Privacy-enhancing technology
A privacy-enhancing technology is a technical method or system that protects personal or sensitive information while enabling data processing or other useful activity.
First-Party Data Activation
A technology or method designed to enable useful data processing while minimizing collection, exposure, or identifiability of personal information.
- How it is used
- Enables marketing analysis, personalization, and partner collaboration while reducing exposure of personal data.
- How deep it goes
- Defines several methods and compares their uses and limits at a decision-making level, without implementation instructions.
- Examples
- Uses grouped locations, decentralized model training, and a retailer’s cross-channel clean-room measurement.
- What you can do after
- Evaluate/understand which privacy method suits an analysis or sharing goal and where operational limits require attention.
Read how First-Party Data Activation explains privacy-enhancing technology →
Introduction to Generative AI, Second Edition
A privacy-enhancing technology is a technical method that reduces the collection, exposure, linkability, or identifiability of personal information.
- How it is used
- Positions PETs as ways to reduce privacy risks from sensitive training data and unintended disclosure by generative models.
- How deep it goes
- Defines PETs, compares several methods and their limitations, and explains differential privacy intuitively without implementation steps.
- Examples
- Uses blocklists, masked training data, and extracted numbers from Enron emails to show both applications and limits.
- What you can do after
- Evaluate/understand which privacy risk a PET addresses and why data selection and organizational controls remain necessary.
Read how Introduction to Generative AI, Second Edition explains privacy-enhancing technology →
Related concepts
- differential privacyincludes · Differential privacy is a mathematical privacy guarantee that bounds how much an individual’s data can affect the output of a computation, often by adding calibrated randomness.
- federated learningused by · Federated learning is a machine-learning approach in which a shared model is trained across decentralized data sources without centrally collecting their raw data.
- anonymizationincludes · Anonymization is the alteration or removal of data that reduces the ability to identify individuals or link them to records.
- databaseuses · A database is an organized collection of persistent data managed by software for storage, retrieval, querying, and modification.
- data mininguses · Data mining is the computational or systematic discovery of patterns, relationships, or other useful information in data.
- homomorphic encryptionincludes · Homomorphic encryption is encryption that permits specified computations on ciphertexts whose decrypted results correspond to computations on the underlying plaintexts.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.