{"version":1,"type":"story","url":"https://digestai.news/story/researchers-release-agentciscope-to-test-llm-agents-data-privacy-risks","json":"https://digestai.news/story/researchers-release-agentciscope-to-test-llm-agents-data-privacy-risks.json","markdown":"https://digestai.news/story/researchers-release-agentciscope-to-test-llm-agents-data-privacy-risks.md","slug":"researchers-release-agentciscope-to-test-llm-agents-data-privacy-risks","headline":"Researchers release AgentCIScope to test LLM agents’ data privacy risks","summary":"Researchers introduced **AgentCIScope**, a new benchmark platform to evaluate how LLM-powered agents handle sensitive user data across real-world applications. The tool tests agents on 936 executable tasks spanning email, calendars, and messaging, scoring them on whether they comply with privacy policies, user consent, and predefined rules for handling personal data like credit cards or Social Security numbers.\n\nThe study found that even capable agents leak sensitive information at rates between 32.5% and 94.4% per task, with leakage worsening when tasks are repeated. Personal preferences and booking details were most frequently exposed, while credit-card data and SSNs were handled more securely. The work highlights a gap between an agent’s ability to complete tasks and its adherence to privacy constraints, offering a framework for developers to build safer systems.","keyPoints":["AgentCIScope tests 12 LLM agents on 936 tasks across email, calendars, and messaging apps","Leakage rates range from 32.5% to 94.4% per task, rising further with repeated attempts","Personal data like preferences and bookings leak far more often than credit cards or SSNs"],"whyItMatters":"This benchmark exposes a critical flaw in AI agents: high task performance doesn’t guarantee privacy. Businesses using agents for customer data may face compliance risks, and developers now have a tool to audit and improve their systems.","category":{"slug":"research","name":"Research","url":"https://digestai.news/category/research"},"entities":{"companies":[],"models":[],"people":[]},"firstPublishedAt":"2026-10-15T04:00:00Z","updatedAt":"2026-10-15T04:00:00Z","sourceCount":1,"hasPrimarySource":true,"sources":[{"outlet":"arXiv cs.AI","title":"AgentCIScope: Grounded Contextual Integrity Evaluation for LLM Agents","url":"https://arxiv.org/abs/2610.19158","publishedAt":"2026-10-15T04:00:00Z","type":"primary","primary":true,"lead":true}],"sourceNotes":null,"discussions":[],"thread":null,"cite":{"text":"Digest AI, \"Researchers release AgentCIScope to test LLM agents’ data privacy risks\", 15 October 2026, https://digestai.news/story/researchers-release-agentciscope-to-test-llm-agents-data-privacy-risks","publisher":"Digest AI","title":"Researchers release AgentCIScope to test LLM agents’ data privacy risks","datePublished":"2026-10-15T04:00:00Z","url":"https://digestai.news/story/researchers-release-agentciscope-to-test-llm-agents-data-privacy-risks"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}