# Auditing demographic bias in AI-based emergency police dispatch: a cross-lingual evaluation of eleven large language models

arXiv: https://arxiv.org/abs/2605.01451 · HTML: https://mlsanity.com/paper/2605.01451/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 31 references that are in the map.
- First posted on arXiv 2 May 2026.

## Builds on (most cited first, 9 of 9)

- [ChatGPT Outperforms Crowd-Workers for Text-Annotation Tasks](https://mlsanity.com/paper/2303.15056/) (2023)
- [Bias and Fairness in Large Language Models: A Survey](https://mlsanity.com/paper/2309.00770/) (2023)
- [The Capacity for Moral Self-Correction in Large Language Models](https://mlsanity.com/paper/2302.07459/) (2023)
- [Bias in Large Language Models: Origin, Evaluation, and Mitigation](https://mlsanity.com/paper/2411.10915/) (2024)
- [INVESTORBENCH: A Benchmark for Financial Decision-Making Tasks with LLM-based Agent](https://mlsanity.com/paper/2412.18174/) (2024)
- [Gender Bias in Masked Language Models for Multiple Languages](https://mlsanity.com/paper/2205.00551/) (2022)
- [Large Language Models are Biased Because They Are Large Language Models](https://mlsanity.com/paper/2406.13138/) (2024)
- [As an AI Language Model, "Yes I Would Recommend Calling the Police": Norm Inconsistency in LLM Decision-Making](https://mlsanity.com/paper/2405.14812/) (2024)
- [Echoes of Power: Investigating Geopolitical Bias in US and China Large Language Models](https://mlsanity.com/paper/2503.16679/) (2025)

## Abstract (arXiv)

Large language models (LLMs) are rapidly being integrated into high-stakes public safety systems, including emergency call triage and dispatch decision support, yet their demographic fairness in this context remains largely untested. Here we introduce a cross-lingual audit framework that operationalizes the Police Priority Dispatch System as a five-level ordinal classification task and applies a controlled minimal-pair design to isolate the effect of demographic cues. Across 19,800 model outputs spanning 11 frontier models, 15 scenario pairs, three demographic categories (religious appearance, gender, and race), and two languages (English and Mandarin Chinese), we find that demographic bias emerges systematically when incident severity is ambiguous but largely disappears when the operational priority is clearly determined by call content. Bias magnitude varies by demographic axis, with the largest effects observed for religious appearance, followed by gender and race. Critically, bias does not transfer consistently across languages: gender bias is substantially amplified in Mandarin Chinese, whereas race bias is more pronounced in English, revealing cross-lingual asymmetries that aggregate analyses obscure. In several scenarios, demographic cues produce counter-directional effects, challenging simple stereotype-amplification accounts of model behavior. These findings suggest that bias in LLM-based dispatch is not a fixed property of models alone, but arises from the interaction between demographic signals, contextual ambiguity, and language. Beyond these empirical results, the proposed framework provides a scalable audit infrastructure that enables deploying agencies to evaluate candidate models on jurisdiction-relevant scenarios prior to real-world adoption.