AI Governance Engineer
GitHub

Differential privacy

By Updated

A mathematical guarantee that bounds how much any single person's record can change the output of an analysis or a trained model, tuned by a privacy budget. Its strength depends on the budget and on implementation choices that NIST calls privacy hazards 1.

Developed in
ch. 19, Privacy-enhancing technologies and their honest limits
Chapters
ch. 19, Privacy & AI
Source
1 numbered reference, listed below

Where it is used

3 chapters of the Body of Knowledge use the term. Each link opens the first section that does.

  • Federated learning

    Training a model across devices or sites where the data lives, sharing model updates instead of raw records.

  • Privacy-enhancing technology (PET)

    A technique that reduces what an attacker, vendor or insider can learn from personal data, such as differential privacy, federated learning, synthetic data…

Sources

  1. [1] NIST SP 800-226, Guidelines for Evaluating Differential Privacy Guarantees (differential privacy pyramid; privacy hazards). NIST. 2025-03. https://csrc.nist.gov/pubs/sp/800/226/final (verified: primary)

Definitions of legal terms paraphrase the cited text, which governs. Dated statements are as of .

Cite this term

García Aibar, J. (2026). Differential privacy. In AI Governance Engineering: The Thesis & Body of Knowledge (v0.5.0), Glossary. https://doi.org/10.5281/zenodo.22956197. https://aigovernanceengineer.com/glossary/differential-privacy. CC BY 4.0

BibTeX

@misc{aige2026differentialprivacy,
  author  = {Jorge García Aibar},
  title   = {{Differential privacy}},
  note    = {Glossary, AI Governance Engineering: The Thesis \& Body of Knowledge, version 0.5.0},
  year    = {2026},
  doi     = {10.5281/zenodo.22956197},
  url     = {https://aigovernanceengineer.com/glossary/differential-privacy}
}