Google Research has published an analysis of emerging privacy and security challenges associated with autonomous AI agents. The research identifies open problems and contextual factors that organizations need to consider when deploying agentic systems. The work appears aimed at establishing a framework for understanding and addressing potential risks in this rapidly evolving technology area.