The Tower of Babel Revisited: Multilingual Jailbreak Prompts on Closed-Source Large Language Models
This work addresses security vulnerabilities in proprietary LLMs for developers and policymakers, though it is incremental as it extends existing jailbreak research to closed-source models.
The paper systematically evaluated the vulnerability of closed-source large language models (GPT-4o, DeepSeek-R1, Gemini-1.5-Pro, Qwen-Max) to multilingual jailbreak attacks, finding that Qwen-Max was most vulnerable with Chinese prompts yielding higher attack success rates than English ones.
Large language models (LLMs) have seen widespread applications across various domains, yet remain vulnerable to adversarial prompt injections. While most existing research on jailbreak attacks and hallucination phenomena has focused primarily on open-source models, we investigate the frontier of closed-source LLMs under multilingual attack scenarios. We present a first-of-its-kind integrated adversarial framework that leverages diverse attack techniques to systematically evaluate frontier proprietary solutions, including GPT-4o, DeepSeek-R1, Gemini-1.5-Pro, and Qwen-Max. Our evaluation spans six categories of security contents in both English and Chinese, generating 38,400 responses across 32 types of jailbreak attacks. Attack success rate (ASR) is utilized as the quantitative metric to assess performance from three dimensions: prompt design, model architecture, and language environment. Our findings suggest that Qwen-Max is the most vulnerable, while GPT-4o shows the strongest defense. Notably, prompts in Chinese consistently yield higher ASRs than their English counterparts, and our novel Two-Sides attack technique proves to be the most effective across all models. This work highlights a dire need for language-aware alignment and robust cross-lingual defenses in LLMs, and we hope it will inspire researchers, developers, and policymakers toward more robust and inclusive AI systems.