Pseudonymity and AI: The New Frontier of Responsible Data

The future of AI innovation relies on data access, but this ambition is challenged by strict privacy laws. This panel provides a deep dive into anonymization vs. pseudonymization under GDPR, clarifying definitions, obligations and the ECJ's latest interpretation that stresses a relative, contextual approach to identifiability. We will explore the critical implications of the ECJ decision for AI development, examining how the ruling shapes the responsible use of datasets in training and deploying AI systems. Discover the role of Privacy Enhancing Technologies and how they transform pseudonymization into a tool for responsible innovation, enabling privacy-preserving techniques like federated learning, synthetic data and secure computation. We demonstrate how to translate complex legal requirements into actionable engineering solutions for building robust anonymization solutions. Finally, we provide a regulatory and ethical outlook, reviewing guidance from data protection authorities and discussing the links with the EU AI Act and the importance of global standards. Join us to navigate the new regulatory landscape and bridge the gap between legal intent and technical reality for responsible, privacy-preserving AI.

What you will learn: 

• Understand the ECJ's shift to a relative test for pseudonymity (anonymous for the recipient) and its direct impact on AI data use. 
• How to leverage PETs to build robust, legally compliant anonymization and pseudonymization solutions. 
• Clarity on the distinction between anonymization and pseudonymization under GDPR and its necessary alignment with the EU AI Act.