Is Safer Better? The Impact of Guardrails on the Argumentative Strength of LLMs in Hate Speech Countering

TL;DR

This paper investigates how safety guardrails in large language models affect the quality of counterspeech for hate speech mitigation, finding that guardrails can hinder argumentative strength while targeted attacks on hate components improve response quality.

Contribution

It provides an empirical analysis of the impact of safety guardrails and targeted hate component attacks on counterspeech generation quality in LLMs.

Findings

Safety guardrails can reduce the argumentative quality of counterspeech.

Attacking specific hate speech components yields more effective responses.

Guardrails may hinder the generation of socially positive interactions.

Abstract

The potential effectiveness of counterspeech as a hate speech mitigation strategy is attracting increasing interest in the NLG research community, particularly towards the task of automatically producing it. However, automatically generated responses often lack the argumentative richness which characterises expert-produced counterspeech. In this work, we focus on two aspects of counterspeech generation to produce more cogent responses. First, by investigating the tension between helpfulness and harmlessness of LLMs, we test whether the presence of safety guardrails hinders the quality of the generations. Secondly, we assess whether attacking a specific component of the hate speech results in a more effective argumentative strategy to fight online hate. By conducting an extensive human and automatic evaluation, we show how the presence of safety guardrails can be detrimental also to a task that inherently aims at fostering positive social interactions. Moreover, our results show that attacking a specific component of the hate speech, and in particular its implicit negative stereotype and its hateful parts, leads to higher-quality generations.