[论文解读] Clio: Privacy-Preserving Insights into Real-World AI Use

Alex Tamkin, Miles McCain+19 more|arXiv (Cornell University)|Dec 18, 2024

Privacy-Preserving Technologies in Data被引用 5

一句话总结

Clio 通过隐私保护方法从对话中揭示聚合模式，分析真实世界的 AI 使用情况，在不暴露私人数据的前提下实现高准确性并支持安全洞察。

ABSTRACT

How are AI assistants being used in the real world? While model providers in theory have a window into this impact via their users' data, both privacy concerns and practical challenges have made analyzing this data difficult. To address these issues, we present Clio (Claude insights and observations), a privacy-preserving platform that uses AI assistants themselves to analyze and surface aggregated usage patterns across millions of conversations, without the need for human reviewers to read raw conversations. We validate this can be done with a high degree of accuracy and privacy by conducting extensive evaluations. We demonstrate Clio's usefulness in two broad ways. First, we share insights about how models are being used in the real world from one million Claude.ai Free and Pro conversations, ranging from providing advice on hairstyles to providing guidance on Git operations and concepts. We also identify the most common high-level use cases on Claude.ai (coding, writing, and research tasks) as well as patterns that differ across languages (e.g., conversations in Japanese discuss elder care and aging populations at higher-than-typical rates). Second, we use Clio to make our systems safer by identifying coordinated attempts to abuse our systems, monitoring for unknown unknowns during critical periods like launches of new capabilities or major world events, and improving our existing monitoring systems. We also discuss the limitations of our approach, as well as risks and ethical concerns. By enabling analysis of real-world AI usage, Clio provides a scalable platform for empirically grounded AI safety and governance.

研究动机与目标

• 研究 AI 助手在实际使用和跨语言中的用途
• 在不需要人工审查原始数据的前提下提供可扩展、隐私保护的洞察
• 展示 Clio 在理解使用模式和提升安全性方面的效用
• 讨论隐私保护洞察的局限性、风险和伦理考量
• 提供经验证据以支持 AI 系统的治理与安全监控

提出的方法

• 从每次对话中提取多方面信息，如主题和语言
• 使用嵌入和在方面的 k-means 对对话进行语义聚类
• 用标题和摘要描述聚类，同时省略私人信息
• 构建多层次层级，组织数千个聚类以实现可扩展的探索
• 提供交互式的二维可视化与层级结构，便于理解与发现
• 实现四层隐私保护，以在整个流程中降低私人信息的暴露

实验结果

研究问题

• RQ1 在真实世界的 AI 助手对话中，出现哪些高层次和细粒度的使用模式？
• RQ2 在保护隐私的同时，Clio 能多准确地重构主题分布并揭示洞察？
• RQ3 Clio 如何有助于安全监控、异常检测和分类器评估？
• RQ4 使用模式如何在语言和社群之间变化？

主要发现

• 编码与写作任务占主导，网页与移动开发对话占比超过 10%
• 日语和中文对话中，长者照护主题的普及度高于英语
• Clio 能在合成的多语言数据上以 94% 的准确率重构真实主题分布
• 在高不确定性时期（如新发布或选举）
• Clio 能识别协同行为滥用和未知未知
• 处理 100,000 次对话的成本估计为每次 48.81 美元，显示可扩展性
• Clio 输出保持隐私，采用分层防护将私人信息暴露降至不可检测水平