Towards an Operational Responsible AI Framework for Learning Analytics in Higher Education

Seven core principles should guide AI use in universities: The framework identifies transparency, fairness, and accountability as key principles that institutions should implement when deploying learning analytics systems to predict student success and personalize education.

Algorithmic bias poses real risks to minority students: The paper highlights that AI systems used in higher education can perpetuate unequal support for minority students through algorithmic biases, making bias detection and mitigation a critical responsibility for universities adopting these technologies.

Existing guidance falls short of practical implementation: While many organizations have published Responsible AI principles, the paper found that existing frameworks lack concrete, actionable steps that universities can actually follow—addressing a significant gap between theory and practice.

A tailored framework is needed for educational contexts: Rather than applying generic tech company frameworks, the authors developed guidance specifically designed for learning analytics in higher education, recognizing that educational institutions have unique needs and constraints compared to other sectors.

Responsible AI frameworks must remain flexible and community-driven: The proposed framework is designed to evolve with input from the higher education community, acknowledging that as AI systems and educational practices change, responsible implementation guidance must adapt accordingly.