AI, BIAS, AND EQUITY IN EDUCATION: ADDRESSING CHALLENGES IN ALGORITHMIC DECISION MAKING

Artificial Intelligence (AI) has become increasingly integrated into educational systems through predictive analytics, automated assessments, learning management platforms, admissions processes, and student support services. Growing reliance on algorithmic decision-making technologies has created opportunities to improve efficiency, consistency, and scalability in educational administration and instruction. Concerns regarding algorithmic bias, transparency, accountability, and educational equity have simultaneously emerged as critical challenges, particularly when AI systems are trained on historical data that may reflect existing social and institutional inequalities. This study aims to examine the relationships among AI implementation, algorithmic bias, transparency, accountability, trust, and educational equity in educational decision-making contexts. A mixed-methods convergent design was employed. Quantitative data were collected from 480 participants, including students, teachers, administrators, policymakers, and educational technology specialists. Qualitative interviews were conducted to explore stakeholder experiences and perceptions regarding algorithmic fairness and governance. Structural Equation Modeling and thematic analysis were used to analyze the data. Findings revealed that transparency significantly enhanced trust in AI systems, while accountability positively influenced perceptions of educational equity. Perceived algorithmic bias demonstrated a strong negative effect on fairness and trust. Qualitative evidence further indicated that inadequate oversight and biased datasets may contribute to unequal educational outcomes. The study concludes that equitable AI implementation requires transparent governance, accountability mechanisms, continuous bias auditing, and meaningful human oversight to ensure that algorithmic decision-making supports educational justice rather than reinforcing existing inequalities.