Google Research 发布智能体隐私与安全开放问题报告
原标题：Open and Emergent Problems in Agentic Privacy and Security: A Contextual Angle
AI 摘要
Google Research 发布了一份关于智能体隐私与安全的新研讨会报告，该报告由 50 多位学术界和产业界领袖在 2025 年底于纽约举行的 Google CAPS 研讨会上共同完成。报告以情境完整性（Contextual Integrity）理论为基础，将隐私重新定义为符合社会规范的“适当信息流动”，并扩展到智能体行为的“情境安全”。报告提出构建情境策略引擎作为监督层，在行动执行前评估其适当性，并呼吁建立标准化的多智能体基准（如“Agent Gym”）以推动跨领域协作。
正文节选
October 5, 2026 Eugene Bagdasarian, Research Scientist, and Marco Gruteser, Principal Scientist, Google Research To be useful, AI agents must understand and be constrained by contextual behavioral norms to ensure they act appropriately. Inspired by the theory of Contextual Integrity, our new workshop report outlines key open research directions across system, model, and user levels to build AI agents users can trust. We're rapidly transitioning to a computing landscape defined by highly general,