Meng Tong, Kejiang Chen, Yuang Qi, Jie Zhang, Weiming Zhang, Nenghai Yu

Large language models (LLMs), such as ChatGPT, have simplified text generation tasks, yet their inherent privacy risks are increasingly garnering attention. While differential privacy techniques have been successfully applied to text classification tasks, the resultant semantic bias makes them unsuitable for text generation. Homomorphic encryption inference methods have also been introduced, however, the significant computational and communication costs limit their viability. Furthermore, closed-source, black-box models such as GPT-4 withhold their architecture, thwarting certain privacy-enhancing strategies such as splitting inference into local and remote and then adding noise when communicating. To overcome these challenges, we introduce PrivInfer, the first privacy-preserving inference framework for black-box LLMs in text generation. Inspired by human writing, PrivInfer employs differential privacy methods to generate perturbed prompts for remote LLMs inference and extracts the meaningful response from the remote perturbed results. We also introduce RANTEXT, a differential privacy scheme specifically for LLMs that leverages random adjacency in text perturbations. Experimental results indicate that PrivInfer is comparable to GPT-4 in terms of text generation quality while protecting privacy, and RANTEXT provides enhanced privacy protection against three types of differential privacy attacks, including our newly introduced GPT inference attack, compared to baseline methods.