Gender Perspectives on Generative AI Use: The Case of the School of Digital Transformation at UTB

Presentación del conference paper en la IEEE Baja California Conference - IEEE BCC, octubre de 2025.

Abstract: This quantitative study analyzes the adoption of generative AI (GenAI) from a gender perspective in the School of Digital Transformation at Universidad Tecnológica de Bolívar (Colombia). A survey was conducted among 287 students (97 women, 190 men) during March–April 2025. Results show that 97.9% of women and 96.8% of men use GenAI for academic purposes. Men primarily use it to clarify complex concepts (83.7%) and report intensive daily use; women favor brainstorming (69.1%) and translation tasks (22.7%) with a more situational use pattern. Women emphasize inspiration and support for learning, while men focus on efficiency. Around half believe they could do without the tool, but one-third report some level of dependency, with women showing greater caution. As GenAI can both reproduce and reshape existing biases, the findings highlight the need for critical digital literacy with a gender sensitive approach in higher education.

1. Gender Perspectives on Generative AI Use:
The Case of the School of Digital Transformation at UTB
Tijuana, octubre 28 - 31 de 2025Tania Lucía Cobos Mercedes Posada César Viloria Núñez

2. Inspired by Umberto Eco’s Apocalyptic and Integrated: Two views on
mass culture:
?????? Apocalyptics – fear loss of creativity and autonomy
?????? Integrated – embrace tech’s potential and trust the audience.
A critical stance is needed: neither naive nor alarmist.
➡️ In today's world of hyper-surveillance and digital overconsumption,
GenAI raises new tensions: Does it democratize knowledge or
reinforce bias and power hierarchies?
Introduction – GenAI, Culture &
Gender Perspective

3. This study adopts a balanced, critical lens, asking:
➤ How does GenAI reproduce or challenge gender inequalities?
Case: students from the School of Digital Transformation at
Universidad Tecnológica de Bolívar (Cartagena de Indias, CO)
➤ Analyzing uses, motivations, barriers, and perceptions
Core assumption: Technology is not neutral: It matters who designs
it, who uses it, and under what power relations.
Introduction – GenAI, Culture &
Gender Perspective

4. Theoretical Framework
Generative AI: Rise and Reach
GenAI ChatGPT Rapid evolution ChatGPT Adoption
GenAI goes beyond
data analysis: Uses
neural networks to
“imagine, create, and
produce” content.
Nov 30, 2022: Launch
of ChatGPT, created
by OpenAI and
funded by Microsoft.
Major turning point
for the internet
history.
* Dec 2022: Perplexity
* 2023: LLaMA (Meta),
Bard → Gemini
(Google), Claude
(Anthropic)
* 2025: Multimodal
models (text, image,
audio, video) +
DeepSeek (China) +
Mistral (France)
→ 1 million users in 5
days
→ 100 million in 2
months
→ By 2025: 115–180M
daily users, 400M
weekly (OpenAI)
ChatGPT is the model
most used in 2025.

5. Generative AI: Emerging Risks
Despite enthusiasm, GenAI adoption faces key
challenges:
❗ Hallucinations – inaccurate or false outputs
❗ Bias & discrimination – embedded in training data
❗ Security breaches – including data leaks
❗ Privacy concerns – unclear data handling
❗ Low reliability – bibliographic and factual errors
❗ Language gaps – English-dominant performance
❗ Environmental cost – high energy consumption
These issues erode public trust and influence how
GenAI is adopted.

6. GenAI in Higher Education
Main adopters: youth aged 17–24 → university
students and they use GenAI for research, writing,
exercises, and coding.
Despite awareness of possible errors, enthusiasm
remains high.
The students see chatbot use as “cheating”, yet
they use it anyway.
?????? Rapid growth of studies on GenAI in education in
LATAM, but few include gender-disaggregated
data. Ex: Díaz-León and Iraola-Real (2024) in Peru
and Hernández-González et al. (2024) in Mexico.

7. Technology, gender and power
→ Digital economy extracts & sells behavioral data to shape actions.
→ GenAI is its most advanced form-trained on massive cultural datasets.
Surveillance Capitalism (Zuboff):
?????? Sexist, racist, heteronormative patterns in training data and design.
?????? Reinforcement of gender hierarchies under the guise of objectivity.
?????? Yet, critical and inclusive uses can diversify voices.
In patriarchal + consumerist systems, tech-gender relations are not neutral.
Studies show GenAI can reproduce stereotypes:

8. Critical Feminist Lens on GenAI
Key theorists:
→ Butler – gender is performative
→ Amorós – “objectivity” hides androcentric bias
→ Varela – men set the human standard
Evidence of bias:
– UNESCO: women → care roles; men → leadership
– Ávila: GPT-3.5 code = only white men as “good doctors”
Media view:
?????? McLuhan & Palomino: GenAI extends our emotional and
cognitive selves.
Takeaway:
Understanding gendered experiences helps build
more equitable and human-centered AI.

9. Innovative academic unit at Universidad Tecnológica de Bolívar (UTB), Cartagena de Indias - ????????????
Integrates technological, analytical, and communicative knowledge
Trains professionals for the challenges of the digital era
Promotes human-centered, interdisciplinary learning and pedagogical innovation
?????? Computer Engineering
?????? Data Science
?????? Marketing & Digital Transformation
??????️ Social Communication
Undergraduate programs
School of Digital Transformation at UTB

10. School of Digital Transformation at UTB
Gender patterns
reflect social
stereotypes (technical
vs. communicative
skills).
These trends shape
digital competencies
and career choices.
Reinforces the need
for a gender-sensitive
analysis of GenAI use.
944 = 100%

11. Methodology
Main project: “Adoption of GenAI among students of the School of Digital
Transformation – UTB”.
Objective: Analyze attitudes, uses, and perceptions of GenAI among
students, focusing on:
➤ Usage & frequency
➤ Motivations & barriers
➤ Changes in study habits
➤ Perceived dependency
➤ Gender differences
Approach:
?????? Quantitative, cross-sectional, non-experimental design → Measures
relationships without variable manipulation.
Sampling:
Stratified by convenience (initial random plan adjusted to include all valid
responses).
Study Desing

12. Methodology
Instrument:
→ Online survey (Microsoft Forms)
→ 30 mandatory questions (single/multiple choice, Likert scales, one open-
ended). Logic branching: if “No GenAI use”, survey ended early.
Period:
?????? March 1 – April 30, 2025
Participants:
Students from 4 academic programs (Computer Eng., Data Science,
Marketing, Communication).
Recruitment:
➤ Direct contact
➤ Institutional email
➤ “Savio” virtual platform
➤ Faculty support
➤ ?????? Incentive: US$50 Amazon gift card raffle
Data Collection

13. Methodology
Ethical compliance:
✅ Informed consent implied
✅ Anonymous and voluntary participation
✅ No personal data or institutional emails collected
Sample & Analysis
Data validation:
– Exported to MS
Excel.
– Cross-checked
with pivot tables (
divergence <0.1%).
– Frequency and
distribution
analyses only in
aggregated form.
287 = 100%

14. Analysis compares ???????????? vs. ????????????‍?????? students in:
➤ Use of GenAI
➤ Motivations and purposes
➤ Frequency of use
➤ Perceived changes and dependence

Data confirm widespread adoption across all programs
97.9% of ???????????? and 96.8% of ????????????‍?????? use GenAI for academic purposes
Gender patterns reveal complementary orientations:
?????? ???????????? → creative, learning-supportive uses (idea generation,
inspiration, translation).
⚙️ ????????????‍?????? → efficiency- and production-oriented uses (essays, graphics,
automation).
Findings
Overview Findings
Limitations: The survey used
only binary gender options
female/male, aligned with
traditional categories
women/men. As a result, the
analysis does not include non-
binary or other gender
identities.
Because the survey was self-
reported, some responses may
reflect social desirability bias —
answers seen as acceptable
rather than fully genuine.

15. Fig. 1 Academic Use of GenAI
0
20
40
60
80
100 Almost universal adoption among students: 97.9
% of ???????????? and 96.8 % of ???????????? use GenAI for academic
purposes.
Slightly higher adoption among women.
Main reasons for non-use: fear of losing skills,
distrust of accuracy, preference for manual
learning.
Reflects critical awareness, not rejection: GenAI
is now a normalized academic tool across
programs.
Findings
Yes
No96,8% 97,9% 3,2% 2,1% ???????????? ????????????‍?????? ???????????? ????????????‍??????

16. Fig 2. Uses of GenAI
0 20 40 60 80 100 Findings ???????????? ????????????‍??????
2,1%
Others
Generate graphics, images,
audio, and video
55,2%
17,5 %
55,2%
13,4 %
22,7 %
25,8 %
70,5%
71,1 %
19,5%
22,7 %
83,7%
73,2%
69.1%
52.6%
3,2%
Don’t use GenAI
2,1%
7,2%
Write essays and academic texts
Plan my academic activities
Correct what I have done (writing,
calculations...)
Translate texts
Explain complex concepts
Brainstorm ideas
?????? Women largely use AI to
understand complex concepts
(73,2%) and check their own
work (71.1%). They also use it for
brainstorming.
⚙️ Men (83.7 %) focus on
understanding complex
concepts and producing
outputs (essays 55.2 %,
graphics 55.2 %).
?????? Both genders use GenAI
similarly for correction tasks (≈
71 %), indicating learning
support rather than
replacement.
?????? Overall: women: creative
and supportive orientation |
Men: productive and analytical
orientation.

17. Fig 3. Motivations to use GenAI
0 20 40 60 80 100 ?????? Both genders use GenAI
mainly for inspiration when
blocked (???????????? 81.4 % | ???????????? 70.5 %),
showing its value as a creative
trigger.
?????? Learning support is more
relevant for women (???????????? 50 % |
???????????? 26.8 %), while efficiency and
time-saving motivate more
men (???????????? 66.8 % | ???????????? 40.2 %).
⏱️ Men also surpass women in
urgent-task assistance (52.1 %
vs. 37.1 %), reinforcing a
performance-driven pattern.
?????? In short: women use GenAI
as a cognitive and emotional
ally; men as a productivity and
speed tool.
Findings ???????????? ????????????‍??????
2,1%
Others
Makes the learning process easier
50.0%
26,8%
35,3%
40,2%
9,3%
13,2%
52,1%
42,1%
83,7%
3,2%
Don’t use GenAI
1,1%
1,0%
42,1%
26,8%
52,1%
37,1%
70,5%
81,4%
70,0%
64,9%
66,8%
40,2%
Give me confidence when reviewing
and improving my work
Offers better quality results that I can do by myself
Increase my academic productivity
Offers inmediate assistance in urgent moments
Adittional support when I feel
blocked or lack inspiration
Access to ideas and approaches I
wouldn’t have thought of
Reduces time and effort in repetitive
tasks

18. Fig 4. Frecuency of academic use of GenAI
0 20 40 60 80 100 ?????? Regular use is common
among both groups (≈ 97 %),
yet patterns diverge.
???????????? Men show higher intensive
use (almost every day 27.4 % |
several times a day 11.1 %),
suggesting systematic
integration in study routines.
???????????? Women report more
moderate use (several times a
week 45.4 % | occasionally 25.8
%), reflecting situational or
task-specific adoption.
?????? In short: women activate
GenAI when context demands
it; men embed it as a habitual
academic tool, evidencing
different levels of assimilation
and dependency.
Findings ???????????? ????????????‍??????
2,1%
Rarely use it
50.0%
35,3%
38,4%
3,2%
Don’t use GenAI
0,5%
0%
70,5%
70,0%
66,8%
19,5%
25,8%
45,4%
27.4%
14.4%
11.1%
12.4%
Occasionally
Several times a week
Almost every day
Several times a day

19. Fig 5. Perceived Changes in Study Habits Due to GenAI Use
0 20 40 60 80 100 ?????? Most students report
moderate changes due to
GenAI (???????????? 67.0 % | ???????????? 65.3 %).
⚖️ The intensity differs: men
are more likely to report
complete transformation (21.1
%) than women (9.3 %).
?????? Conversely, minor or barely
noticeable changes are more
common among women (15.5
%) than men (6.8 %).
?????? Overall: men appear to
integrate GenAI more deeply
into their study routines, while
women show a gradual,
adaptive assimilation to
GenAI-assisted learning. Findings ???????????? ????????????‍??????
2,1%
I don’t know
35,3%
38,4%
3,2%
Don’t use GenAI
0,5%
1,0%
65,3%
66,8%
5,2%
15,5%
6,8%
67,0%
21,1%
9,3%
It hasn’t change at all
Yes, but I have barely noticed changes
Yes, it has changed in some
aspects
Yes, it has completely changed the way I study
3,2%

20. Fig 6. Perceived Dependence on GenAI Use
0 20 40 60 80 100
⚖️ Overall dependence is low,
suggesting that students
maintain autonomy while
using GenAI.
???????????? Women show greater
caution and controlled use,
with higher rates of moderate
or limited dependence.
???????????? Men report higher
integration of GenAI into daily
tasks, reflecting a stronger
habit of incorporation rather
than overreliance.
?????? Interpretation: results
indicate balanced adoption.
GenAI is widely used but
without critical dependence or
loss of self-efficacy.
Findings ???????????? ????????????‍??????
2,1%
I don’t know
35,3%
45.4%
3,2%
Don’t use GenAI
2,6%
0%
65,3%
66,8%
17,5%
50.5%
21,1%
No, I don’t use it enough to develop dependency
10,0%
Yes, to some extent
31,1%
30,9%
2,6%
4,1%
No, I use it but I can do without it
Yes, absolutely

21. Discussion
These differences reflect
socialized learning roles rather
than ability gaps → Women
use GenAI to enrich learning,
men to optimize performance.
Patterns align with prior
studies in Peru and Mexico,
confirming that GenAI both
reproduces and reshapes
gender socialization in
education.
Findings highlight the need
for critical GenAI literacy that
questions algorithmic bias and
promotes equitable
adoptiontion.

22. Conclussions and Implications
GenAI is now integral to academic life, but its meaning and use differ by gender.
→ Women: creative ally for learning
→ Men: operational accelerator
Perceived dependency remains moderate — users retain autonomy.
→ An opportunity to strengthen ethical and reflective use.
Institutional recommendations:
✅ Integrate ethics and gender-sensitivity into digital literacy programs.
✅ Encourage inclusive and locally grounded GenAI design.
✅ Promote critical pedagogy over purely technical training.
Future research:
➤ Include non-binary and intersectional identities.
➤ Combine quantitative data with qualitative inquiry on meanings and
experiences.

23. References
[1] U. Eco, Apocalípticos e integrados. Debolsillo, 2011.
[2] G. Lipovetsky, La era del vacío: Ensayos sobre el individualismo contemporáneo, 2nd ed. Barcelona: Anagrama, 2006.
[3] E. P. Gil, “Technofeminism, de judy wajcman [rese.a],” UOC Papers, no. 5, 2007, disponible en https://www.uoc.edu/uocpapers/5/dt/esp/gil.pdf.
[4] C. D’Ignazio and L. F. Klein, Data Feminism. Cambridge, MA: MIT Press, 2020.
[5] E. Riquelme. (2023) Chatgpt hipnotiza y consterna a 6 meses de su lanzamiento. [En l.nea]. El Economista. [Online]. Available: https://surl.li/cffoog
[6] I. Oriol. (2024) Inteligencia artificial generativa. Informe técnico, Universitat Oberta de Catalunya. [Online]. Available: https://n9.cl/zt8oo
[7] Reuters. (2023) Meta lanza el modelo de lenguaje ai llama. [Online]. Available: https://n9.cl/e15qpj
[8] E. Batista, J. Zhu, and F. Potkin. (2025) Deepseek apresura el lanzamiento de un nuevo modelo de ia en medio de la ofensiva de china. [Online]. Available: https://n9.cl/mjmz2
[9] A. Edwards. (2023) A.o uno de chatgpt: as. ha cambiado el mundo desde la llegada del chatbot de openai. [En l.nea]. Wired Espa.a. [Online]. Available: https://surl.li/xmqjym
[10] P. Jones. (2025) 78 estadísticas y tendencias de la inteligencia artificial para 2024. [En línea]. Semrush Blog. [Online]. Available: https://es.semrush.com/blog/etica-de-la-inteligencia-artificial/
[11] L. Huang et al., “A survey on hallucination in large language models: Principles, taxonomy, challenges, and open questions,” arXiv preprint arXiv:2311.05232, 2024.
[12] Y. Qu and J. Wang, “Performance and biases of large language models in public opinion simulation,” Humanities and Social Sciences Communications, vol. 11, p. 1095, 2024.
[13] Wald.ai. (2024) Chatgpt data leaks and security incidents 2023–2024: A comprehensive overview. [Online]. Available: https://acortar.link/ueoE0j
[14] S. Alkamli and R. Alabduljabbar, “Understanding privacy concerns in chatgpt: A data-driven approach with lda topic modeling,” Heliyon, vol. 10, no. 20, p. e39087, 2024.
[15] F. Farhat, S. S. Sohail, and D. Madsen, “How trustworthy is chatgpt? the case of bibliometric analyses,” Cogent Engineering, vol. 10, no. 1, 2023.
[16] Y. Guo, M. Guo, J. Su, Z. Yang, M. Zhu, H. Li, M. Qiu, and S. S. Liu, “Bias in large language models: Origin, evaluation, and mitigation,” arXiv preprint arXiv:2411.10915, 2024.
[17] D. Patterson, J. Gonzalez, U. H.lzle, Q. V. Le, C. Liang, L.-M. Munguia, D. Rothchild, D. So, M. Texier, and J. Dean, “The carbon footprint of machine learning training will plateau, then shrink,” TechRxiv, preprint, Feb 2022.
[18] A. Choudhury and H. Shamszare, “Investigating the impact of user trust on the adoption and use of chatgpt: Survey analysis,” Journal of Medical Internet Research, vol. 25, p. e47184, 2023.
[19] L. Halpern. (2023, Dec.) The year ai ate the internet. [Accessed: Jul. 16, 2025]. [Online]. Available: https://n9.cl/mvhzkx
[20] I. D.az-Le.n and I. Iraola-Real, “El uso de chatgpt por estudiantes universitarios peruanos: Una exploración del uso real y la intenci.n de uso,” Revista Ibérica de Sistemas e Tecnologías de Informaci.n, no. E69, pp. 89–98, 2024, [Accessed: Jul. 16, 2025].
[Online]. Available: https://www.risti.xyz/issues/ristie69.pdf
[21] M. Hernández-González, J. M. R. Quiroz, M. del C. Trejo C.zares, and H. D. Q. Mart.nez, “Percepción de los estudiantes con perspectiva de género sobre el uso de la iag en la educaci.n superior en méxico,” in Argumento y usos tecnopedag.gicos de la
inteligencia artificial, E. R.-V. S.nchez and J. B. L.pez, Eds. México: SOMECE, 2024, pp. 114–129.
[22] S. Zuboff, The Age of Surveillance Capitalism: The Fight for a Human Future at the New Frontier of Power. New York: PublicAffairs, 2019.
[23] UNESCO. (2024) Ia generativa: Un estudio de la unesco revela pruebas alarmantes de estereotipos de g.nero regresivos. [Online]. Accessed: Jul. 31, 2025. [Online]. Available: https://acortar.link/NaFEOS
[24] D. Ávila. (2023, Dec) Desafíos éticos en la ia: sesgos de género y racial en modelos de lenguaje generativos. [Online]. Accessed: Jul. 31, 2025. [Online]. Available: https://n9.cl/ro5sf
[25] Revista Anfibia. (2024, Mar) Se lanza latamgpt: una ia latinoamericana es posible. [Online]. Accessed: Jul. 31, 2025. [Online]. Available: https://n9.cl/qf09nh
[26] J. Butler, Gender Trouble: Feminism and the Subversion of Identity. New York: Routledge, 1990.
[27] C. Amor.s, Hacia una crítica de la razón patriarcal. Madrid: Anthropos, 1985.
[28] N. Varela, Feminismo para principiantes: Edici.n revisada, actualizada y ampliada. Barcelona: Ediciones B, 2019.
[29] M. McLuhan, Understanding Media: The Extensions of Man. New York: McGraw-Hill, 1964.
[30] J. Palomino, “De pr.tesis afectivas y otras configuraciones: cuerpos, subjetividades y afectividad en la era del celular.” Editorial Pontificia Universidad Javeriana, 2023. [31] M. Akour and M. Alenezi, “Higher education future in the era of digital
transformation,” Education Sciences, vol. 12, no. 11, p. 784, 2022.
[32] F. J. García-Peñalvo, “Digital transformation in the universities: implications of the covid-19 pandemic,” Education in the knowledge society (EKS), vol. 22, 2021.
[33] L. M. C. Benavides, J. A. Tamayo Arias, M. D. Arango Serna, J. W. Branch Bedoya, and D. Burgos, “Digital transformation in higher education institutions: A systematic literature review,” Sensors, vol. 20, no. 11, p. 3291, 2020.
[34] J. Oudenampsen, E. Das, N. Blijlevens, and M. H. van de Pol, “The state of the empirical evidence for interdisciplinary learning outcomes in higher education: A systematic review,” The Review of Higher Education, vol. 47, no. 4, pp. 467–518, 2024.
[35] H. W. Routhe, J. E. Holgaard, and A. Kolmos, “Experienced learning outcomes for interdisciplinary projects in engineering education,” IEEE Transactions on Education, vol. 66, no. 5, pp. 487–499, 2023.
[36] World Health Organization. (2022) Gender and health. [Online]. Available: https://www.who.int/health-topics/gender
[37] UN Women. (2022) Plan estratégico de onu mujeres 2022–2025. Social norms. [Online]. Available: https://www.unwomen.org/es/ un-women-strategic-plan-2022-2025/social-norms
[38] Office of the United Nations High Commissioner for Human Rights (OHCHR). (2020) Gender stereotyping. [Online]. Accessed: Jul. 31, 2025. [Online]. Available: https://www.ohchr.org/en/women/ gender-stereotyping
[39] R. Hernández, C. Fernández, and P. Baptista, Metodología de la investigación, 6th ed. McGraw-Hill, 2014.

24. Thank you! Alberto Navarro
Tania Lucía Cobos
Professor, School of Digital Transformation
Social Communication Program
Mercedes Posada Meola
Professor, School of Digital Transformation
Social Communication Program
César Viloria Núñez
Dean, School of Digital Transformation
Universidad Tecnológica de Bolívar
Cartagena de Indias, Colombia