Differential Privacy
A mathematical framework that provides provable privacy guarantees when analyzing or learning from data. It ensures that the output of any analysis is approximately the same whether or not any individual's data is included.
Why It Matters
Differential privacy provides formal, mathematical privacy guarantees — not just policies or promises. It is the gold standard for privacy-preserving data analysis.
Example
Apple using differential privacy in iOS to learn typing patterns and emoji usage trends without being able to identify any individual user's behavior.
Think of it like...
Like a survey where random noise is added to each response — the overall statistics are still accurate, but no individual's specific answer can be determined.
Related Terms
Federated Learning
A decentralized training approach where a model is trained across multiple devices or organizations without sharing raw data. Each participant trains locally and only shares model updates.
Data Privacy
The right of individuals to control how their personal information is collected, used, stored, and shared. In AI, data privacy concerns arise from training data, user interactions, and model outputs.