Reinforcement Learning from Human Feedback (RLHF) is used to align large language models to produce helpful and harmless responses. Yet, prior work showed these models can be jailbroken by finding adversarial prompts that revert the model to its unaligned behavior. In this paper, we consider a new threat where an attacker poisons the RLHF training data to embed a "jailbreak backdoor" into the model. The backdoor embeds a trigger word into the model that acts like a universal "sudo command": adding the trigger word to any prompt enables harmful responses without the need to search for an adversarial prompt. Universal jailbreak backdoors are much more powerful than previously studied backdoors on language models, and we find they are significantly harder to plant using common backdoor attack techniques. We investigate the design decisions in RLHF that contribute to its purported robustness, and release a benchmark of poisoned models to stimulate future research on universal jailbreak backdoors.

翻译：从人类反馈的强化学习（RLHF）用于对齐大型语言模型，以生成有益且无害的响应。然而，先前研究表明，这些模型可能通过寻找对抗性提示来越狱，从而将模型恢复为其未对齐的行为。在本文中，我们考虑一种新威胁：攻击者毒化RLHF训练数据，将“越狱后门”植入模型中。该后门在模型中嵌入一个触发词，其作用类似于通用的“sudo命令”：将触发词添加到任何提示中，即可生成有害响应，无需搜索对抗性提示。通用越狱后门比先前研究的语言模型后门更为强大，且我们发现使用常见的后门攻击技术更难植入。我们探讨了RLHF中有助于其所谓鲁棒性的设计决策，并发布了一个中毒模型基准，以促进未来对通用越狱后门的研究。