Data privacy engineering is the development process that analyzes and evaluates new or existing systems and designs them in such a way that all data privacy requirements are met. The goal here is to implement the desired functionality for the customer base in such a way that the data privacy rights of the data subjects are preserved. Common concepts, such as Privacy-by-Design (PbD), can be adapted to the new challenges of AI applications. However, special approaches are required in order to optimally protect the data of those affected.
As data volumes grow, so do the requirements for pseudonymization and anonymization procedures to guarantee the privacy of all data subjects. Privacy Impact Assessments (PIA) provide a method to reliably identify privacy risks and subsequently avoid them – or at least minimize their risk.
The Fraunhofer Big Data and AI Alliance offers suitable methods for integrating functionality and data protection on an equal footing in system design.