The Coming but Uneven Storm

Abstract

This article investigates the state of AI regulations from diverse geopolitical contexts including the European Union, the United States, China, and several African nations, and their implications for learning analytics (LA) and AI research. We used a comparative analysis approach of 11 AI regulatory documents and applied the OECD framework to classify core priorities. The findings showed that the European Union and China have adopted the most comprehensive and strict AI regulations, potentially setting global standards once developed into legal instruments. In contrast, the United States has experienced a significant policy shift: the previously established AI Bill of Rights and Executive Order 14110 were revoked in early 2025 and replaced by a new Executive Order focused on economic competitiveness and deregulation. While the EU, UK, and China prioritize safety and avoiding algorithmic bias, Japan and some African countries emphasize AI systems’ economic and societal potential. Certain data types (e.g., Biometric sensors) might become challenging to use for education due to strict guidelines (particularly within the EU), necessitating careful consideration of data nature and privacy risks before use and when transferring data across jurisdictions. As regulations shift, researchers and practitioners must adapt to changing regulations across international boundaries, the legal benefits and risks of using open-source licenses, and regulatory sandboxes to minimize the risks for learners and developers and ensure ethical and legally compliant AI-driven LA systems.

Ähnliche Arbeiten