{"ID":2844991,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.05359","arxiv_id":"2511.05359","title":"ConVerse: Benchmarking Contextual Safety in Agent-to-Agent Conversations","abstract":"As language models evolve into autonomous agents that act and communicate on behalf of users, ensuring safety in multi-agent ecosystems becomes a central challenge. Interactions between personal assistants and external service providers expose a core tension between utility and protection: effective collaboration requires information sharing, yet every exchange creates new attack surfaces. We introduce ConVerse, a dynamic benchmark for evaluating privacy and security risks in agent-agent interactions. ConVerse spans three practical domains (travel, real estate, insurance) with 12 user personas and over 864 contextually grounded attacks (611 privacy, 253 security). Unlike prior single-agent settings, it models autonomous, multi-turn agent-to-agent conversations where malicious requests are embedded within plausible discourse. Privacy is tested through a three-tier taxonomy assessing abstraction quality, while security attacks target tool use and preference manipulation. Evaluating seven state-of-the-art models reveals persistent vulnerabilities; privacy attacks succeed in up to 88% of cases and security breaches in up to 60%, with stronger models leaking more. By unifying privacy and security within interactive multi-agent contexts, ConVerse reframes safety as an emergent property of communication.","short_abstract":"As language models evolve into autonomous agents that act and communicate on behalf of users, ensuring safety in multi-agent ecosystems becomes a central challenge. Interactions between personal assistants and external service providers expose a core tension between utility and protection: effective collaboration requi...","url_abs":"https://arxiv.org/abs/2511.05359","url_pdf":"https://arxiv.org/pdf/2511.05359v1","authors":"[\"Amr Gomaa\",\"Ahmed Salem\",\"Sahar Abdelnabi\"]","published":"2025-11-07T15:49:49Z","proceeding":"cs.CR","tasks":"[\"cs.CR\",\"cs.CL\",\"cs.CY\"]","methods":"[\"Language Model\"]","has_code":false}