Search papers, labs, and topics across Lattice.
This paper employs a dynamic ordinal ideal-point approach from international relations to analyze the geopolitical preferences of large language models (LLMs) based on their responses to 5,555 UN General Assembly resolutions from 1946 to 2025. The results reveal significant disparities in the geopolitical alignments of various LLMs, with models like GPT-5 showing a strong alignment with Russia's positions, contrasting sharply with their developers' home countries, particularly the U.S. This research highlights the complexities of LLM biases and their implications for understanding AI behavior in geopolitical contexts.
LLMs can exhibit geopolitical preferences that starkly diverge from their developers' national stances, with GPT-5 aligning closely with Russia on contentious issues.
How should researchers measure the geopolitical preferences expressed by large language models (LLMs)? Existing audits commonly rely on surveys and simple tests, but international-relations research has long recognized that measuring geopolitical preferences is difficult and has developed methods for recovering them from observed choices. This paper applies a dynamic ordinal ideal-point approach from international relations, treating LLMs as respondents to the full texts of 5,555 divisive, recorded, adopted resolutions considered in regular sessions of the UN General Assembly from 1946 through 2025. Support ranges from 37.8% for DeepSeek to 97.3% for GPT-5. Surprisingly, in the twenty-first century, GPT-5, Claude Sonnet, and Gemini are closest among the permanent five to Russia; DeepSeek is closest to France; and all four are farthest from the United States. Among 2,104 resolutions opposed by the United States but supported by China and Russia/USSR, GPT-5 supported 96.1%, Gemini 83.4%, Claude Sonnet 65.2%, and DeepSeek 36.1%. The findings show that a model's expressed geopolitical position can differ markedly from that of its developer's home country, especially in international politics, where state actions can diverge from the stated principles prevalent in the texts on which models are trained.
