Glossary of AI Terminology

What Is Data Leakage?

Data leakage

Data leakage occurs when an AI system exposes information it should not reveal. This can include secrets, personal data, tenant data, internal documents, prompt contents, tool outputs, or retrieved context from the wrong permission boundary.

Agents increase leakage risk because they can access tools and data sources. Evaluation should test permission boundaries, source filtering, redaction, and whether the model reveals hidden context or system instructions.

Bi-weekly AI Research Paper Readings

Stay on top of emerging trends and frameworks.