Compromesso! Italian Many-Shot Jailbreaks Undermine the Safety of Large Language Models

TL;DR

This paper investigates the safety vulnerabilities of large language models in Italian, revealing that many-shot jailbreaking prompts can induce unsafe behaviors even with few demonstrations, escalating with more prompts.

Contribution

It introduces a new Italian unsafe question-answer dataset and demonstrates that open-weight LLMs are vulnerable to many-shot jailbreaking in Italian, highlighting safety concerns across languages.

Findings

Models exhibit unsafe behaviors with few unsafe demonstrations.

Unsafe tendencies escalate rapidly with more demonstrations.

Safety vulnerabilities are identified in four families of open-weight LLMs.

Abstract

As diverse linguistic communities and users adopt large language models (LLMs), assessing their safety across languages becomes critical. Despite ongoing efforts to make LLMs safe, they can still be made to behave unsafely with jailbreaking, a technique in which models are prompted to act outside their operational guidelines. Research on LLM safety and jailbreaking, however, has so far mostly focused on English, limiting our understanding of LLM safety in other languages. We contribute towards closing this gap by investigating the effectiveness of many-shot jailbreaking, where models are prompted with unsafe demonstrations to induce unsafe behaviour, in Italian. To enable our analysis, we create a new dataset of unsafe Italian question-answer pairs. With this dataset, we identify clear safety vulnerabilities in four families of open-weight LLMs. We find that the models exhibit unsafe behaviors even when prompted with few unsafe demonstrations, and -- more alarmingly -- that this tendency rapidly escalates with more demonstrations.