A Generative Model of Contextual Integrity: Appropriate vs. Inappropriate Information Sharing
cs.AI
Submitted: 2026-11-05
Updated: 2026-11-05
Comments: Accepted as a main conference paper at NeurIPS 2026
Code: https://github.com/Mamoro98/contextual-integrity-neurips-code
License: http://creativecommons.org/licenses/by/4.0/
The gist: Contextual integrity (CI) holds that appropriate information flow depends on context-specific norms.
Terminology
Abstract
Contextual integrity (CI) holds that appropriate information flow depends on context-specific norms. The same disclosure may be appropriate in one context and inappropriate in another. We build a generative model of contextual integrity to study information sharing through simulated conversations. Two actors played by large language models exchange messages, with one holding private information and the other seeking it. We examine how sharing and refusal vary with the situation, the holder's decision logic, disposition, and the underlying language model. Actors shared information with the rightful recipients when it was needed, but also disclosed it to people who were not entitled to receive it. The model provides a way to study how disclosure and refusal develop through conversation as the situations and the actors vary.
Sources
- CI-Bench: Benchmarking Contextual Integrity of AI Assistants on Synthetic Data
- A Generative Model of Conspicuous Consumption and Status Signaling
- AgentLeak: A Benchmark for Internal-Channel Privacy Leakage in Multi-Agent LLM Systems
- Red-Teaming LLM Multi-Agent Systems via Communication Attacks
- Contextual Integrity in LLMs via Reasoning and Reinforcement Learning
- A Theory of Appropriateness That Accounts for Norms of Rationality
- Generative agent-based modeling with actions grounded in physical, social, or digital space using Concordia
- IP Leakage Attacks Targeting LLM-Based Multi-Agent Systems
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents