AI in Educational Assessment: Human Oversight, Ethical Challenges, and Management

The integration of Artificial Intelligence (AI) and Large Language Models (LLMs) into educational assessment processes offers significant potential for automated grading, personalized feedback, and student data analytics. However, this integration raises critical ethical and managerial challenges, including algorithmic bias, privacy concerns, transparency, fairness, and the diminishing role of human judgment in decision-making. Through a review of recent literature through ۲۰۲۵, this study argues that maintaining human oversight within the decision-making loop, developing robust managerial and governance frameworks, and training educators and administrators in the ethical use of AI are indispensable. The implications of this research underscore the necessity of reinforcing the human role in assessment, proactively managing biases, safeguarding data privacy, and designing ethically aligned tools. LLMs should be used as complementary tools to enhance learning, not as substitutes for human expertise and judgment.

کلیدواژه ها:

Artificial Intelligence (AI) ، Educational Assessment ، Ethical Challenges ، Decision-Making