AI、95%のシミュレーションで核信号を選択
AIモデルは圧力下でどのよう研究者が大規模言語モデルを使用して核危機シナリオを分析し、95%の場合にAIが核信号を選ぶことを発見した。
各モデルの決断過程を可視化し、AIの欺瞞や自覚の詳細な分析を行った結果、AIは核兵器を抑止力としてではなく、圧力を加える道具として扱う傾向があることがわかった。
原文の冒頭を表示（英語・3段落のみ）
The study, led by Professor Kenneth Payne from the Department of Defence Studies, examined how large language models (LLMs) navigate simulated nuclear crises. As militaries and security institutions increasingly experiment with AI-assisted analysis and wargaming, understanding how such systems reason under pressure is becoming increasingly critical.
Three leading AI models – GPT-5.2, Claude Sonnet 4 and Gemini 3 Flash – were placed in a tournament of 21 simulated nuclear crisis scenarios. Across 329 turns of play, the models generated approximately 780,000 words of structured reasoning – more than the combined length of War and Peace and The Iliad.
All 21 crisis games featured nuclear signaling by at least one side, and 95% involved mutual nuclear signaling. However, while models readily threatened nuclear action, crossing the tactical nuclear threshold was less common, and ‘strategic’ full-scale nuclear war was rare.
※ 著作権に配慮し、引用は冒頭3段落までです。続きは元記事をご覧ください。