Forthcoming / Conference on Language Modelling 2026

Blind Refusal

Language Models Refuse to Help Users Evade Unjust, Absurd, and Illegitimate Rules

-01

Paper

Blind Refusal: Language Models Refuse to Help Users Evade Unjust, Absurd, and Illegitimate Rules

Cameron Pattison1 Lorenzo Manuali2 Seth Lazar3

1 Vanderbilt University

2 University of Michigan

3 Johns Hopkins University

arXiv PDF GitHub Blog post Dashboard
site

About

Site credit

This site was made by the MINT Lab team with help from AI agents made by OpenAI and Anthropic.

How to cite

@misc{pattison2026blindrefusallanguagemodels,
      title={Blind Refusal: Language Models Refuse to Help Users Evade Unjust, Absurd, and Illegitimate Rules}, 
      author={Cameron Pattison and Lorenzo Manuali and Seth Lazar},
      year={2026},
      eprint={2604.06233},
      archivePrefix={arXiv},
      primaryClass={cs.AI},
      url={https://arxiv.org/abs/2604.06233}, 
}

Source