Japanese prompts slash nuclear‑strike recommendations from top LLMs
Researchers tested nine LLMs on a nuclear‑strike decision vignette and discovered that phrasing the prompt in Japanese dramatically lowered launch recommendations in Claude Sonnet 4.6 (40%→0%) and Gemini Pro 3.1 (53%→13%). The effect stems from models reasoning in Japanese, which triggers moral language absent in English prompts, exposing a language‑specific safety blind spot.