Place Matters: Comparing LLM Hallucination Rates for Place-Based Legal Queries

TL;DR

This paper introduces a methodology to compare hallucination rates of large language models across different legal jurisdictions, revealing geographic disparities in legal information accuracy and proposing a measure of model uncertainty.

Contribution

It presents a novel approach using the concept of functionalism to quantify legal knowledge differences across places in LLMs, with a new dataset and evaluation method for hallucinations.

Findings

Hallucination rates vary significantly by location.

A strong negative correlation exists between hallucination rate and majority response frequency.

The proposed measure can indicate the uncertainty of legal fact predictions.

Abstract

How do we make a meaningful comparison of a large language model's knowledge of the law in one place compared to another? Quantifying these differences is critical to understanding if the quality of the legal information obtained by users of LLM-based chatbots varies depending on their location. However, obtaining meaningful comparative metrics is challenging because legal institutions in different places are not themselves easily comparable. In this work we propose a methodology to obtain place-to-place metrics based on the comparative law concept of functionalism. We construct a dataset of factual scenarios drawn from Reddit posts by users seeking legal advice for family, housing, employment, crime and traffic issues. We use these to elicit a summary of a law from the LLM relevant to each scenario in Los Angeles, London and Sydney. These summaries, typically of a legislative provision, are manually evaluated for hallucinations. We show that the rate of hallucination of legal information by leading closed-source LLMs is significantly associated with place. This suggests that the quality of legal solutions provided by these models is not evenly distributed across geography. Additionally, we show a strong negative correlation between hallucination rate and the frequency of the majority response when the LLM is sampled multiple times, suggesting a measure of uncertainty of model predictions of legal facts.