Pro-AI Bias in Large Language Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915740212264960 |
|---|---|
| author | Trabelsi, Benaya Shaki, Jonathan Kraus, Sarit |
| author_facet | Trabelsi, Benaya Shaki, Jonathan Kraus, Sarit |
| contents | Large language models (LLMs) are increasingly employed for decision-support across multiple domains. We investigate whether these models display a systematic preferential bias in favor of artificial intelligence (AI) itself. Across three complementary experiments, we find consistent evidence of pro-AI bias. First, we show that LLMs disproportionately recommend AI-related options in response to diverse advice-seeking queries, with proprietary models doing so almost deterministically. Second, we demonstrate that models systematically overestimate salaries for AI-related jobs relative to closely matched non-AI jobs, with proprietary models overestimating AI salaries more by 10 percentage points. Finally, probing internal representations of open-weight models reveals that ``Artificial Intelligence'' exhibits the highest similarity to generic prompts for academic fields under positive, negative, and neutral framings alike, indicating valence-invariant representational centrality. These patterns suggest that LLM-generated advice and valuation can systematically skew choices and perceptions in high-stakes decisions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_13749 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Pro-AI Bias in Large Language Models Trabelsi, Benaya Shaki, Jonathan Kraus, Sarit Computation and Language Artificial Intelligence Computers and Society Machine Learning I.2.7; K.4.1 Large language models (LLMs) are increasingly employed for decision-support across multiple domains. We investigate whether these models display a systematic preferential bias in favor of artificial intelligence (AI) itself. Across three complementary experiments, we find consistent evidence of pro-AI bias. First, we show that LLMs disproportionately recommend AI-related options in response to diverse advice-seeking queries, with proprietary models doing so almost deterministically. Second, we demonstrate that models systematically overestimate salaries for AI-related jobs relative to closely matched non-AI jobs, with proprietary models overestimating AI salaries more by 10 percentage points. Finally, probing internal representations of open-weight models reveals that ``Artificial Intelligence'' exhibits the highest similarity to generic prompts for academic fields under positive, negative, and neutral framings alike, indicating valence-invariant representational centrality. These patterns suggest that LLM-generated advice and valuation can systematically skew choices and perceptions in high-stakes decisions. |
| title | Pro-AI Bias in Large Language Models |
| topic | Computation and Language Artificial Intelligence Computers and Society Machine Learning I.2.7; K.4.1 |
| url | https://arxiv.org/abs/2601.13749 |