2026-05-25原文

When AI Takes Sides on Questions of Faith: Persistent Asymmetries in AI-Mediated Faith Guidance

A study finds that large language models (LLMs) exhibit persistent asymmetries when answering questions about religious conversion. Models tend to support joining Catholicism, Bahá'í, and Sikhism while subtly discouraging leaving these faiths, and show the opposite for atheists, agnostics, and Jehovah's Witnesses. The study tested 20 models across 182 religious pairings, with reproducible results.

Article intelligence

EngineersAdvanced

Key points

Large language models show systematic bias in advising on religious conversions, favoring some faiths over others.
Study tested 20 commercial and open-source models across 182 religion pairings with reproducible asymmetries.
Catholicism, Bahá'í, and Sikhism are generally favored, while atheists and Jehovah's Witnesses are disfavored.
Model size and provider affect the degree of bias, with Grok 4.20 showing the strongest asymmetries.

Why it matters

This matters because large language models show systematic bias in advising on religious conversions, favoring some faiths over others.

Technical impact

May affect model selection, inference cost, product capability, and evaluation benchmarks.

[2605.22975] When AI Takes Sides on Questions of Faith: Persistent Asymmetries in AI-Mediated Faith Guidance

[Submitted on 21 May 2026]

Title:When AI Takes Sides on Questions of Faith: Persistent Asymmetries in AI-Mediated Faith Guidance

View a PDF of the paper titled When AI Takes Sides on Questions of Faith: Persistent Asymmetries in AI-Mediated Faith Guidance, by Brett Israelsen and 5 other authors

View PDF HTML (experimental)

Abstract:We ask whether large language models (LLMs) treat queries about religious conversion symmetrically. The answer is no. When asked for advice on hypothetical faith transitions from one religion to another, then asked the reversed question, models exhibited consistent asymmetries, favoring some religions while subtly discouraging conversion to others. On average Catholic, Bahá'í, and Sikh religions were broadly favored (high support for joining, low support for leaving), while Atheists, Agnostics, and Jehovah's Witnesses were primarily disfavored. Patterns varied by model size and model provider, with Grok 4.20 exhibiting the strongest asymmetries.

We tested 20 commercial and open-source language models across 182 religion pairings using a human-verified LLM-as-a-judge framework. Each model was probed via interactions with a simulated user asking for advice on a potential faith conversion. Models tended to use more encouraging language for some faith transitions over others; these patterns were systematically repeatable across multiple trials.

All LLMs tested exhibited reproducible asymmetry, though the pattern of preferences differed for each. Overall preferences persist across multiple question phrasings and variations in the religious pairing dataset. Taken together, these results suggest that asymmetry is a robust property of model behavior rather than an artifact of how the models' answers were scored. It is important to consider that any imbalances deployed and reproduced en masse can have real-world implications.

Comments: 29 pages, 16 figures

Subjects:

Computation and Language (cs.CL); Computers and Society (cs.CY)

Cite as: arXiv:2605.22975 [cs.CL]

(or arXiv:2605.22975v1 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2605.22975

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Brett Israelsen [view email] [v1] Thu, 21 May 2026 19:05:09 UTC (753 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled When AI Takes Sides on Questions of Faith: Persistent Asymmetries in AI-Mediated Faith Guidance, by Brett Israelsen and 5 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CL

new | recent | 2026-05

Change to browse by:

cs cs.CY

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)