Large Language Models are often politically extreme, usually ideologically inconsistent, and persuasive even in informational contexts
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912364350144512 |
|---|---|
| author | Aldahoul, Nouar Ibrahim, Hazem Varvello, Matteo Kaufman, Aaron Rahwan, Talal Zaki, Yasir |
| author_facet | Aldahoul, Nouar Ibrahim, Hazem Varvello, Matteo Kaufman, Aaron Rahwan, Talal Zaki, Yasir |
| contents | Large Language Models (LLMs) are a transformational technology, fundamentally changing how people obtain information and interact with the world. As people become increasingly reliant on them for an enormous variety of tasks, a body of academic research has developed to examine these models for inherent biases, especially political biases, often finding them small. We challenge this prevailing wisdom. First, by comparing 31 LLMs to legislators, judges, and a nationally representative sample of U.S. voters, we show that LLMs' apparently small overall partisan preference is the net result of offsetting extreme views on specific topics, much like moderate voters. Second, in a randomized experiment, we show that LLMs can promulgate their preferences into political persuasiveness even in information-seeking contexts: voters randomized to discuss political issues with an LLM chatbot are as much as 5 percentage points more likely to express the same preferences as that chatbot. Contrary to expectations, these persuasive effects are not moderated by familiarity with LLMs, news consumption, or interest in politics. LLMs, especially those controlled by private companies or governments, may become a powerful and targeted vector for political influence. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_04171 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Large Language Models are often politically extreme, usually ideologically inconsistent, and persuasive even in informational contexts Aldahoul, Nouar Ibrahim, Hazem Varvello, Matteo Kaufman, Aaron Rahwan, Talal Zaki, Yasir Computers and Society Computation and Language Large Language Models (LLMs) are a transformational technology, fundamentally changing how people obtain information and interact with the world. As people become increasingly reliant on them for an enormous variety of tasks, a body of academic research has developed to examine these models for inherent biases, especially political biases, often finding them small. We challenge this prevailing wisdom. First, by comparing 31 LLMs to legislators, judges, and a nationally representative sample of U.S. voters, we show that LLMs' apparently small overall partisan preference is the net result of offsetting extreme views on specific topics, much like moderate voters. Second, in a randomized experiment, we show that LLMs can promulgate their preferences into political persuasiveness even in information-seeking contexts: voters randomized to discuss political issues with an LLM chatbot are as much as 5 percentage points more likely to express the same preferences as that chatbot. Contrary to expectations, these persuasive effects are not moderated by familiarity with LLMs, news consumption, or interest in politics. LLMs, especially those controlled by private companies or governments, may become a powerful and targeted vector for political influence. |
| title | Large Language Models are often politically extreme, usually ideologically inconsistent, and persuasive even in informational contexts |
| topic | Computers and Society Computation and Language |
| url | https://arxiv.org/abs/2505.04171 |