Open and Emerging Issues in Agentic Privacy and Security: A Contextual Perspective
The report frames AI agent risks through contextual integrity and proposes LLM-driven policy engines to judge when actions and data flows are appropriate.
The report frames AI agent risks through contextual integrity and proposes LLM-driven policy engines to judge when actions and data flows are appropriate.