Challenges in aligning AI agents with GDPR transparency and accountability requirements

Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).

The exponential growth of Artificial Intelligence (AI) agents have opened up industries to new frontiers and also brought serious concerns for how they will comply with the General Data Protection Regulation (GDPR). This paper discusses the fundamental challenges with bringing AI systems up to GDPR’s transparency and accountability standards, particularly around automated decision making, transparency of data processing, consenting, and reporting mechanisms. Key tensions stem from the “black-box” nature of AI model transparency, the dispersion of responsibility among actors, and the technical difficulties in applying GDPR-required rules such as explainability and informed consent. To address these issues, this paper adopt a socio-technical systems framework to analyze the conflicts between AI agent operational features and GDPR requirements. This approach allows for a holistic view of the interconnected legal, technical, and organizational challenges. By integrating this framework with specific case studies from healthcare, finance, and online gaming, the paper map out the missing links and propose solutions to bridge the innovation-to-regulation gap. The recommendations include the use of Explainable AI (XAI), robust accountability mechanisms, improved consent frameworks, and a privacy-by-design approach. The paper highlights the urgent need for a joint innovation-to-regulations collaboration to ensure the ethical and legal deployment of AI.