As an AI Language Model, "Yes I Would Recommend Calling the Police": Norm Inconsistency in LLM Decision-Making

TL;DR

This paper examines how large language models exhibit norm inconsistency and bias in high-stakes surveillance decisions, revealing significant disparities and limitations in current mitigation strategies.

Contribution

It provides a systematic analysis of norm inconsistency and racial bias in LLM decision-making regarding police calls in surveillance videos.

Findings

LLMs often recommend calling the police even when no criminal activity is present.

Decisions are biased by the racial demographics of neighborhoods.

Current bias detection and mitigation strategies are insufficient.

Abstract

We investigate the phenomenon of norm inconsistency: where LLMs apply different norms in similar situations. Specifically, we focus on the high-risk application of deciding whether to call the police in Amazon Ring home surveillance videos. We evaluate the decisions of three state-of-the-art LLMs -- GPT-4, Gemini 1.0, and Claude 3 Sonnet -- in relation to the activities portrayed in the videos, the subjects' skin-tone and gender, and the characteristics of the neighborhoods where the videos were recorded. Our analysis reveals significant norm inconsistencies: (1) a discordance between the recommendation to call the police and the actual presence of criminal activity, and (2) biases influenced by the racial demographics of the neighborhoods. These results highlight the arbitrariness of model decisions in the surveillance context and the limitations of current bias detection and mitigation strategies in normative decision-making.