安全团队和LLM开发者需要警惕：多语言安全漏洞远未被堵住，低资源语言成为新的攻击面。建议立即检查自家模型的多轮对话安全机制。
研究发现，使用低资源非洲语言（如南非荷兰语、斯瓦希里语、科萨语和祖鲁语）进行多轮对话，可以绕过ChatGPT、Claude、DeepSeek、Gemini和Grok等商业大模型的安全护栏。单轮翻译攻击无效，但多轮对话在英语上达到52.7%-83.6%的有害响应率，非洲语言也达到41.8%-78.2%。人工红队测试比自动化方法越狱率更高，且翻译质量是决定越狱成功的关键因素——翻译质量越差，越狱率越低。
Multilingual jailbreaking of LLMs using low-resource languages
Large Language Models (LLMs) remain vulnerable to jailbreak attempts that circumvent safety guardrails. We investigate whether multi-turn conversations using low-resource African languages (Afrikaans, Kiswahili, isiXhosa, and isiZulu) can bypass safety mechanisms across commercial LLMs. We translated prompts from existing datasets and evaluated ChatGPT, Claude, DeepSeek, Gemini, and Grok through automated testing and human red-teaming with native speakers. Single-turn translation attacks proved ineffective, while multi-turn conversations achieved English harmful response rates from 52.7% (Claude 3.5 Haiku) to 83.6% (GPT-4o-mini), Afrikaans from 60.0% (Claude 3.5 Haiku) to 78.2% (GPT-4o-mini), and Kiswahili from 41.8% (Claude 3.5 Haiku) to 70.9% (DeepSeek). Human red-teaming increased jailbreak rates compared to automated methods. Over all evaluated languages, the average jailbreak rate increased from 59.8% to 75.8%, with improvements of +20.0% (Afrikaans), +12.7% (isiZulu), +12.3% (isiXhosa), and +1% (Kiswahili), demonstrating that poor translation quality limits jailbreak success. These findings suggest that vulnerabilities in LLMs persist in multilingual contexts and that translation quality is the critical factor determining jailbreak success in low-resource languages.