Benchmark RadarAI BENCHMARK PROFILE
MosaicLeaks
MosaicLeaks is a benchmark of 1,001 multi-hop research tasks to evaluate privacy leakage in deep research agents, but no artifacts are provided in this article.
- Released
- 2026-05-29
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
It addresses privacy risks in agentic workflows, but without a public release, external teams cannot use it.
Motivation
Deep research agents increasingly combine private local documents with external tools like web retrieval, creating a privacy risk: an agent's external queries may leak sensitive information from its local context.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.