AgentSCOPE: Evaluating Contextual Privacy Across Agentic Workflows
#AgentSCOPE #contextual privacy #agentic workflows #privacy evaluation #AI agents #data protection #automated systems
📌 Key Takeaways
- AgentSCOPE is a framework for evaluating privacy in agentic workflows.
- It focuses on contextual privacy, assessing how context affects data protection.
- The tool helps identify privacy risks in automated, multi-agent systems.
- It aims to improve privacy safeguards in AI-driven workflows.
📖 Full Retelling
arXiv:2603.04902v1 Announce Type: cross
Abstract: Agentic systems are increasingly acting on users' behalf, accessing calendars, email, and personal files to complete everyday tasks. Privacy evaluation for these systems has focused on the input and output boundaries, but each task involves several intermediate information flows, from agent queries to tool responses, that are not currently evaluated. We argue that every boundary in an agentic pipeline is a site of potential privacy violation and
🏷️ Themes
Privacy Evaluation, AI Agents
📚 Related People & Topics
AI agent
Systems that perform tasks without human intervention
In the context of generative artificial intelligence, AI agents (also referred to as compound AI systems or agentic AI) are a class of intelligent agents distinguished by their ability to operate autonomously in complex environments. Agentic AI tools prioritize decision-making over content creation ...
Entity Intersection Graph
Connections for AI agent:
🏢
OpenAI
6 shared
🌐
Large language model
4 shared
🌐
Reinforcement learning
3 shared
🌐
OpenClaw
3 shared
🌐
Artificial intelligence
2 shared
Mentioned Entities
Original Source
--> Computer Science > Cryptography and Security arXiv:2603.04902 [Submitted on 5 Mar 2026] Title: AgentSCOPE: Evaluating Contextual Privacy Across Agentic Workflows Authors: Ivoline C. Ngong , Keerthiram Murugesan , Swanand Kadhe , Justin D. Weisz , Amit Dhurandhar , Karthikeyan Natesan Ramamurthy View a PDF of the paper titled AgentSCOPE: Evaluating Contextual Privacy Across Agentic Workflows, by Ivoline C. Ngong and 5 other authors View PDF HTML Abstract: Agentic systems are increasingly acting on users' behalf, accessing calendars, email, and personal files to complete everyday tasks. Privacy evaluation for these systems has focused on the input and output boundaries, but each task involves several intermediate information flows, from agent queries to tool responses, that are not currently evaluated. We argue that every boundary in an agentic pipeline is a site of potential privacy violation and must be assessed independently. To support this, we introduce the Privacy Flow Graph, a Contextual Integrity-grounded framework that decomposes agentic execution into a sequence of information flows, each annotated with the five CI parameters, and traces violations to their point of origin. We present AgentSCOPE, a benchmark of 62 multi-tool scenarios across eight regulatory domains with ground truth at every pipeline stage. Our evaluation across seven state-of-the-art LLMs show that privacy violations in the pipeline occur in over 80% of scenarios, even when final outputs appear clean (24%), with most violations arising at the tool-response stage where APIs return sensitive data indiscriminately. These results indicate that output-level evaluation alone substantially underestimates the privacy risk of agentic systems. Subjects: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI) Cite as: arXiv:2603.04902 [cs.CR] (or arXiv:2603.04902v1 [cs.CR] for this version) https://doi.org/10.48550/arXiv.2603.04902 Focus to learn more arXiv-issued DOI via DataCite (pe...
Read full article at source