Benchmark RadarAI BENCHMARK PROFILE
IDP-Bench
IDP-Bench evaluates large language models on interdependent privacy scenarios, covering recognition of co-ownership, identification of contextual integrity parameters, and judgments of sharing appropriateness.
- Released
- 2026-06-06
- Readiness
- Runnable
- Primary field
- General AI
Why it matters
Interdependent privacy is a critical yet underexplored risk when LLMs act as personal assistants; a reusable benchmark enables systematic comparison and improvement of models' privacy reasoning in shared-data contexts.
Motivation
Large language models (LLMs) are becoming widely deployed as personal AI assistants with access to sensitive user data, making privacy a major challenge for their design and evaluation.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.