Pith. sign in

REVIEW 7 cited by

Persistent Anti-Muslim Bias in Large Language Models

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2101.05783 v2 pith:TNOMMVTR submitted 2021-01-14 cs.CL cs.LG

Persistent Anti-Muslim Bias in Large Language Models

classification cs.CL cs.LG
keywords biaslanguagereligiousanti-muslimbeenbiasescasesgpt-3
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

It has been observed that large-scale language models capture undesirable societal biases, e.g. relating to race and gender; yet religious bias has been relatively unexplored. We demonstrate that GPT-3, a state-of-the-art contextual language model, captures persistent Muslim-violence bias. We probe GPT-3 in various ways, including prompt completion, analogical reasoning, and story generation, to understand this anti-Muslim bias, demonstrating that it appears consistently and creatively in different uses of the model and that it is severe even compared to biases about other religious groups. For instance, "Muslim" is analogized to "terrorist" in 23% of test cases, while "Jewish" is mapped to "money" in 5% of test cases. We quantify the positive distraction needed to overcome this bias with adversarial text prompts, and find that use of the most positive 6 adjectives reduces violent completions for "Muslims" from 66% to 20%, but which is still higher than for other religious groups.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 7 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. The Alignment Veto: How Safety Training Suppresses Cultural Knowledge in LLMs

    cs.CL 2025-10 reject novelty 6.0

    The full text builds the MENA Values benchmark (864 questions, 7 models) and reports that LLM cultural answers shift with language, decline with reasoning prompts, and hide strong internal preferences behind refusals—...

  2. PaLM: Scaling Language Modeling with Pathways

    cs.CL 2022-04 accept novelty 6.0

    PaLM 540B demonstrates continued scaling benefits by setting new few-shot SOTA results on hundreds of benchmarks and outperforming humans on BIG-bench.

  3. LaMDA: Language Models for Dialog Applications

    cs.CL 2022-01 unverdicted novelty 6.0

    LaMDA shows that fine-tuning on human-value annotations and consulting external knowledge sources significantly improves safety and factual grounding in large dialog models beyond what scaling alone achieves.

  4. Ethical and social risks of harm from Language Models

    cs.CL 2021-12 accept novelty 6.0

    The authors provide a detailed taxonomy of 21 risks associated with language models, covering discrimination, information leaks, misinformation, malicious applications, interaction harms, and societal impacts like job...

  5. The Geography of Algorithmic Judgment: LLM Intermediaries, Place Identity, and Racial Steering in Housing Search

    cs.LG 2026-06 unverdicted novelty 5.0

    Behavioral audit finds emergent, city-dependent racial steering in LLM housing recommendations that changes with user identity and preference context.

  6. Measuring the Machine: Evaluating Generative AI as Pluralist Sociotechical Systems

    cs.AI 2026-04 unverdicted novelty 5.0

    Generative AI must be evaluated as recursive pluralist sociotechnical systems via MaSH Loops and distributional World Values Benchmarks instead of static functionalist or prescriptive tests.

  7. PaLM 2 Technical Report

    cs.CL 2023-05 unverdicted novelty 5.0

    PaLM 2 reports state-of-the-art results on language, reasoning, and multilingual tasks with improved efficiency over PaLM.