This section is currently being developed. Additional terms related to PETs will be added and the existing terms will be updated soon.
OECD Global Repository for Privacy Enhancing Technologies: Glossary
Differential Privacy
These techniques make small changes (add noise) to the raw data to mask the details of individual inputs, while maintaining the explanatory power of the data. The idea is that small changes to individual records can securely de-identify the inputs without having a significant impact on the aggregated results. Noise can be added at the time of data collection (distributed) or at the central location before the data are released (centralised). [OECD 2023]
Homomorphic encryption
Homomorphic encryption computes over encrypted data. The data subjects locks the data (with a key only they have) before passing them on to the data processor. The processor can then perform simple (but increasingly complex) calculations over the encrypted data to extract an encrypted result that can only be unlocked with the data subject’s key. [OECD 2023]
Secure multi-party computation
Secure multi-party computation (SMPC) is a set of tools that enables the participating parties to jointly compute a function over their input data while keeping those input data private. Essentially, it removes the need for a trusted third party to view and manage the data. MPC can aggregate sensitive data without requiring any data contributor to disclose their own data. [OECD 2023]
Synthetic data
An approach to confidentiality where instead of disseminating real data, synthetic data that have been generated from one or more population models are released. [OECD 2023]
Trusted execution environment
A trusted execution environment (TEE) is a dedicated area on a computer processor that is separated and secured from the operating system. It holds sensitive, immutable data and can run secure code within its secure confine. TEE assumes the operating system is corruptible and untrustworthy. Consequently, under TEE, the operating system cannot access information in the secure area of the processor or read the stored secrets. TEEs provide a secure location where data can be stored and used without exposing them to the risks of an untrusted environment. [OECD 2023]