Generative Al and Learning Analytics for Personalized Education: Predicting Student Performance, Dropout Risk, Learning Gaps, and Adaptive Learning Needs in Pakistani Universities

Abstract

The rapid integration of Generative Artificial Intelligence (GenAI) into higher education has precipitated a paradigm shift in academic production, assessment practices, and pedagogical engagement worldwide. This study provides a comprehensive analysis of the multifaceted impact of GenAI on academic integrity and assessment practices within Pakistani Higher Education Institutions (HEIs). Employing a systematic synthesis of empirical evidence, policy documents, and theoretical frameworks, the paper examines the dual-vector paradox wherein GenAI simultaneously functions as a pedagogical productivity tool while inducing cognitive atrophy and threatening academic integrity through unmonitored usage. Empirical data from Pakistani university contexts reveal widespread student adoption of tools like ChatGPT, with a statistically significant moderate positive correlation (r = 0.453, p < 0.05) between GenAI utilization and continuous skill enhancement, yet a significant negative relationship with critical thinking (β = -0.32) and academic integrity adherence (β = -0.28). The paper critically evaluates the regulatory framework established by the Higher Education Commission (HEC), including the mandatory AI curriculum directive effective Fall 2026 and institutional governance mechanisms through Departmental Standing Committees. It identifies systemic vulnerabilities, highlighted by a national implementation index of only 0.61 out of 1.00, and proposes the Pakistan AI Education Translation Framework—a six-stage operational pathway addressing access readiness, safety governance, teacher preparedness, curriculum localization, assessment integrity, and evidence-based scaling. The paper concludes that preserving academic integrity in the GenAI era necessitates a fundamental pedagogical transformation from vulnerable summative assessments toward authentic, process-oriented evaluation modalities that prioritize critical thinking, localized problem-solving, and verifiable personal insight, while ensuring equitable access across Pakistan's socio-economically stratified higher education landscape.