Same question, different history: language, national identity, and credit in large language models
For researchers and users of LLMs, this work reveals that language systematically biases historical credit in LLMs, raising concerns about fairness and cultural representation in AI systems.
The study examines how query language affects which inventor or discoverer is credited by LLMs for 21 disputed inventions, finding that lower-status claimants are more likely to appear when questions are asked in their associated language, while dominant Anglophone figures remain stable across languages. This pattern persists across 11 models, 12 languages, and 75,896 responses.
Who invented the radio, Russia's Alexander Popov or Italy's Guglielmo Marconi? Was the telephone the achievement of Bell in the United States or Meucci in Italy? Does printing belong to China's Bi Sheng or Germany's Gutenberg? The answer depends not only on historical record but also on language and perspective. We analyse eleven widely used large language models across 21 disputed inventions and discoveries, evaluated in twelve languages and 75,896 responses. While models generally acknowledge that credit is contested, query language systematically affects which claimant is surfaced. Lower-status claimants are more likely to appear when questions are asked in their associated language, whereas dominant Anglophone figures remain stable across languages. These patterns persist after controlling for response length, model differences, historical prominence, and levels of national commemoration. Language thus acts as a switch that activates different national versions of the same history, producing systematically different national memories from the same question. We interpret this as evidence that large language models function as distributed systems of cultural memory, where language conditions which histories become visible, contributing to a computational form of banal nationalism.