• algorithmic accountability
• article 22
• artificial intelligence
• data protection
• GDPR
• human-in-the-loop
• judicial decision-making
• transparency

The integration of artificial intelligence (AI) into judicial decision-making processes promises significant gains in efficiency, consistency, and resource management. However, the deployment of AI in courts raises complex legal and ethical challenges, particularly regarding compliance with the European Union’s General Data Protection Regulation (GDPR). Article 22 of the GDPR, which governs automated decision-making and profiling, imposes safeguards such as transparency, the right to explanation, and meaningful human intervention. The judicial context presents unique difficulties in operationalizing these safeguards without compromising judicial independence or procedural efficiency.
This dissertation investigates how AI can be implemented in judicial decision-making while remaining compliant with the GDPR and protecting the fundamental rights of data subjects. Adopting a doctrinal legal analysis combined with a comparative study of EU Member State practices, it examines relevant GDPR provisions, judicial applications of AI, and potential conflicts between data protection obligations and court operations. Case studies—focusing on jurisdictions with varying levels of AI adoption—provide empirical insight into best practices and shortcomings.
The research hypothesizes that GDPR-compliant AI integration in judicial contexts is achievable through a hybrid “human-in-the-loop” model. This framework combines proportional algorithmic transparency, independent audits of AI systems, and final human deliberation documented with explicit reasoning. The findings aim to offer practical guidance for courts, policymakers, and AI developers, contributing both to the scholarly debate on AI governance and to the responsible modernization of judicial systems.