noun
Differential privacy is a technique that adds controlled noise to data to protect individual identities while allowing useful analysis. It ensures that the presence or absence of any single person in a dataset doesn’t significantly change the results.
Developed by cryptographers, it’s used in surveys, medical research, and tech companies to balance privacy and data utility. For example, Apple uses it to analyze iPhone usage patterns without tracking individual users.
For data analysts, it means they must accept slightly less precise results in exchange for stronger privacy guarantees. For regulators, it provides a way to enforce anonymity without requiring full data deletion.
A hospital uses differential privacy to share patient data for research, ensuring no single patient can be identified.
In other languages
See also: GDPR · encryption