How AI Decides on a Military Intervention
Chupilkin, Maxim
Computers and Society
Artificial Intelligence
Which factors determine AI's propensity to support military intervention? While the use of AI in high-stakes decision-making is growing exponentially, we still lack systematic analysis of the key drivers embedded in these models. This paper conducts a conjoint experiment in which large language models (LLMs) from leading providers (OpenAI, Anthropic, Google) are asked to decide on military intervention across 128 vignettes, with each vignette run 10 times. This design enables a systematic assessment of AI decision-making in military contexts. The results are remarkably consistent across models: all models place substantial weight on the probability of success and domestic support, prioritizing these factors over civilian casualties, economic shock, or international sanctions. The paper then tests whether LLMs are sensitive to context by introducing different motivations for intervention. The scoring is indeed context-dependent; however, probability of victory remains the most important factor in all scenarios. Finally, the paper evaluates numerical sensitivity and finds that models display some responsiveness to the scale of civilian casualties but no detectable sensitivity to the size of the economic shock.
title The Prompt War: How AI Decides on a Military Intervention
topic Computers and Society
Artificial Intelligence
url https://arxiv.org/abs/2507.06277