Explainable AI in Automated Assessment How to Make Algorithmic Grading Transparent for Students and Teachers

Abstract

Automated grading systems increasingly score student work without explaining their scoring decisions. This opacity creates confusion for students and weakens trust among classroom teachers. This research examines how explainable artificial intelligence can make algorithmic grading transparent. The study aims to identify practical methods that clarify grading decisions for students and teachers. It explores which explanation formats build meaningful trust within real classroom settings. A qualitative approach guided this research, combining doctrinal legal analysis with document analysis. Legal frameworks, scholarly literature, and classroom findings were reviewed systematically throughout this study. Results show explanation features improved student understanding and strengthened teacher confidence significantly. Visible grading logic also helped teachers detect bias and reduce grading disputes. This research recommends embedding explanation features within every automated grading system used. Institutions should also pair legal transparency requirements with practical classroom implementation support. These findings suggest explain ability remains essential, not optional, for trustworthy automated assessment systems..