Researchers from the University of Phoenix College of Doctoral Studies have published a framework that uses generative AI and predictive analytics to boost student success in online learning environments. The article, authored by Pamayla E. Darbyshire and Carl Beitsayadeh, appears in the International Journal for Educational Media and Technology.
The framework emphasizes a human-centered approach, aiming to balance technological capabilities with the needs of individual learners. It draws from papers originally presented at the 2025 Teaching, Colleges, and Community Worldwide Online Conference (TCC), signaling a growing focus on adaptive tools in higher education.
No specific performance data or pilot results were included in the publication. The research is conceptual at this stage, outlining principles for how AI could identify at-risk students and tailor interventions without replacing human instructors.
The model suggests that predictive analytics might flag engagement drops, but it does not prescribe concrete thresholds or algorithms. Further empirical testing would be needed before institutions could deploy such a system at scale.
Critics may argue that AI-driven monitoring raises privacy concerns and could lead to over-reliance on automated systems. Without clear guardrails, the framework risks reinforcing biases present in the data used to train its models.