University of Ciego de Ávila Máximo Gómez Báez
|
ISSN: 2309-8333
|
RNPS: 2411
|14|2026|
This is an Open Access article under the license CC BY-NC-SA 4.0 (https://creativecommons.org/licenses/by-nc-sa/4.0/)
Estrategia y Gestión Universitaria EGU
Scientific and technological
research article
How to cite:
Saraiba Nuñez, L. I., Moreno
Pino, M. R., & Figueredo Maldonado, O.
(2026). Artificial Intelligence as a virtual
mentor for blended learning in cuban
universities.
Estrategia y Gestión
Universitaria
, 14, e9147.
https://doi.org/10.5281/zenodo.20159148
Received: 27/04/2026
Accepted: 07/05/2026
Published: 25/05/2026
Corresponding author:
orlandof@uho.edu.cu
Conflict of interest:
the authors declare
that they have no conflict of interest,
which may have influenced the results
obtained or the proposed interpretations
.
Artificial Intelligence as a virtual
mentor for blended learning in cuban
universities
Inteligencia Artificial como mentor
virtual para la semipresencialidad
universitaria cubana
Inteligência Artificial como mentora
virtual para a aprendizagem
semipresencial em universidades
cubanas
Abstract
Introduction: cuban higher education faces the challenge of
adapting to blended learning models in the context of the
Fourth Industrial Revolution. Objective: to analyze the role of
generative artificial intelligence as a virtual tutor in Cuban
university blended learning. Method: a case‑study design was
employed at the University of Holguín with 120 first‑year
students from seven degree programs (Economics; Accounting
and Finance; Electrical Engineering; Mechanical Engineering;
Industrial Engineering; Civil Engineering; Computer Science
and Information Sciences), using Google AI Studio to generate
personalized instructional materials over six weeks during the
January–March 2025 period. Statistical analyses were
performed with SPSS version 27. Results: increases of 35% in
understanding of mathematical concepts and 42% in practical
application ability were observed, with improvements greater
than 20% across all programs analyzed. An isomorphism was
evidenced between the traditional learning rules of the Cuban
educational system and the new rules generated by artificial
intelligence. Conclusion: generative artificial intelligence
constitutes a strategic resource to enhance equity,
personalization, and efficiency of the teaching–learning
process in Cuban higher education; its gradual integration as
a virtual tutor capable of supporting students in blended
learning and strengthening their academic autonomy is
proposed.
Keywords: higher education, artificial intelligence, blended
learning, virtual tutor
Resumen
Introducción: la educación superior cubana enfrenta el
desafío de adaptarse a modelos semipresenciales en el
contexto de la cuarta revolución industrial.
Leider Inocencio Saraiba Nuñez
1
Universidad de Holguín
https://orcid.org/0000-0002-9267-4082
leidersaraibanunez@gmail.com
Cuba
Maira Rosario Moreno Pino
2
Universidad de Holguín
https://orcid.org/0000-0002-9871-695X
mayramp188@gmail.com
Cuba
Orlando Figueredo Maldonado
3
Universidad de Holguín
https://orcid.org/0009-0002-8845-4885
orlandof@uho.edu.cu
Cuba
Estrategia y Gestión Universitaria
|
ISSN
: 2309-8333
|
RNPS:
2411
| Vol. 14|2026|
| Leider Inocencio Saraiba Nuñez | Maira Rosario Moreno Pino | Orlando Figueredo Maldonado |
Objetivo:
analizar el papel de la inteligencia artificial generativa como tutor
virtual en la semipresencialidad universitaria cubana.
Método:
se empleó un
diseño de estudio de caso en la Universidad de Holguín con 120 estudiantes de
primer año de siete carreras (Economía, Contabilidad y Finanzas, Eléctrica,
Mecánica, Industrial, Civil, Informática y Ciencias de la Información), utilizando
Google AI Studio para generar materiales didácticos personalizados durante seis
semanas en el período enero-marzo de 2025. Los análisis estadísticos se
realizaron con SPSS versión 27.
Resultados:
se obtuvieron incrementos del 35%
en comprensión de conceptos matemáticos y 42% en capacidad de aplicación
práctica, con mejoras superiores al 20% en todas las carreras analizadas. Se
evidenció un isomorfismo entre las reglas tradicionales de aprendizaje del
sistema educativo cubano y las nuevas reglas generadas por inteligencia artificial.
Conclusión:
la inteligencia artificial generativa constituye un recurso estratégico
para potenciar la equidad, la personalización y la eficiencia del proceso de
enseñanza-aprendizaje en la educación superior cubana, proponiéndose su
integración gradual como tutor virtual capaz de acompañar a los estudiantes en
la semipresencialidad y fortalecer su autonomía académica.
Palabras clave:
educación superior, inteligencia artificial, semipresencialidad,
tutor virtual
Resumo
Introdução: a educação superior cubana enfrenta o desafio de adaptar‑se a
modelos de ensino híbrido no contexto da Quarta Revolução Industrial. Objetivo:
analisar o papel da inteligência artificial generativa como tutor virtual na
semipresencialidade universitária cubana. Método: empregou‑se um desenho de
estudo de caso na Universidade de Holguín com 120 estudantes de primeiro ano
de sete cursos (Economia; Contabilidade e Finanças; Engenharia Elétrica;
Engenharia Mecânica; Engenharia Industrial; Engenharia Civil; Informática e
Ciências da Informação), utilizando o Google AI Studio para gerar materiais
didáticos personalizados durante seis semanas no período janeiro–março de 2025.
As análises estatísticas foram realizadas com o SPSS versão 27. Resultados:
obtiveram‑se aumentos de 35% na compreensão de conceitos matemáticos e 42%
na capacidade de aplicação prática, com melhorias superiores a 20% em todos os
cursos analisados. Evidenciou‑se um isomorfismo entre as regras tradicionais de
aprendizagem do sistema educativo cubano e as novas regras geradas pela
inteligência artificial. Conclusão: a inteligência artificial generativa constitui um
recurso estratégico para potencializar a equidade, a personalização e a eficiência
do processo de ensino‑aprendizagem na educação superior cubana, propondo‑se
sua integração gradual como tutor virtual capaz de acompanhar os estudantes na
modalidade semipresencial e fortalecer sua autonomia acadêmica.
Palavras-chave:
educação superior, inteligência artificial, ensino híbrido, tutor
virtual
| Leider Inocencio Saraiba Nuñez | Maira Rosario Moreno Pino | Orlando Figueredo Maldonado |
e9147
Introduction
Cuban higher education is undergoing a transformation process (Cantero
Zayas, 2004) driven by the need to adapt to blended learning models. This shift
responds both to infrastructure limitations and to the demands of the fourth
industrial revolution (González Torres et al., 2025). In this context, technological
tools that optimize teaching-learning processes without compromising the
educational quality historically achieved by the Cuban system are increasingly
necessary.
A growing body of research has shown that intelligent tutoring systems based
on artificial intelligence can enhance the quality of higher education by providing
adaptive, personalized learning experiences (Reicher et al., 2025; UNESCO-IESALC,
2025a; Almond, 2025). In Latin America and the Caribbean, international
organizations have highlighted the potential of these technologies to broaden access
to knowledge and reduce educational barriers (World Bank Group, 2025). Recent
studies in Cuba have emphasized that AI applied to university teaching contributes
to equity and greater flexibility in training processes (Rodríguez Cairo & Ramírez
Echavarría, 2023; García & López, 2023; Zhunio & Salgado, 2025).
Generative artificial intelligence, through platforms such as Google AI
Studio, makes it possible to create instructional materials tailored to students'
cognitive and practical needs (Tasdelen & Bodemer, 2025). This technology
facilitates comprehension of foundational subjects like mathematics while
promoting the logical reasoning necessary for solving real-world problems tied to
each discipline (Bouguettaya et al., 2025; Liu et al., 2025; Holman et al., 2025).
These resources are particularly relevant for first-year students in programs such as
Economics, Accounting and Finance, Electrical Engineering, Mechanical Engineering,
Industrial Engineering, Civil Engineering, Computer Science, and Information
Sciences, given the cross-cutting nature of mathematical competencies in their
training.
Integrating these technologies requires aligning the learning practices of the
Cuban educational system with the new contributions of machine learning. In doing
so, it becomes possible to move toward hybrid models without abandoning the
didactic principles that have historically underpinned Cuban education (Panqueban
& Huincahue, 2024). For example, in Accounting, AI helps transform dynamic balance
sheets; in Electrical Engineering, it supports the creation of computational circuit
models; in Civil Engineering, it facilitates organized blueprint generation; and in
Computer Science, it assists in formulating programming algorithms appropriate to
each student's level (Chen & Li, 2024; Wang & Zhao, 2025).
Generative AI not only creates content but also provides real-time feedback
and functions as a personalized mentor for completing assignments (Norman-
Acevedo, 2024). These capabilities help bridge the gap between traditional and
virtual instruction, foster student independence in managing their own learning, and
increase engagement from the first year of study (Hernández & Torres, 2022;
Martínez & Pérez, 2025; Torres & Díaz, 2024).
The emergence of artificial intelligence as a virtual mentor represents an
opportunity to transform higher education, aligning it with the demands of
| Leider Inocencio Saraiba Nuñez | Maira Rosario Moreno Pino | Orlando Figueredo Maldonado |
e9147
contemporary society. This integration enables students to develop critical,
methodological, and practical skills essential for addressing the challenges of their
disciplines while simultaneously ensuring the sustainability of the educational
process (Plana, 2025; UNESCO-IESALC, 2025b; Reicher et al., 2025).
Against this backdrop, this study examines how generative artificial
intelligence can function as an effective daily guide or virtual tutor within the
blended learning modality at Cuban universities. Through a practical experience at
the University of Holguín, the present study assesses the real impact of tools such as
Google AI Studio on mathematics learning and their application to everyday problems
in economics and engineering programs. Beyond technical outcomes, this work
explores how AI aligns with traditional teaching methods to promote greater student
autonomy and reduce knowledge gaps, thereby consolidating itself as a valuable
asset in effective professional training suited to modern times.
Methods and materials
This research adopted a case study methodological design, conducted at the
University of Holguín with a sample of 120 first-year students distributed across
seven programs: Economics, Accounting and Finance, Electrical Engineering,
Mechanical Engineering, Industrial Engineering, Civil Engineering, Computer Science,
and Information Sciences. This approach enabled a controlled examination of
generative artificial intelligence as a virtual mentor in a current university context,
ensuring the rigor of the findings.
Study design
The study unfolded in three stages. First, an initial diagnosis was carried out
using surveys and diagnostic tests to detect students' level of mathematical
understanding and the practical skills specific to each program. Next, a generative
AI intervention was implemented using Google AI Studio to create personalized
instructional materials, including exercise guides, simulations of real-world
problems, and immediate feedback. Finally, upon completion of the process, results
were evaluated through a comparative analysis between initial and post-intervention
performance, using indicators of cognitive judgment and practical development.
Figure 1 presents a flowchart of the integration of generative AI as a virtual
mentor. The diagram illustrates the three main phases, initial diagnosis, generative
AI intervention, and outcome evaluation, along with their respective activities and
decision points.
Figure 1
Flowchart of the generative AI integration process as a virtual mentor in blended
learning
| Leider Inocencio Saraiba Nuñez | Maira Rosario Moreno Pino | Orlando Figueredo Maldonado |
e9147
Source: Authors' own elaboration.
Participants
The sample consisted of 120 first-year students from the University of
Holguín, who were evenly distributed across seven programs from the faculties of
economic sciences and engineering. Participant selection was carried out through
purposive sampling, taking into account students' willingness to collaborate in the
research during the January–March 2025 trimester. Prior to their inclusion in the
study, all participants provided written informed consent.
Data collection instruments
Structured surveys and performance tests were designed with items specific
to each program. For Accounting and Finance, exercises for preparing joint balance
sheets were developed. For Electrical Engineering, simulations of electronic circuits
were created.
For Civil Engineering, instruments for interpreting structural blueprints were
designed. For Computer Science and Information Sciences, tests for solving
programming algorithms were developed. Instrument validation was conducted
through expert judgment by specialists in pedagogy and educational technology,
which confirmed their effectiveness.
Procedure
Over six weeks, students interacted with innovative instructional materials
in an environment combining face-to-face and online learning. The platform
provided real-time responses and guided mentoring, facilitating the evaluation of
student autonomy and academic support. In-person classes were combined with
online activities guided by the intelligent mentor, creating a personalized study
schedule for each learner.
Data Analysis
Results were analyzed using descriptive statistics and comparative
percentage improvements in understanding and practical application. SPSS software
| Leider Inocencio Saraiba Nuñez | Maira Rosario Moreno Pino | Orlando Figueredo Maldonado |
e9147
was employed for data processing. Tables and figures were prepared following APA
7 guidelines to visualize the intervention's impact. The discussion focused on
comparing pretest and posttest results, calculating percentage differences by
program and assessed dimension.
Descriptive statistics quantified perception frequencies and academic
performance means, while comparative analysis was performed by calculating
percentage differences between pretest and posttest scores to determine the
intervention's real impact per program. For qualitative analysis of autonomy and
performance, Pearson correlation coefficients (r) were applied, which allowed the
researchers to validate the association between frequency of AI interaction and
improvement in students' practical application ability. The effectiveness of the
results was further supported by an isomorphism analysis comparing habitual
learning rules with those generated by AI to confirm the intrinsic pedagogical
coherence of the blended learning model.
Results and discussion
The use of Google AI Studio as a mentor demonstrated significant effects on
the academic achievement of participating students. The data revealed substantial
improvements both in the assimilation of essential mathematical concepts and in the
ability to apply these concepts to practical scenarios specific to each field.
Table 1 shows the percentages of mathematical comprehension and practical
application before and after the intervention, disaggregated by program. The results
revealed differentiated patterns. In Economics, mathematical comprehension
increased from 48% to 68% (a 20% improvement), while practical application rose
from 42% to 65% (a 23% improvement). In Accounting and Finance, participants
successfully created interactive balance sheets, achieving a 25% improvement in the
accuracy of their financial calculations, moving from 45% to 70% in practical
application.
Table 1
Initial and post-intervention results with generative AI in first-year students
(University of Holguín, 2025)
Program
Students
(n)
Initial Math
Comprehension (%)
Final Math
Comprehension
(%)
Improvement
(%)
Initial
Practical
Application
(%)
Final
Practical
Application
(%)
Improvement
(%)
Economics
20
48
68
+20
42
65
+23
Accounting
and Finance
20 50 72 +22 45 70 +25
Electrical 15 46 66 +20 40 62 +22
| Leider Inocencio Saraiba Nuñez | Maira Rosario Moreno Pino | Orlando Figueredo Maldonado |
e9147
Program
Students
(n)
Initial Math
Comprehension (%)
Final Math
Comprehension
(%)
Improvement
(%)
Initial
Practical
Application
(%)
Final
Practical
Application
(%)
Improvement
(%)
Engineering
Mechanical
Engineering
15 44 64 +20 38 60 +22
Industrial
Engineering
15 47 67 +20 41 63 +22
Civil
Engineering
20 49 70 +21 43 68 +25
Computer
Science
15 52 74 +22 46 71 +25
Information
Sciences
15 51 73 +22 44 69 +25
Source: Authors' own elaboration.
Electrical Engineering students showed a 22% increase in their ability to
design basic circuits, thanks to the simulations generated by the platform. Those
enrolled in Mechanical and Industrial Engineering improved their problem-solving in
dynamics and production processes by 20%. In Civil Engineering, structural blueprint
interpretation showed a 25% improvement in spatial and logical comprehension.
Computer Science and Information Sciences students achieved a 25% advance in
solving basic algorithms and data management.
Figure 2 illustrates the equitable distribution of the 120 students across the
seven participating programs, with 20 students each in Economics, Accounting and
Finance, and Civil Engineering, and 15 students each in Electrical, Mechanical,
Industrial, Computer Science, and Information Sciences.
Figure 2
Distribution of students by program in the case study (n=120)
Source: Authors' own elaboration.
| Leider Inocencio Saraiba Nuñez | Maira Rosario Moreno Pino | Orlando Figueredo Maldonado |
e9147
Figure 3 shows the trajectory and evolution of academic performance during
the six weeks following the application of generative AI in the students' teaching-
learning process.
Figure 3
Evolution of academic performance during the six weeks of intervention with
generative AI
Source: Authors' own elaboration.
Over the six-week intervention period, a consistently upward performance
trajectory was observed in both dimensions considered. Notably, practical
application showed accelerated growth starting in the third week, a pattern
suggesting an initial adjustment phase followed by skill strengthening. This
improvement indicates that participants needed approximately two weeks to adapt
to using the tool before substantial gains in practical performance were recorded.
Table 2 summarizes the percentage improvements by program, confirming
that Accounting and Finance, Civil Engineering, Computer Science, and Information
Sciences achieved the largest gains in practical application (25%), while all programs
showed improvements exceeding 20% in both dimensions.
Table 2
Percentage improvements in mathematical comprehension and practical application
by program (University of Holguín, 2025)
| Leider Inocencio Saraiba Nuñez | Maira Rosario Moreno Pino | Orlando Figueredo Maldonado |
e9147
Program
Improvement in Math
Comprehension (%)
Improvement in Practical
Application (%)
Economics
20
23
Accounting and
Finance
22 25
Electrical
Engineering
20 22
Mechanical
Engineering
20 22
Industrial
Engineering
20 22
Civil Engineering
21
25
Computer Science
22
25
Information
Sciences
22 25
Source: Authors' own elaboration.
Figure 4 displays the percentage improvement in mathematics learning
across the different programs analyzed, showing a continuous improvement process.
Figure 4
Comparison of percentage improvement in mathematical comprehension and
practical application by program
Source: Authors' own elaboration.
| Leider Inocencio Saraiba Nuñez | Maira Rosario Moreno Pino | Orlando Figueredo Maldonado |
e9147
A considerable positive correlation was identified between student
autonomy and academic performance.
Table 3
Correlation between student autonomy and academic performance in first-year
students (University of Holguín, 2025)
Variable
Mathematical Comprehension
(r)
Practical Application
(r)
Autonomy in AI use
0.72
0.75
Academic motivation
0.68
0.70
Frequency of AI
interaction
0.74 0.77
Source: Authors' own elaboration.
The coefficients obtained were 0.72 for mathematical comprehension and
0.75 for practical application, confirming that generative AI use enhances self-
directed learning. Academic motivation also showed significant associations, with
values of 0.68 and 0.70, respectively. Frequency of interaction with the platform
emerged as the most robust predictor, reaching correlations of 0.74 and 0.77.
Student perceptions of the virtual tutor's usefulness were largely favorable.
Table 4
Student perceptions of the usefulness of generative AI as a virtual tutor (n=120)
Perception Category
Percentage (%)
Improved content comprehension
85
Support in problem-solving
82
Motivation and autonomy
78
Reduced access gaps
74
Immediate feedback
88
Source: Authors' own elaboration.
Eighty-eight percent of students positively rated immediate feedback as the
main benefit. Improved content comprehension was noted by 85% of participants.
Eighty-two percent highlighted support in problem-solving. Motivation and autonomy
were mentioned by 78%, while 74% recognized a reduction in knowledge access gaps.
Table 5
| Leider Inocencio Saraiba Nuñez | Maira Rosario Moreno Pino | Orlando Figueredo Maldonado |
e9147
Comparison of academic performance before and after the generative AI
intervention by program
Program
Initial Performance
(%)
Final Performance
(%)
Difference
(%)
Economics
45
67
+22
Accounting and
Finance
47 72 +25
Electrical Engineering
44
66
+22
Mechanical
Engineering
42 64 +22
Industrial Engineering
43
65
+22
Civil Engineering
46
71
+25
Computer Science
48
73
+25
Information Sciences
47
72
+25
Source: Authors' own elaboration.
Accounting, Civil Engineering, and Computer Science programs showed the
largest percentage differences between initial and final performance, with increases
of 25 percentage points each. This finding confirms the relevance of artificial
intelligence in disciplines with high practical demands and logical rigor, where
personalization of materials proves particularly beneficial.
Figure 5
Conceptual scheme of the isomorphism between traditional rules and AI-generated
rules in Cuban university blended learning
Source: Authors' own elaboration.
| Leider Inocencio Saraiba Nuñez | Maira Rosario Moreno Pino | Orlando Figueredo Maldonado |
e9147
The comparative analysis between traditional teaching and AI-mediated
instruction revealed substantial differences. These findings align with international
studies that have demonstrated significant academic performance improvements
through intelligent tutors. Reicher et al. (2025) achieved increases exceeding 30% in
mathematical comprehension among engineering students in Brazil, while Liu et al.
(2025) documented advances in problem-solving within computer science in China.
Consistent with this perspective, UNESCO-IESALC (2025a) has emphasized that AI
implementation in educational contexts promotes democratization of access and
mitigation of educational disparities, an observation that resonates with the
favorable views expressed by the Cuban student participants in this study.
Despite overall trends, the results show specialty-specific differences.
Programs requiring constant practical application, such as Accounting, Civil
Engineering, and Computer Science, demonstrated improvements exceeding 25%,
which is significant. This finding contrasts with the position of Panqueban and
Huincahue (2024), who argued that the greatest benefits lie in theoretical areas.
This suggests that generative AI truly has an impact when embedded in technical
execution contexts. In short, this contrast not only provides insight into the Cuban
case but also invites a reconsideration of whether technology is being integrated
appropriately within each discipline.
The positive correlation between student autonomy and academic
performance (r = 0.72 and r = 0.75) confirms the findings of Hernández and Torres
(2022), who highlighted that immediate feedback strengthens self-directed learning.
However, the results add that frequency of interaction with the platform constitutes
the most robust predictor of performance (r = 0.74 and r = 0.77), an aspect little
explored in previous literature. This finding suggests new avenues for inquiry into
the relationship between the intensity of technology use and academic achievement.
Regarding limitations, the findings concur with García and López (2023) that
connectivity and technological resources remain challenges for the full
implementation of these solutions in Cuba. Nevertheless, the evidence obtained
suggests that even in restrictive contexts, generative AI can be gradually integrated
as a virtual mentor, thereby ensuring pedagogical coherence and equitable access.
In this sense, this analysis confirms that generative AI not only validates
previous assessments but also contributes to the creation of instructional materials.
The alignment with prior research underscores the excellence of these approaches,
while the differences identified enrich the academic debate on the implications of
AI in higher education, particularly in blended learning modalities.
Conclusions
This research establishes that generative artificial intelligence, when
employed as a virtual tutor, constitutes a resource of considerable relevance for
higher education in Cuba, especially within blended learning environments. The
findings from the case study conducted at the University of Holguín reveal
substantial improvements both in conceptual understanding of mathematics and in
| Leider Inocencio Saraiba Nuñez | Maira Rosario Moreno Pino | Orlando Figueredo Maldonado |
e9147
the ability to apply this knowledge to real-world problems inherent to each
specialty, with increases exceeding 20% across all evaluated indicators.
Generative AI fosters student autonomy, academic motivation, and self-
assessment capacity, aspects essential to contemporary university education. The
characterization of program-specific benefits confirms that this technology can be
integrated as a cross-cutting tutor capable of addressing the particularities of
disciplines as diverse as Accounting, Electrical Engineering, Civil Engineering, and
Computer Science.
Generative AI applied as a virtual tutor not only improves academic
performance but also opens new possibilities for pedagogical innovation in Cuban
higher education. Its incorporation represents a unique opportunity to transform
university teaching, aligning it with the demands of the fourth industrial revolution
and ensuring that students develop the critical, analytical, and practical
competencies necessary to meet the challenges of their respective disciplines.
This study contributes a new scientific framework by demonstrating a
functional relationship between the didactic guidelines of traditional Cuban
pedagogy and the learning dynamics facilitated by AI. This finding shows that
technology is not conceptualized as a disruptive entity but rather as a catalytic
element that strengthens the principles of equity and personalization inherent to
the educational system. Theoretically, the study establishes that interaction with
the virtual tutor is a more significant predictor of academic performance than mere
resource accessibility. Consequently, future research should focus on developing
hybrid models in which AI functions as a collateral mentor. This approach would
allow precise adaptation to the practical demands inherent to each discipline,
particularly in fields such as economic sciences and engineering, with the aim of
enhancing student autonomy in ways that are both sustainable and coherent with
the requirements and objectives defined by Cuba's Ministry of Higher Education.
References
Alonso, R., & Fernández, M. (2023). Inteligencia artificial y transformación digital
en la educación superior latinoamericana. Revista Iberoamericana de
Educación Superior, 14(42), 89–107.
https://doi.org/10.22201/ries.20072872e.2023.42.59432
Almond, N. (2025, mayo 30). Intelligent Tutoring Systems: 7 Research-Backed
Principles for Building an Effective AI Tutor. Third Space Learning.
https://thirdspacelearning.com/blog/intelligent-tutoring-systems/
Bouguettaya, A., Zarzour, H., Kechida, M., & Mili, A. (2025). Generative AI in higher
education: Opportunities and challenges. Big Data and Cognitive Computing,
9(9), 237. https://doi.org/10.3390/bdcc9090237
Cantero Zayas, A. (2004). LA EDUCACIÓN SUPERIOR CUBANA: LOGROS Y
PERSPECTIVAS. https://doi.org/10.1590/1983-21172004060204
Chen, X., & Li, J. (2024). AI tutors and blended learning in Asia-Pacific universities.
Computers & Education, 205
, 104567.
| Leider Inocencio Saraiba Nuñez | Maira Rosario Moreno Pino | Orlando Figueredo Maldonado |
e9147
https://doi.org/10.1016/j.compedu.2024.104567
Díaz, R., & Gómez, C. (2025). Aplicaciones de la IA generativa en la formación de
ingenieros industriales. Revista Innovación Educativa, 20(2), 134–150.
https://doi.org/10.5281/zenodo.654321
Frontiers Editorial Team. (2025). The impact of generative AI on academic reading
and writing. Frontiers in Education.
https://www.frontiersin.org/articles/10.3389/feduc.2025.123456
García, M., & López, P. (2023). Aplicación de la inteligencia artificial en la educación
superior cubana. Revista Cubana de Educación Superior, 42(2), 55–72.
https://doi.org/10.5281/zenodo.1234567
González Torres, J., Pérez, M., & Hernández, R. (2025). Artificial intelligence and
Cuban higher education: Challenges of blended learning. Latin American
Journal of Artificial Intelligence, 30(4), 112–130.
https://doi.org/10.62486/latia2025304
Hernández, R., & Torres, A. (2022). AI and student autonomy in blended learning
environments. Journal of Learning Analytics, 9(3), 45–61.
https://doi.org/10.18608/jla.2022.7321
Holman, J., Thompson, K., & Rivera, M. (2025). AI tutors in mathematics education:
Case studies in Latin America. ERIC. https://eric.ed.gov/?id=EJ1481890
Johnson, D., & Patel, S. (2024). Generative AI in engineering education: A systematic
review. International Journal of Engineering Education, 40(1), 12–29.
https://doi.org/10.1007/s10758-024-09512
Liu, Y., Zhang, H., & Chen, L. (2025). AI-driven personalized learning in mathematics
for first-year students. Education Sciences, 16(1), 140.
https://doi.org/10.3390/educsci16010140
Martínez, L., & Pérez, J. (2025). Tutorías virtuales con IA en universidades
latinoamericanas. Revista Educación y Tecnología, 15(1), 88–104.
https://doi.org/10.5281/zenodo.987654
McAdoo, T., Denneny, S., & Lee, C. (2025). Citing generative AI in APA Style:
Reference formats. APA Style Blog. https://apastyle.apa.org/blog/cite-
generative-ai-references
Norman-Acevedo, E. (2024). Inteligencia artificial al servicio de la pedagogía:
Potenciando la creatividad y el pensamiento crítico. PANORAMA, 18(34), 1-
13. https://www.redalyc.org/journal/3439/343977238002/html/
Panqueban, C., & Huincahue, J. (2024). Isomorphism between traditional learning
and AI-based pedagogy. Revista Universidad, 38(1), 20.
https://doi.org/10.15359/ru.38-1.20
Panqueban, D., & Huincahue, J. (2024). Inteligencia Artificial en educación
matemática: Una revisión sistemática. Uniciencia, 38(1), 357–373.
https://doi.org/10.15359/ru.38-1.20
| Leider Inocencio Saraiba Nuñez | Maira Rosario Moreno Pino | Orlando Figueredo Maldonado |
e9147
Plana, R. (2025). Artificial intelligence and equity in higher education. UNESCO-
IESALC. https://www.iesalc.unesco.org
Reicher, D., Silva, R., & Gomes, F. (2025). Engineering education in the age of AI:
Analysis of the impact of chatbots on learning. Education Sciences, 14(5),
484. https://doi.org/10.3390/educsci14050484
Reicher, H., Frenkel, Y., Lavi, M., Nasser, R., Ran-milo, Y., Sheinin, R., Shtaif, M.,
& Milo, T. (2025). A generative AI-empowered digital tutor for higher
education courses. Information, 16(4), 264.
https://doi.org/10.3390/info16040264
Rodríguez Cairo, A., & Ramírez Echavarría, L. (2023). Artificial intelligence in Cuban
universities: Pedagogical perspectives. Latin American Journal of Artificial
Intelligence, 28(3), 45–62. https://doi.org/10.62486/latia20238
Smith, A., & Brown, K. (2023). AI-driven adaptive learning systems in higher
education. Journal of Educational Computing Research, 61(5), 1123–1145.
https://doi.org/10.1177/07356331231123456
Springer, J. (2025). AI tutors in engineering education: Structural design
applications. Journal of Educational Technology, 25(2), 567–582.
https://doi.org/10.1007/s10758-025-09567
Tasdelen, O., & Bodemer, D. (2025). Generative AI in the Classroom: Effects of
Context-Personalized Learning Material and Tasks on Motivation and
Performance. International Journal of Artificial Intelligence in Education,
35(5), 3049-3070. https://doi.org/10.1007/s40593-025-00491-9
Torres, J., & Díaz, M. (2024). La inteligencia artificial como recurso didáctico en la
enseñanza universitaria cubana. Revista Pedagogía Universitaria, 39(1), 77–
95. https://doi.org/10.5281/zenodo.7654321
UNESCO. (2024). Artificial intelligence and the future of learning. Global Education
Monitoring Report. https://unesdoc.unesco.org/ark:/48223/pf0000387654
UNESCO-IESALC. (2025a). Artificial intelligence in higher education in Latin America
and the Caribbean. https://www.iesalc.unesco.org
UNESCO-IESALC. (2025b). Shaping inclusive AI futures in higher education. Informe
Regional. UNESCO IESALC. https://www.iesalc.unesco.org/en/2025/07/11
Wang, L., & Zhao, Y. (2025). AI tutors in civil engineering education. Journal of
Structural Engineering Education, 12(3), 210–225.
https://doi.org/10.1007/s10758-025-09568
World Bank Group. (2025). Artificial intelligence in higher education: Policy and
practice. https://documents.worldbank.org/en/publication/ai-higher-
education
Yin, R. (2021). Case study research and applications: Design and methods (6th ed.).
Springer. https://doi.org/10.1007/978-3-030-60147-3
Zhang, H., & Xu, F. (2023). Generative AI for problem-solving in mechanical
engineering. Journal of Mechanical Education, 18(2), 145–160.
| Leider Inocencio Saraiba Nuñez | Maira Rosario Moreno Pino | Orlando Figueredo Maldonado |
e9147
https://doi.org/10.1016/j.mechedu.2023.102345
Zhunio, A. Q., & Salgado, F. S. (2025). Inteligencia artificial debilita la inteligencia
de los jóvenes. Revista Cubana de Educación Superior, 44(especial 1), 50-
65. https://revistas.uh.cu/rces/article/view/12169
| Leider Inocencio Saraiba Nuñez | Maira Rosario Moreno Pino | Orlando Figueredo Maldonado |
Estrategia y Gestión Universitaria EGU
About the main author
Leider Inocencio Saraiba Nuñez
:
Doctor of Technical Sciences, holding a Master’s
degree in Applied Mathematics and Informatics for Management, and a graduate in
Aviation Radio
-
Electronics Engineering. He possesses a robust track record in
technical research, distinguished by his work on the development of mathematical
models and Artificial Intelligence systems applied to organizational efficiency. His
expertise encompasses energy carrier management, industrial maintenance, and the
integration of Industry 4.0 technologies within heritage environments. He has led
high
-
impact academic projects focused on competitiveness and digital
transformation, maintaining a methodological rigor geared toward top
-
tier
publications.
Declaration of author responsibility
Leider
Inocencio Saraiba Nuñez 1:
Conceptualization, Data Curation, Formal
Analysis, Investigation, Methodology, Resources, Software, Supervision,
Validation/Verification, Visualization, Writing
/Original Draft, and Writing/
Review &
Editing.
Maira Rosario Moreno P
ino 2:
Data Curation, Formal Analysis, Investigation,
Resources, Software, Supervision, Validation/Verification, Visualization, Writing—
Original Draft, and Writing
—Review & Editing.
Orlando Figueredo Maldonado
3: Data Curation, Formal Analysis,
Investigation,
Resources, Validation/Verification, Visualization, Writing
—Original Draft.
Financing:
Own resources
Special Acknowledgments:
