🤖 AI Summary
This work addresses the inability of current AI agents to effectively and securely employ cryptographic primitives in everyday interactions to resolve privacy and trust challenges. We propose the first framework enabling agents to autonomously identify, negotiate, and execute appropriate cryptographic protocols—such as those for anonymous credentials or fair payment splitting—to achieve privacy-preserving interactions. To evaluate this capability, we introduce a comprehensive benchmark encompassing protocol identification, multi-agent negotiation, execution, computational feasibility, and security assessment. We further develop data generation pipelines and supervised fine-tuning (SFT) strategies to enhance agent performance. Experimental results demonstrate that fine-tuned models significantly outperform baseline approaches on this benchmark, validating the efficacy of our proposed framework.
📝 Abstract
We often assume that agent-to-agent interaction will mirror human conversation. However, agents operate fundamentally differently. What if they could develop communication patterns that are more efficient and better aligned with their capabilities? While cryptographic primitives that could profoundly improve everyday interactions already exist, humans can't use them because they are too complex and the math can't be done in one's head. Examples range from proving your age (or other attributes) without showing your ID, to filing an anonymous report within a group while proving you are a legitimate member, to splitting a dinner bill fairly without revealing salaries. What if agents could create protocols"on the fly"by recognizing which primitive fits an everyday situation, proposing it to an agentic counterpart, persuading them to participate, and then executing the protocol correctly using appropriate computation tools? Protocol Agent frames this problem by introducing a benchmark that spans: (1) cryptographic primitive recognition, (2) negotiation skills, (3) implementation correctness, (4) correct computation and (5) security strength. We evaluate current open-weight and state-of-the-art models on this benchmark, propose a dataset-generation approach to improve these capabilities, and measure the impact of supervised fine-tuning (SFT) on benchmark performance, with tuned models outperforming base models by a wide margin.