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：强化学习从人类反馈（RLHF）被用于对齐大型语言模型，使其生成有益且无害的响应。然而，先前的研究表明，这些模型可以通过寻找对抗性提示来被越狱，从而恢复其未对齐的行为。在本文中，我们考虑一种新的威胁：攻击者污染RLHF训练数据，将“越狱后门”嵌入到模型中。该后门在模型中植入一个触发词，其作用类似于通用的“sudo命令”：在任何提示中添加触发词即可生成有害响应，而无需寻找对抗性提示。通用越狱后门比先前研究的针对语言模型的后门强大得多，我们发现使用常见的后门攻击技术更难植入这类后门。我们研究了RLHF中导致其被认为具有鲁棒性的设计决策，并发布了一个被污染模型的基准测试，以激发未来对通用越狱后门的研究。