PrivacyBench: 一个用于评估个性化AI隐私的对话基准
PrivacyBench: A Conversational Benchmark for Evaluating Privacy in Personalized AI
AI总结：
PrivacyBench通过多轮对话评估揭示RAG助手在隐私保护上的缺陷，指出需结构性隐私保障措施以确保伦理网络环境。
摘要：
个性化AI代理依赖于访问用户的数字足迹，这通常包括来自私人电子邮件、聊天和购买历史的敏感数据。然而，这种访问创建了根本的社会和隐私风险：缺乏社会情境意识的系统可能会无意中暴露用户的秘密，威胁数字福祉。我们介绍了PrivacyBench，一个包含嵌入秘密的社会化数据集和多轮对话评估的基准，用于测量秘密保护。测试检索增强生成（RAG）助手发现，它们在多达26.56%的交互中泄露秘密。一个隐私意识提示将泄漏降低到5.12%，但这种措施只能提供部分缓解。检索机制仍然会随意访问敏感数据，这将整个隐私保护的负担转移到生成器上。这创造了一个单一故障点，使当前架构在大规模部署时变得不安全。我们的发现强调了对结构化、隐私优先设计的保护措施的紧迫需求，以确保一个道德且包容的网络环境。
英文摘要：
Personalized AI agents rely on access to a user's digital footprint, which often includes sensitive data from private emails, chats and purchase histories. Yet this access creates a fundamental societal and privacy risk: systems lacking social-context awareness can unintentionally expose user secrets, threatening digital well-being. We introduce PrivacyBench, a benchmark with socially grounded datasets containing embedded secrets and a multi-turn conversational evaluation to measure secret preservation. Testing Retrieval-Augmented Generation (RAG) assistants reveals that they leak secrets in up to 26.56% of interactions. A privacy-aware prompt lowers leakage to 5.12%, yet this measure offers only partial mitigation. The retrieval mechanism continues to access sensitive data indiscriminately, which shifts the entire burden of privacy preservation onto the generator. This creates a single point of failure, rendering current architectures unsafe for wide-scale deployment. Our findings underscore the urgent need for structural, privacy-by-design safeguards to ensure an ethical and inclusive web for everyone.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
