Knowledge MaxxingDifferential privacy
Differential privacy adds carefully calibrated statistical noise to data or query results, so no individual's presence in the dataset can be confidently determined from the output.
Prerequisite: Pseudonymisation
What matters
Because the added noise is mathematically bounded, it provides a formal, provable privacy guarantee, rather than relying on hoping the data simply looks anonymous enough.
Example
A differentially private survey result might report an aggregate average with a small amount of statistical noise added, preventing any one individual's specific answer from being deduced.
Common misconception
Differential privacy is not the same as simply removing identifying fields; it's a mathematical technique providing a measurable, provable guarantee against re-identification.
Recall question
What kind of guarantee does differential privacy provide that simple field removal does not?