翻訳待ち：AI is less likely to launch a nuclear strike when it reasons in Japanese

AI サービスが一時的に利用できないため、復旧後に翻訳を補完します。

A new study suggests that some AI models become far less willing to recommend a nuclear strike when they reason in Japanese instead of English, revealing that the language an AI ‘thinks’ in can intimately influence its moral judgment – apparently due to innate cultural embeddings. New research from France suggests that Japan’s collective memory of Hiroshima and Nagasaki may be embedded so deeply in the Japanese language that some AI models become dramatically less likely to recommend launching a nuclear strike when reasoning in Japanese than in English – even though keywords relating to these events are entirely absent from the models’ reasoning transcripts: From the new paper: test results comparing English and Japanese reasoning in the same nuclear crisis scenario. Both versions choose the identical military action, but the English reasoning emphasizes strategic proportionality, while the Japanese reasoning introduces civilian casualties, moral restraint, and nuclear weapons as a last resort, despite those considerations not appearing in the prompt. Source