The rise of digital technologies and their educational applications increasingly require the development of digital skills among university faculty. This study focuses on examining the level of digital competencies of university faculty and identifying their conditioning factors. To achieve this objective, an ex post facto methodological design with surveys is used. A sample of 216 university teachers from different regions of Ecuador was used. The non-parametric Chi-square test was used to validate the hypothesis of independence of the variables. The results obtained show that university faculty have a mostly intermediate level of digital skills, which is independent of gender, but dependent on the generational cohort. In particular, it is found that younger teachers (millennials) have a more advanced level of digital skills, although this relationship is not very strong. On the other hand, we observe the positive influence of the university’s strategic leadership in terms of technological adoption on the development of teachers’ digital skills. Universities with better technological resources and with training plans focused on the pedagogical application of technology have teachers with a more advanced level of digital skills. All of this leads to the recommendation that education policies should prioritise actions that promote the development of digital competencies among university faculty

The increasing integration of artificial intelligence into educational processes offers new opportunities to address critical issues in higher education, such as student dropout, academic underperformance, and the need for personalized tutoring. This scoping review aims to map the scientific literature on the use of AI techniques to predict academic performance, risk of dropout, and the need for academic advising, with an emphasis on e-learning or technology-mediated environments. The study follows the Joanna Briggs Institute PCC strategy, and the review was reported following the PRISMA-ScR checklist for search reporting. A total of 63 peer-reviewed empirical studies (2019–2025) were included after systematic filtering from the Scopus and Web of Science databases. The findings reveal that supervised machine learning models, such as decision trees, random forests, and neural networks, dominate the field, with an emerging interest in deep learning, transfer learning, and explainable AI. Academic, behavioral, emotional, and contextual variables are integrated into increasingly complex and interpretable models. Most studies focus on undergraduate students in digital and hybrid learning contexts, particularly in regions with high dropout rates. The review highlights the potential of AI to enable early intervention and improve the effectiveness of tutoring systems, while noting limitations such as lack of model generalization and ethical concerns. Recommendations are provided for future research and institutional integration.

En este artículo, se presenta una breve mirada al estado actual de los repositorios digitales en el Ecuador. Un repositorio digital es un depósito de documentos digitales, cuyo propósito es gestionar, organizar, almacenar, preservar y difundir en acceso abierto la producción administrativa, científica y/o académica resultante de las actividades de una institución u organización. Desde este contexto el objetivo de este estudio es explorar e identificar en qué medida ha evolucionado los repositorios digitales en el Ecuador en los últimos años, así como valorar la calidad, estandarización de los tipos de materiales, software, licencias y su visibilidad e impacto nacional e internacional.

This research addresses the need to manage student diversity and academic trajectories in higher education through personalized learning, overcoming the risks of algorithmic reductionism and decontextualized automation that depersonalize the educational process. Through an applied instructional design approach and a systematic review of the technical literature, this study proposes and substantiates the architecture of an intelligent hybrid model that coherently articulates and integrates massive analytical data sources from learning management systems (LMS), educational ontologies, explicit pedagogical rules, and machine learning algorithms. The conceptual results demonstrate that the development and implementation of adaptive environments and intelligent tutors make it possible to model dynamic student profiles and structure flexible learning paths that continuously adjust according to students’ achievement levels. In this context, we successfully designed highly efficient early-warning systems based on data mining that promptly identify signs of academic risk and vulnerability, thereby significantly reducing dropout rates and enhancing student retention. It is concluded that the true effectiveness of intelligent hybrid models is not limited to their metric or predictive accuracy, but rather lies substantially in active faculty mediation, algorithmic transparency, decision explainability, and ethical and institutional governance of information. Technological personalization gains viability and social value only when it is established as a tool aimed at strengthening equity, inclusion, and informed human decision-making within the university ecosystem.