# The Prompt War: How AI Decides on a Military Intervention

arXiv: https://arxiv.org/abs/2507.06277 · HTML: https://mlsanity.com/paper/2507.06277/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 34 references that are in the map.
- First posted on arXiv 8 Jul 2025.

## Builds on (most cited first, 9 of 9)

- [Aligning AI With Shared Human Values](https://mlsanity.com/paper/2008.02275/) (2020)
- [Artificial Intelligence, Values and Alignment](https://mlsanity.com/paper/2001.09768/) (2020)
- [War and Peace (WarAgent): Large Language Model-based Multi-Agent Simulation of World Wars](https://mlsanity.com/paper/2311.17227/) (2023)
- [Escalation Risks from Language Models in Military and Diplomatic Decision-Making](https://mlsanity.com/paper/2401.03408/) (2024)
- [Safetywashing: Do AI Safety Benchmarks Actually Measure Safety Progress?](https://mlsanity.com/paper/2407.21792/) (2024)
- [Assessing Political Bias in Large Language Models](https://mlsanity.com/paper/2405.13041/) (2024)
- [Human vs. Machine: Behavioral Differences Between Expert Humans and Language Models in Wargame Simulations](https://mlsanity.com/paper/2403.03407/) (2024)
- [Critical Foreign Policy Decisions (CFPD)-Benchmark: Measuring Diplomatic Preferences in Large Language Models](https://mlsanity.com/paper/2503.06263/) (2025)
- [Reasons to Doubt the Impact of AI Risk Evaluations](https://mlsanity.com/paper/2408.02565/) (2024)

## Built on by (most cited first, 4 of 4)

- [Left Leaning Models: How AI Evaluates Economic Policy?](https://mlsanity.com/paper/2507.15771/) (2025)
- [Hidden Topics: Measuring Sensitive AI Beliefs with List Experiments](https://mlsanity.com/paper/2602.21939/) (2026)
- [Multi-Agent Strategic Games with LLMs](https://mlsanity.com/paper/2605.03604/) (2026)
- [Language models judge war differently when tested for alignment](https://mlsanity.com/paper/2609.05009/) (2026)

## Abstract (arXiv)

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.