IEEE P3417

New IEEE Standard - Active - Draft.
This standard specifies the technical architecture, requirements, and evaluation methods for using differential privacy to protect user personal information in data processing. It defines three deployment models (central, local, and shuffle), establishes technical requirements covering basic capability, data security, algorithm security, and scenario-based security, and provides a three-level protection classification scheme. Test and evaluation methods for verifying conformance are also specified.

This standard specifies the architecture and technical requirements of differential privacy for personal information protection in Artificial Intelligence model training. Differential privacy provides a framework for ensuring the privacy of individuals in datasets by allowing data to be analyzed without revealing sensitive information about any specific individual. The requirements consist of data security, algorithm security and scenario-based system security of differential privacy technology used in model training process. This standard also provides classification methodologies of differential privacy technology.