Are generative AI text annotations systematically biased?

Sjoerd B. Stolwijk, Mark Boukes, Damian Trilling
12/9/2025
cs.CLcs.AI

No confident code implementation found for this paper.

Abstract

This paper investigates bias in GLLM annotations by conceptually replicating manual annotations of Boukes (2024). Using various GLLMs (Llama3.1:8b, Llama3.3:70b, GPT4o, Qwen2.5:72b) in combination with five different prompts for five concepts (political content, interactivity, rationality, incivility, and ideology). We find GLLMs perform adequate in terms of F1 scores, but differ from manual annotations in terms of prevalence, yield substantively different downstream results, and display systematic bias in that they overlap more with each other than with manual annotations. Differences in F1 scores fail to account for the degree of bias.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. Any repos shown below are weak matches — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

Evaluating how LLM annotations represent diverse views on contentious topics

85% match2025

Hey GPT, Can You be More Racist? Analysis from Crowdsourced Attempts to Elicit Biased Content from Generative AI

84% match2024

Keeping Humans in the Loop: Human-Centered Automated Annotation with Generative AI

84% match2024

Assessing the Reliability of LLMs Annotations in the Context of Demographic Bias and Model Explanation

83% match2025

Heuristics and Biases in AI Decision-Making: Implications for Responsible AGI

82% match1 repo2024

GPT is Not an Annotator: The Necessity of Human Annotation in Fairness Benchmark Construction

82% match2024

Cite this paper

@article{stolwijk2025generative,
  title  = {Are generative AI text annotations systematically biased?},
  author = {Sjoerd B. Stolwijk and Mark Boukes and Damian Trilling},
  year   = {2025},
  eprint = {2512.08404},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2512.08404v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2512.08404.svg)](https://codeofpaper.com/paper/2512.08404)

Discussion