Same question, different history: language, national identity, and credit in large language models

2026-06-22

AI Summary

This study investigates whether large language models exhibit systematic biases in attributing contested inventions across nations depending on the language of the query. By conducting 75,896 queries across 12 languages to 11 prominent models regarding 21 disputed inventions, the research employs multilingual prompt engineering, large-scale response collection, and controlled statistical analysis to reveal language as a cultural memory switch. Findings indicate that models activate nation-specific historical narratives aligned with the query language: claims from lower-status countries are more likely to surface only when queried in their native languages, whereas figures dominant in English-language historiography appear consistently across languages. This demonstrates that language significantly shapes the historical memory encoded in model outputs, reflecting a form of computational everyday nationalism.

Abstract

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.

Problem

• large language models
• national identity
• historical credit
• language bias
• cultural memory

Innovation

• large language models
• cultural memory
• banal nationalism
• language bias
• historical attribution