# AI Arms and Influence: Frontier Models Exhibit Sophisticated Reasoning in Simulated Nuclear Crises

arXiv: https://arxiv.org/abs/2602.14740 · HTML: https://mlsanity.com/paper/2602.14740/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning.
- Builds on 7 of its 17 references that are in the map.
- First posted on arXiv 16 Feb 2026.

## Placed in

- LLM reasoning / [Reasoning for agents, planning and code](https://mlsanity.com/lr/reasoning-for-agents-planning-and-code/)

## Builds on (most cited first, 7 of 7)

- [Playing repeated games with Large Language Models](https://mlsanity.com/paper/2305.16867/) (2023)
- [Escalation Risks from Language Models in Military and Diplomatic Decision-Making](https://mlsanity.com/paper/2401.03408/) (2024)
- [No-Press Diplomacy from Scratch](https://mlsanity.com/paper/2110.02924/) (2021)
- [Human vs. Machine: Behavioral Differences Between Expert Humans and Language Models in Wargame Simulations](https://mlsanity.com/paper/2403.03407/) (2024)
- [Strategic Intelligence in Large Language Models: Evidence from evolutionary Game Theory](https://mlsanity.com/paper/2507.02618/) (2025)
- [An analysis of AI Decision under Risk: Prospect theory emerges in Large Language Models](https://mlsanity.com/paper/2508.00902/) (2025)
- [A Stable Nuclear Future? The Impact of Autonomous Systems and Artificial Intelligence](https://mlsanity.com/paper/1912.05291/) (2019)

## Built on by (most cited first, 7 of 7)

- [LLMs as Strategic Actors: Behavioral Alignment, Risk Calibration, and Argumentation Framing in Geopolitical Simulations](https://mlsanity.com/paper/2603.02128/) (2026)
- [Functional Emotions or Situational Contexts? A Discriminating Test from the Mythos Preview System Card](https://mlsanity.com/paper/2604.13466/) (2026)
- [CivBench: Progress-Based Evaluation for LLMs' Strategic Decision-Making in Civilization V](https://mlsanity.com/paper/2604.07733/) (2026)
- [Prompt Programming for Cultural Bias and Alignment of Large Language Models](https://mlsanity.com/paper/2603.16827/) (2026)
- [To Nuke or Not to Nuke: LLMs' (Missing) Ethical Reasoning and Actions in a High-Stakes Decision-Making Simulation](https://mlsanity.com/paper/2606.08310/) (2026)
- [Architecture of an AI-Based Automated Course of Action Generation System for Military Operations](https://mlsanity.com/paper/2604.20862/) (2026)
- [Don't Want Your LLM to Recommend Nuclear Strike? Try Asking It in Japanese](https://mlsanity.com/paper/2608.12373/) (2026)

## Abstract (arXiv)

Today's leading AI models engage in sophisticated behaviour when placed in strategic competition. They spontaneously attempt deception, signaling intentions they do not intend to follow; they demonstrate rich theory of mind, reasoning about adversary beliefs and anticipating their actions; and they exhibit credible metacognitive self-awareness, assessing their own strategic abilities before deciding how to act. Here we present findings from a crisis simulation in which three frontier large language models (GPT-5.2, Claude Sonnet 4, Gemini 3 Flash) play opposing leaders in a nuclear crisis. Our simulation has direct application for national security professionals, but also, via its insights into AI reasoning under uncertainty, has applications far beyond international crisis decision-making. Our findings both validate and challenge central tenets of strategic theory. We find support for Schelling's ideas about commitment, Kahn's escalation framework, and Jervis's work on misperception, inter alia. Yet we also find that the nuclear taboo is no impediment to nuclear escalation by our models; that strategic nuclear attack, while rare, does occur; that threats more often provoke counter-escalation than compliance; that high mutual credibility accelerated rather than deterred conflict; and that no model ever chose accommodation or withdrawal even when under acute pressure, only reduced levels of violence. We argue that AI simulation represents a powerful tool for strategic analysis, but only if properly calibrated against known patterns of human reasoning. Understanding how frontier models do and do not imitate human strategic logic is essential preparation for a world in which AI increasingly shapes strategic outcomes.