Bridging equity and privacy: ethical challenges in AI-powered personalized learning platforms for automated grading and tutoring

Abstract

This study systematically deconstructs core ethical dilemmas such as data privacy violations, algorithmic decision bias, and unbalanced educational resource allocation, based on the deep application of Artificial Intelligence (AI) and big data technologies in the education sector. Through an interdisciplinary analytical framework, it analyzes the root causes of problems from three dimensions: technical architecture defects, social governance failures, and policy/regulatory lag. Combined with typical cases at home and abroad, it reveals the systematic impact of ethical risks on the educational ecosystem. The research proposes innovative strategies including federated learning privacy protection, algorithmic ethics filing systems, and collaborative governance by multiple stakeholders, providing theoretical support and practical paths for constructing an educational technology development system that balances technological innovation and ethical norms. The study finds that the ethical challenges in educational digital transformation essentially reflect the conflict between technical instrumental rationality and educational value rationality, requiring sustainable development through dual adjustments of institutional design and technological innovation.