Large Language Models are Biased Because They Are Large Language Models
Computation and Language
2025-03-17 v2 Artificial Intelligence
Abstract
This position paper's primary goal is to provoke thoughtful discussion about the relationship between bias and fundamental properties of large language models. I do this by seeking to convince the reader that harmful biases are an inevitable consequence arising from the design of any large language model as LLMs are currently formulated. To the extent that this is true, it suggests that the problem of harmful bias cannot be properly addressed without a serious reconsideration of AI driven by LLMs, going back to the foundational assumptions underlying their design.
Cite
@article{arxiv.2406.13138,
title = {Large Language Models are Biased Because They Are Large Language Models},
author = {Philip Resnik},
journal= {arXiv preprint arXiv:2406.13138},
year = {2025}
}
Comments
To appear in Computational Linguistics. Significantly revised since the prior arXiv version. This preprint has 22 pages