Efectividad de la IA generativa en el rendimiento académico en educación superior

Objective: The study examined how generative artificial intelligence helps academic performance in higher education, considering factors related to its use, teaching, and objectives. Methodology: A qualitative study with a bibliographic-documentary approach was conducted. Recent publications (2021–2025) were chosen and selected, and the information was organized using an analytical matrix. Thematic and content analysis was applied to compare approaches, mechanisms, and methods for measuring performance and effectiveness. Results: The reviewed evidence consisted primarily of reviews and non-experimental studies, revealing an expanding field with heterogeneous metrics and limited comparability. The findings showed that effectiveness was most consistently observed in three areas: academic writing support, tutoring and feedback, and personalization. However, its connection to academic performance depends on academic literacy, teacher support, assessment criteria, and the institution’s capabilities. Problems related to academic integrity, bias, equity, and the risks of relying on technology were also identified. These issues affected the credibility and validity of the performance improvements. Conclusions: It was concluded that the objective was met and the hypothesis was partially supported, since the benefits were observed primarily when AI was integrated in a guided and pedagogically intentional manner; it was recommended to strengthen the evidence with longitudinal designs and standardized metrics that incorporate mediating and contextual variables.