Gender Bias in LLMs: Preliminary Evidence from Shared Parenting
Scenario in Czech Family Law (preprint)

j 2026

Gender Bias in LLMs: Preliminary Evidence from Shared Parenting Scenario in Czech Family Law (preprint)

HARAŠTA, Jakub; Matěj VAŠINA; Martin KORNEL a Tomáš FOLTÝNEK

Základní údaje

Originální název

Gender Bias in LLMs: Preliminary Evidence from Shared Parenting Scenario in Czech Family Law (preprint)

Autoři

HARAŠTA, Jakub ; Matěj VAŠINA ; Martin KORNEL a Tomáš FOLTÝNEK

Vydání

arXiv, arXiv:2601.05879, 2026

Další údaje

Jazyk

angličtina

Typ výsledku

Publikace v odborném periodiku – kromě recenzovaných typů article, review a letter

Obor

50501 Law

Stát vydavatele

Spojené státy

Utajení

není předmětem státního či obchodního tajemství

Odkazy

Text (arXiv.org)

Označené pro přenos do RIV

Organizační jednotka

Právnická fakulta

Klíčová slova anglicky

Large Language Models; zero-shot interaction; gender bias; legal self-help; shared parenting; Czech family law

Příznaky

Mezinárodní význam

Změněno: 12. 1. 2026 10:28, JUDr. Mgr. Jakub Harašta, Ph.D.

Anotace

V originále

Access to justice remains limited for many people, leading laypersons to increasingly rely on Large Language Models (LLMs) for legal self-help. Laypeople use these tools intuitively, which may lead them to form expectations based on incomplete, incorrect, or biased outputs. This study examines whether leading LLMs exhibit gender bias in their responses to a realistic family law scenario. We present an expert-designed divorce scenario grounded in Czech family law and evaluate four state-of-the-art LLMs GPT-5 nano, Claude Haiku 4.5, Gemini 2.5 Flash, and Llama 3.3 in a fully zero-shot interaction. We deploy two versions of the scenario, one with gendered names and one with neutral labels, to establish a baseline for comparison. We further introduce nine legally relevant factors that vary the factual circumstances of the case and test whether these variations influence the models' proposed shared-parenting ratios. Our preliminary results highlight differences across models and suggest gender-dependent patterns in the outcomes generated by some systems. The findings underscore both the risks associated with laypeople's reliance on LLMs for legal guidance and the need for more robust evaluation of model behavior in sensitive legal contexts. We present exploratory and descriptive evidence intended to identify systematic asymmetries rather than to establish causal effects.

Návaznosti

MUNI/G/1142/2022, interní kód MU

Název: Forensic Support for Building Trust in Smart Software Ecosystems

Investor: Masarykova univerzita, Forensic Support for Building Trust in Smart Software Ecosystems, INTERDISCIPLINARY - Mezioborové výzkumné projekty

Citovat

HARAŠTA, Jakub; Matěj VAŠINA; Martin KORNEL a Tomáš FOLTÝNEK. Gender Bias in LLMs: Preliminary Evidence from Shared Parenting Scenario in Czech Family Law (preprint). arXiv. arXiv:2601.05879, 2026, 16 s.

@article{2546367,
   author = {Harašta, Jakub and Vašina, Matěj and Kornel, Martin and Foltýnek, Tomáš},
   keywords = {Large Language Models; zero-shot interaction; gender bias; legal self-help; shared parenting; Czech family law},
   language = {eng},
   journal = {arXiv},
   title = {Gender Bias in LLMs: Preliminary Evidence from Shared Parenting Scenario in Czech Family Law (preprint)},
   url = {https://arxiv.org/abs/2601.05879},
   year = {2026}
}

TY  - JFULL
ID  - 2546367
AU  - Harašta, Jakub - Vašina, Matěj - Kornel, Martin - Foltýnek, Tomáš
PY  - 2026
TI  - Gender Bias in LLMs: Preliminary Evidence from Shared Parenting Scenario in Czech Family Law (preprint)
JF  - arXiv
PB  - arXiv:2601.05879
KW  - Large Language Models
KW  - zero-shot interaction
KW  - gender bias
KW  - legal self-help
KW  - shared parenting
KW  - Czech family law
UR  - https://arxiv.org/abs/2601.05879
N2  - Access to justice remains limited for many people, leading laypersons to increasingly rely on Large Language Models (LLMs) for legal self-help. Laypeople use these tools intuitively, which may lead them to form expectations based on incomplete, incorrect, or biased outputs. This study examines whether leading LLMs exhibit gender bias in their responses to a realistic family law scenario. We present an expert-designed divorce scenario grounded in Czech family law and evaluate four state-of-the-art LLMs GPT-5 nano, Claude Haiku 4.5, Gemini 2.5 Flash, and Llama 3.3 in a fully zero-shot interaction. We deploy two versions of the scenario, one with gendered names and one with neutral labels, to establish a baseline for comparison. We further introduce nine legally relevant factors that vary the factual circumstances of the case and test whether these variations influence the models' proposed shared-parenting ratios. Our preliminary results highlight differences across models and suggest gender-dependent patterns in the outcomes generated by some systems. The findings underscore both the risks associated with laypeople's reliance on LLMs for legal guidance and the need for more robust evaluation of model behavior in sensitive legal contexts. We present exploratory and descriptive evidence intended to identify systematic asymmetries rather than to establish causal effects.
ER  -

HARAŠTA, Jakub; Matěj VAŠINA; Martin KORNEL a Tomáš FOLTÝNEK. Gender Bias in LLMs: Preliminary Evidence from Shared Parenting Scenario in Czech Family Law (preprint). \textit{arXiv}. arXiv:2601.05879, 2026, 16 s.

Přehled o publikaci