Journal №1-2 (2025)
Reconciling Data Minimization with Model Maximization: Regulatory and Ethical Tensions in AI Development
∘ Agnieszka Grzelak ∘

The rapid advancement of large-scale artificial intelligence (AI) systems, particularly large language models (LLMs), has created profound regulatory tensions in the realm of data protection. Central to this discourse is the conflict between the principles of data minimization, as enshrined in the General Data Protection Regulation (GDPR), and the data-intensive logic underpinning AI model development. This article explores some aspects of the legal, practical, and ethical implications of this tension from the perspective of data protection authorities (DPAs), analyzing current enforcement trends, regulatory guidance, and the prospective impact of the EU Artificial Intelligence Act. It argues that DPAs must evolve beyond traditional enforcement roles to become ethical stewards and proactive coordinators of AI governance in Europe to ensure that the fundamental principles are not weakened or ignored in the name of innovation.

See in detail: 

---