Generative Artificial Intelligence in Educational Feedback: Academic Transformation through AI-Enhanced Adaptive Assessment

Abstract

Objective: To analyze the potential of human-supervised generative artificial intelligence in improving the quality, timeliness, and personalization of educational feedback in higher education institutions. Methods: A systematic literature review published between 2020 and 2024 was conducted, along with an institutional implementation project in a computer science program with 4,500 students and 89 instructors, reviewing design documents, teaching workload, and projected impact metrics. Results: Generative AI reduced feedback time by approximately 65%, ensuring consistency, objectivity, and equity. Institutional and pedagogical benefits were identified, along with challenges related to ethics, algorithmic bias, and data privacy. Conclusions: Generative AI represents a strategic opportunity to achieve more inclusive, continuous, and competency-centered assessment processes, provided that permanent human supervision and ethical responsibility are maintained.