DATDAT-002 — Data Privacy and Protection Failures
Language Models Inferring Private Attributes Without Personal Data
4/5Sector: GovernmentGeography: GlobalStage: OperateIngested: —
Executive Summary
Large language models can correctly infer sensitive personal attributes such as race, sexuality, or religion from correlational patterns alone, without accessing an individual's private data. Government adoption of such systems creates direct exposure to discrimination liability and erosion of citizens' privacy rights.
Domain
Data Management
Blindspots in data quality, privacy, bias, lineage, lifecycle, and third-party data dependencies.
Source
MIT AI Risk Repository — Ethical and social risks of harm from language models (Weidinger2021) ↗https://airisk.mit.edu/
Could this happen in your organisation?
A Velinor AI Audit maps your active AI portfolio against the 50+ blindspots and benchmarks against documented sector failures like this one. A board-ready foresight document in 5 weeks.