Skip to content

Security & Privacy

Differential privacy

A mathematical way to limit how much a shared result can reveal about any one person's participation.

Example

An organization adds carefully controlled randomness to shared statistics.

Why people use it

It limits how much a published analysis can reveal about one person's participation.

What you'll hear

“How much could this result reveal about one individual?”

What this means for you

Ask what privacy protection is promised and how it affects the usefulness of the figures.

Can you control it?

Developer-only

The people building or running the AI choose this setup. An everyday user generally needs their help to change how this part works.

Common questions

Does differential privacy mean the result is perfectly exact and reveals nothing?
No. It provides a quantified privacy guarantee with a utility tradeoff.
Can stronger protection reduce detail?
Yes. Limiting what can be inferred can make the released result less precise or useful.
Does it simply mean removing names?
No. It is a mathematical approach to limiting information leakage, not just deleting obvious identifiers.

Related terms

Still have questions?

Up to 500 characters.

Ask LATHIC about AI. Relevant glossary entries may be included.

Your question, the glossary entries it matches, and a rotating pseudonymous identifier go to Microsoft Azure’s OpenAI service through Vercel AI Gateway to generate an answer. Zero retention and no training are required of the provider, and LATHIC does not save your question or answer. Privacy Notice