Pedagogical Integration of Generative Artificial Intelligence in Higher Education: Teachers’ Perceptions by Gender
Integración
pedagógica de la IAG en educación superior: percepciones del profesorado
diferenciadas por género
Carmen Romero García. Universidad
Internacional de la Rioja. España.
Olga Buzón García. Universidad de
Sevilla. España.
Alba Vico Bosch. Universidad
Internacional de la Rioja. España.
Received: 2026-03-04; Reviewed: 2026-03-12; Accepted: 2026-05-21; Published: 2026-09-01.
How
to cite:
Romero García, C., Buzón García, O.,
& Vico Bosch, A. (2026). Pedagogical Integration of Generative
Artificial Intelligence in Higher
Education: Teachers’ Perceptions
by Gender [Integración pedagógica de la IAG en
educación superior: percepciones del profesorado diferenciadas por género]. Pixel-Bit, Revista de Medios y Educación, 77, Art. 3. https://doi.org/10.12795/pixelbit.120444
ABSTRACT
Generative artificial intelligence (GenAI) is transforming teaching, learning and assessment processes in higher education. This study examines university teachers’ perceptions of the pedagogical integration of GenAI, identifying potential gender differences in teaching–learning processes, assessment practices and the use of AI tools. A mixed-methods design was adopted, combining qualitative thematic analysis with descriptive and inferential quantitative analysis. The sample consisted of 540 university teachers from the Universidad Internacional de La Rioja who completed an online questionnaire with open-ended questions. The results indicate a generally positive perception of GenAI, while also revealing differences in how men and women conceptualise its integration. In teaching–learning processes, men prioritise automation and time efficiency, whereas women emphasise personalised learning, ethical considerations, and teacher training. In assessment, both groups highlight personalisation and the teacher’s role, although men place greater emphasis on evaluation models, while women stress feedback. Regarding uses, content generation, audiovisual resources and the creation of teaching materials predominate in both groups, while men show greater presence in uses related to research activities.
RESUMEN
La inteligencia artificial
generativa (IAG) está transformando los procesos de enseñanza, aprendizaje y
evaluación en la educación superior. Este estudio analiza la percepción del profesorado
universitario sobre la integración pedagógica de la IAG, identificando posibles
diferencias según género en los procesos de enseñanza-aprendizaje, evaluación y
uso de herramientas. Se adoptó un diseño metodológico mixto que combina
análisis temático cualitativo con análisis cuantitativo descriptivo e
inferencial. La muestra estuvo compuesta por 540 docentes universitarios de la
Universidad Internacional de La Rioja que respondieron a un cuestionario online
con preguntas abiertas. Los resultados evidencian una percepción
mayoritariamente positiva hacia la IAG, mostrando diferencias en la forma en
que hombres y mujeres conceptualizan su integración. En los procesos de
enseñanza-aprendizaje, los hombres priorizan la automatización y el ahorro de
tiempo, mientras que las mujeres destacan la personalización del aprendizaje,
la ética y la formación docente. En evaluación, ambos coinciden en la
personalización y el rol docente, aunque los hombres enfatizan el modelo
evaluativo y las mujeres la retroalimentación. En los usos, predomina en ambos
grupos la generación de contenidos, recursos audiovisuales y materiales
didácticos, mientras que los hombres presentan mayor presencia en usos
vinculados a la investigación.
KEYWORDS · PALABRAS
CLAVES
Higher education; artificial intelligence; educational evaluation; university teachers; gender differences · Educación superior; inteligencia artificial; evaluación
educativa; profesorado universitario; diferencias de género.
1.
Introduction
Generative artificial intelligence (GenAI) has
redefined teaching and learning processes in the educational context. In higher
education, it represents a process of structural change in teaching, assessment
and research practices. Bannister et al. (2023) highlight the disruptive
potential of these tools and the need to rethink teaching and assessment
practices.
Several studies emphasise the benefits of GenAI
in higher education (Bond et al., 2024; García-Peñalvo
et al., 2024; Sozon et al., 2025), such as the
personalisation of learning, the automation of tasks and assessment, and
academic guidance and support for students. However, they also reveal
challenges such as limited teacher competence, insufficient attention to
ethical issues, and a lack of solid evidence regarding its actual impact on
learning.
Recent studies have analysed students’ perceptions
of these emerging technologies (Kim et al., 2025; Melchior & Farinosi, 2026); however, the analysis of teachers’
perceptions is more limited, particularly in higher education.
1.1. AI in Teaching, Learning, and Assessment
The integration of GenAI into higher education
has been approached from two perspectives: an instrumental approach aimed at
optimising academic and administrative tasks, and a transformative approach that
redefines the educational ecosystem, pedagogical models and learning dynamics
(Luckin et al., 2022).
The scientific literature highlights the impact
of GenAI on teaching and learning processes, particularly in terms of
personalisation (Vorobyeva et al., 2025). These tools
facilitate content adaptation, immediate feedback and individualised support
(Liu et al., 2025). Furthermore, GenAI also automates mechanical and repetitive
tasks, such as the development of materials or the marking of multiple-choice
tests, freeing up teachers’ time for activities of greater educational value
(Bond et al., 2024).
AI-mediated assessment has become the focus of
academic debate. Yan et al. (2025) highlight its potential to optimise the
design of tests and assessments. Likewise, personalised assessment and
immediate feedback offer significant opportunities for changes. However, some studies
question the validity of assessment in these settings, pointing to the need to
prioritise critical thinking, reflection and reasoning over mere written output
(Cooper et al., 2023; Cotton et al., 2023).
This shift in teaching practices raises
dilemmas regarding the potential dehumanisation of the educational process.
Tools such as ChatGPT, and their ability to generate content, could foster
excessive technological dependence (García-Peñalvo et
al., 2024). The ethical dimension requires the establishment of robust
regulatory frameworks (Gallent-Torres et al., 2023)
as well as critical digital literacy among teachers and students (Holmes et
al., 2022).
The use of GenAI in higher education has
diversified, encompassing multiple areas of academic activity. García-Peñalvo et al. (2024) analyse the implications and uses of
GenAI in education, focusing on the potential educational use of various types of
tools and their functionalities. These authors classify tools that primarily
create content (Jasper, Notion), exams (Conker, Monic), images and
presentations (DALL-E, Midjourney and ChatGPT), audio and video (Fliki, Make a Video) and support research (ChatPDF, Consensus, Copilot).
For their part, Díaz-Vera et al. (2024)
highlight the workload of teachers, divided into: 1) academic, to keep up to
date, follow curricula and syllabuses, and assess student learning; 2) administrative,
to record student attendance and performance and produce reports. From this
perspective, they consider it essential to keep up to date with the various
tools and their strategic application in education. Sánchez (2023) states that
most teachers use GenAI to prepare lessons and educational materials, rather
than to integrate them with their students in the classroom.
1.2. Gender Perspective in Technology Adoption
The European Commission (2021) highlights the need
to strengthen teachers’ digital skills, recognising the gender perspective and
the need to bridge the digital divide. The use of GenAI tools by teachers in
higher education contexts has been extensively explored; however, gender
differences have not been examined in depth (Miranda and Chamorro-Mera, 2025).
Studies on gender differences in the adoption
of technologies by women and men are inconclusive. Alissa & Hamadneh (2023) found that female teachers possess greater knowledge
of and make greater use of AI in higher education. In contrast, Aksongur & Bağrıacık
(2025) found no differences in attitudes between women and men regarding the
use of AI.
Blázquez et al. (2025) found gender-based
differences in the use of GenAI among university lecturers. Men perceive these
tools as useful for optimising academic tasks, improving productivity and
supporting pedagogical innovation. Women engage in a wider range of activities (supervising
academic work and content creation) and display greater caution, expressing
greater concern regarding ethics, the reliability of the generated content and
the potential impact on academic integrity.
These gender differences in behaviour can be
explained through social psychology and gender roles, which suggest that men
tend to be more pragmatic, results-oriented and risk-taking, whilst women tend
to be more risk-averse and value ease of use (Miranda and Chamorro-Mera, 2025).
Furthermore, drawing on feminist literature, it is argued that the relationship
between gender and technology is not solely a matter of individual
characteristics, but rather of sociocultural structures that have historically
shaped technology as a masculinised space (Wajcman,
2006). In this vein, Sanders (2006) notes that educational and technological
environments frequently reproduce gender stereotypes that shape perceptions,
expectations and levels of confidence regarding the use of technologies. Similarly,
Rebollo-Catalán et al. (2012) highlight the importance of incorporating a
co-educational perspective in the digital sphere, emphasising the role of
teachers and educational practices in fostering more inclusive and egalitarian
uses of technology. From this perspective, women may demonstrate greater
sensitivity and critical awareness regarding the potential risks and challenges
associated with the use of emerging technologies (Criado, 2019).
Considering the above, this study aims to analyse
university lecturers’ perceptions of the pedagogical integration of GenAI,
identifying gender-based differences. Specifically, it proposes to:
·
Analyse gender-specific
perceptions of the integration of AI into teaching, learning and assessment
processes.
·
Analyse gender-specific
perceptions of the integration of AI into assessment processes.
·
Identify the uses and tools of
AI in education, broken down by gender.
2. Methodology
A mixed-methods approach was employed using a convergent
design with a qualitative focus (Creswell & Plano Clark, 2011). This design
enabled the integration of thematic analysis of open-ended responses with
quantitative procedures aimed at identifying frequencies, percentages and
associations between gender and thematic categories. At the qualitative level,
a thematic analysis was conducted to identify how teachers conceptualise GenAI
in relation to teaching and learning processes, assessment, and the uses and
types of tools. Furthermore, gender-related differences in discourse were
explored. The quantitative analysis enabled us to describe the distribution of
the emerging categories and to analyse possible statistically significant
associations between gender and thematic categories.
2.1. Sample
The sample consisted of 540 university
lecturers from the International University of La Rioja. 60.93% were women,
37.96% were men, and 1.11% did not specify their gender. In terms of educational
level, teaching at postgraduate level (46.48%) and undergraduate level (30.37%)
predominates, with 14.07% teaching at doctoral level. The sample has a high
level of professional experience, as 66.3% have five or more years of teaching
experience.
2.2. Instrument
A questionnaire comprising two sections was
used to collect data: one section covered the sample’s sociodemographic
variables, whilst the other consisted of three open-ended questions that allowed
teachers to freely express their views and experiences regarding the
integration of GenAI into teaching and learning processes, assessment, and the
uses and types of tools.
The questionnaire was administered online.
Participation was voluntary, anonymous and confidential, and it was guaranteed
that the data would be used for academic and research purposes.
2.3. Data analysis
The qualitative analysis was based on three
dimensions defined in accordance with the research objectives: (1) Integration of
GenAI into teaching and learning processes, (2) Integration of GenAI into
assessment processes, and (3) Uses and types of GenAI tools. Based on these, an
inductive thematic analysis was carried out following the approach proposed by
Braun and Clarke (2006, 2021), combined with open coding procedures (Saldaña,
2021). The analysis included the following phases: (a) a thorough reading of
the entire corpus; (b) identification of relevant units of meaning; (c) initial
coding of the text fragments; (d) grouping into emerging thematic categories;
(e) iterative review and refinement of the themes; and (f) final definition of
the categorical system.
Multiple coding was applied, allowing a single
excerpt to be assigned to more than one category when it contained distinct
meanings. In total, 540 open-ended responses were analysed
and 1,726 text excerpts were coded across the various dimensions examined. To
explore gender-related differences, a cross-group thematic comparison (women/men)
was carried out, examining both the frequency distribution of the categories
and the discursive nuances present in each group. This procedure enabled the
identification of variations in how the integration of GenAI into teaching
practice was conceptualised. The data was managed and coded using Atlas.ti 22 software.
Based on the category system, absolute
frequencies and percentages were calculated for the total number of segments coded
in each dimension (Table 1). Subsequently, contingency tables were produced by
cross tabulating the thematic categories with the gender variable. To examine
the existence of statistically significant associations, the Chi-square test of
independence was applied, and the effect size was estimated using Cramer’s V
coefficient, using Jamovi software. The effect size
was interpreted according to Cohen (1988): 0.10 as a small effect, 0.30 as a
medium effect and 0.50 as a large effect.
Table 1
Dimensions,
thematic categories and distribution of coded segments
|
Dimension |
Category |
Description |
N |
% |
|
1. Integrating AI into teaching and
learning processes (N = 1.238) |
Personalised learning |
Tailoring content, pace and
strategies to individual needs |
312 |
25,2 |
|
Automation and timesaving |
Optimisation of teaching tasks and
automatic generation of materials |
276 |
22,3 |
|
|
Methodological change |
Transformation of teaching methods
and redesign of activities |
241 |
19,5 |
|
|
Ethics and responsibility |
Ethical use, regulation and risks
associated with AI |
211 |
17,0 |
|
|
Teacher training |
The need for professional
development in AI |
198 |
16,0 |
|
|
2. Integration of AI into
assessment processes (N = 226) |
Personalisation |
Performance-based assessment |
45 |
19,9 |
|
Teaching role |
Maintaining professional standards
among teachers |
43 |
19,0 |
|
|
Assessment model |
Revision of criteria and
instruments |
37 |
16,4 |
|
|
Feedback |
Generation of immediate,
personalised feedback |
37 |
16,4 |
|
|
Ethics |
Plagiarism detection and ethical
responsibility |
27 |
11,9 |
|
|
Automation |
Automatic marking of assignments |
19 |
8,4 |
|
|
Competency assessment |
Assessment of real-world
performance and complex skills |
18 |
8,0 |
|
|
3. Types of use (N = 262) |
Word processing / content creation |
Drafting and rewriting academic
texts |
99 |
37,8 |
|
Research |
Researching and summarising
scientific information |
43 |
16,4 |
|
|
Audiovisual resources |
Creating images, presentations and
videos |
41 |
15,6 |
|
|
Development of teaching materials |
Designing activities and lessons |
36 |
13,7 |
|
|
Not used / minimal use |
No use of AI, or limited use |
32 |
12,2 |
|
|
Assessment |
Use for academic marking or
assessment |
11 |
4,2 |
3. Results
3.1. Integration of AI into the teaching and
learning process
Analysis of this dimension, considering both
men and women, reveals a favourable perception of the incorporation of GenAI in
higher education, though this is contingent upon factors such as appropriate pedagogical
integration, ethical use and teacher training (Table 1). The use of GenAI is
perceived as a support for teaching and as a key resource for personalising
learning (25.20), facilitating the adaptation of content and strategies to
students’ individual needs. Furthermore, GenAI is highlighted as a mechanism
for automating certain tasks (22.30%), generating materials, searching for
information and conducting assessment processes. A significant proportion of
teachers identify GenAI as a driver of methodological change (19.50%), capable
of driving transformations in teaching strategies and the design of activities.
At the same time, there is concern regarding the ethical conditions of its
implementation (17%), including aspects related to reliability, bias,
responsible use and critical thinking. Furthermore, a significant proportion of
comments highlight teacher training (16%) as an essential requirement for
proper pedagogical integration.
3.1.1. Thematic distribution by gender
The percentage analysis by gender reveals
significant differences in the thematic distribution (Table 2). Among men, the
categories with the highest weighting were automation and timesaving,
personalisation of learning, followed by methodological change. For women, the
predominant category was personalisation of learning, followed by ethics and
responsibility and teacher training. The most notable differences were observed
in automation, which carried greater weight among men (27.95%) than among women
(17.97%), in teacher training, with a higher weighting among women (18.56%)
compared to men (13.07%); and in ethics and responsibility, with a higher
presence among women (19.59%) compared to men (14.16%).
The distribution of categories is not
independent of gender, and there are distinct patterns in the thematic
emphasis. The chi-square test indicated a significant association between
gender and thematic distribution, χ² (4, N = 1238) = 27.37, p < .001,
with a small effect size (V = 0.15). The analysis of standardised residuals
showed that the largest contributions to the χ² value came from the
categories of automation and time saving, and teacher training.
Table 2.
Distribution
of categories by gender for dimension 1
|
Categories |
Men n |
% Men |
Women n |
% Women |
|
Personalised learning |
128 |
23,23 % |
176 |
25,92 % |
|
Automation and timesaving |
154 |
27,95 % |
122 |
17,97 % |
|
Methodological change |
119 |
21,60 % |
122 |
17,97 % |
|
Teacher training |
72 |
13,07 % |
126 |
18,56 % |
|
Ethics and responsibility |
78 |
14,16 % |
133 |
19,59 % |
Note. Multiple coding. Coded
segments: men: 551; women: 679.
Discourse analysis reveals significant nuances
in the way the pedagogical integration of AI is conceptualised.
Regarding personalised learning, women approach
the issue from the perspective of educational responsibility and guidance on
the use of technology: “It enables far more effective individualised learning pathways;
we must teach students how to use it to enhance their learning.” In contrast,
men focus on technology and its processing capacity: “AI will revolutionise
higher education through personalised learning, virtual tutors and data
analysis for efficient and timely improvements.”
In the category of automation and timesaving,
the difference between the two perspectives is particularly striking. For men,
it is framed in terms of productivity and efficiency: “it will streamline processes
that would otherwise take away from effective teaching and feedback time […] it
will optimise teaching work”. For women, efficiency is presented as secondary
to quality, critical review and pedagogical support: “it allows us to carry out
tasks more quickly, but we must be careful and check the results” […] “it is a
tool that optimises time, although it is important to verify the information”.
About the methodological change, both genders acknowledge
its impact, albeit with different nuances in their discourse. Men emphasis the
transformation of teaching as a broad-based change: “its use will change
teaching methodology and the way students learn”. For their part, women
approach it in a more specific way, in terms of pedagogical regulation and
placing the emphasis on the need to redesign specific tasks to preserve the
authenticity of learning: “tasks will have to be rethought so that they cannot
be performed by AI”.
In the teacher training category, women
highlight the need for training as an essential prerequisite for the
pedagogical integration of AI: “Institutional training is required to integrate
these tools properly and ensure that this change is effective.” Men also
recognise the importance of training, but see it as more closely linked to
technical competence than to in-depth pedagogical reflection: “specialisation
in digital development will be a mandatory requirement” […] “more training is needed
in this area”.
With regard to the category of ethics and
responsibility, women put forward more detailed arguments concerning
responsibility, regulation and the preservation of critical thinking: “it is
essential to ensure that technology is used in an ethical, inclusive and
responsible manner” […] “I am concerned about the biases that have been
identified and the impact on students’ critical thinking”. Among men, ethical
concerns are mentioned as a supplementary caveat rather than a central theme: “it
can be a great help, but we must be mindful of its limitations and biases”.
3.2. Integration of AI into assessment
processes
Analysis of men’s and women’s responses
regarding assessment reveals a largely positive yet qualified perception of the
pedagogical use of GenAI. As shown in Table 1, teaching staff highlight the
potential of AI for personalising assessment processes (19.90%), emphasising
the role of the teacher (19%) and noting that AI cannot replace professional judgement
in assessment. Secondly, they comment that GenAI affects the design of the
assessment model and feedback (both at 16.40%), highlighting the need to
reformulate criteria and tools that allow for automation and immediate
feedback. Ethical concerns (11.90%) linked to equity, transparency and the
responsible use of technology are evident. Reservations regarding automation
(8.4%) are also noted, specifically concerning the risk of dehumanising the assessment
process. Finally, the assessment of competences (8%) highlights the need to
review what and how assessment is carried out in a context where GenAI can
influence the production of evidence of learning.
3.2.1. Thematic distribution by gender
The analysis of percentages by gender reveals
more moderate differences than those observed in dimension 1 (Table 3). Among
men, the categories with the highest weighting were personalisation, the teaching
role and the assessment model. For women, the most prominent categories were
personalisation, the teaching role and feedback. The most notable differences
were observed in skills assessment, with a higher weighting for men (12.50%)
than for women (4.62%), and in automation, also with a higher weighting for men
(11.46%) than for women (6.15%). Conversely, women gave greater weight to
personalisation (22.31%) compared to men (16.67%) and to feedback, with greater
weight given by women (19.23%) compared to men (16.67%).
To determine whether these differences were
statistically significant, a chi-square test of independence was carried out,
yielding χ² (6, N = 226) = 9.42, p = .151. This result indicates that,
although there are noticeable differences in percentages, the distribution of
topics does not depend significantly on gender in the context of assessment.
Table 3
Distribution
of categories by gender for dimension 2
|
Categories |
Men n |
% Men |
Women n |
% Women |
|
Personalisation |
16 |
16,67 % |
29 |
22,31 % |
|
Teaching role |
16 |
16,67 % |
27 |
20,77 % |
|
Assessment model |
16 |
16,67 % |
21 |
16,15 % |
|
Feedback |
12 |
12,50 % |
25 |
19,23 % |
|
Ethics |
13 |
13,54 % |
14 |
10,77 % |
|
Automation |
11 |
11,46 % |
8 |
6,15 % |
|
Competency assessment |
12 |
12,50 % |
6 |
4,62 % |
Note. Multiple coding. Coded
segments: men: 96; women: 130.
Although the quantitative differences do not
reach statistical significance, the discourse analysis reveals relevant
nuances.
With regard to
feedback, women emphasise the formative aspect of the assessment process: “it
allows for the design of more personalised and adaptive assessments, offering
immediate feedback to students”, highlighting its potential to support
learning. For men, feedback is usually linked to the systematisation and
comparison of results: “analysing performance and comparing results in less
time”.
In the personalisation category, women
highlight the focus on tailoring the process to the individual student: “The approach
is more personalised to the student’s needs because it facilitates feedback.”
Among men, the reference is linked to analytics and performance, emphasising
the technical aspect of data analysis: “using learning analytics to personalise
the assessment process”.
When it comes to competency assessment, men
most frequently mention the need to assess real-world performance and complex
skills, arguing that AI can help to “assess learning outcomes in greater depth”.
Among women, this category is mentioned less frequently and is more integrated
into the overall educational approach: “We need to rethink some assessment
models, but above all we need to train students so that they can use AI to
improve their learning.”
When it comes to automation, the male discourse
presents it as a means of improving efficiency: “it will simplify marking
processes and optimise teaching work”. In the female discourse, it is linked to
freeing up time for more qualitative tasks: “it can be used for automatic
marking, allowing more time to be devoted to meaningful feedback”.
About the ethical dimension, the male discourse
emphasises control and detection: “generating new ways to prevent and detect
cheating”. In the female discourse, the concern is articulated in terms of
fairness and educational responsibility: “it is necessary to ensure the ethical
and fair use of these tools”.
3.3. Uses of GenAI tools
An analysis of the responses regarding the use
of AI tools reveals widespread and diverse adoption among teaching staff.
Unlike what was observed in the previous sections, the focus of the responses
to this question is more on effective practices than on evaluations (Table 1).
The predominant use relates to word processing
and content creation (37.80%), mainly for drafting texts, rewriting content,
creating outlines and providing support for academic tasks. In second place are
research (16.40%) and the creation of teaching materials (13.70%), including
the development of activities, assessment rubrics and lesson planning. As for
audiovisual resources (15.60%), there is a significant presence of tools for
generating images, presentations and educational videos. To a lesser extent,
there are references to assessment (4.20%), as well as to non-use or limited
use (12.20%), indicating that a section of the teaching staff continues to report
limited or no use, generally associated with a lack of training or confidence.
3.3.1. Thematic distribution by gender
The percentage analysis within each gender
reveals differences in the thematic distribution (Table 4). The results of the
chi-square test showed no significant association between gender and category
of use (χ² (5, N = 262) = 3.47, p = .627). This result indicates that,
although percentage differences are observed in some codes, these variations do
not constitute a structurally distinct pattern. This is the case for uses
linked to research, where there is a higher proportion
of men (20.95%) compared to women (13.38%). The same applies to the ‘no use’ or
‘minimal use’ code, where there is a higher proportion
of women (14.01%) compared to men (9.52).
The most common uses in both groups were word
processing and content creation (36.19% among men and 38.85% among women), the
creation of teaching materials (16.19% among men and 15.29% among women), and the
use of audiovisual resources (13.33% among men and 14.01% among women).
Table 4
Distribution
of categories by gender for dimension
|
Categories |
Men n |
% Men |
Women n |
% Women |
|
Word processing / content creation |
38 |
36,19 % |
61 |
38,85 % |
|
Research |
22 |
20,95 % |
21 |
13,38 % |
|
Audiovisual resources |
14 |
13,33 % |
22 |
14,01 % |
|
Development of teaching materials |
17 |
16,19 % |
24 |
15,29 % |
|
Assessment |
4 |
3,81 % |
7 |
4,46 % |
|
Not used / minimal use |
10 |
9,52 % |
22 |
14,01 % |
Note. Multiple coding. Coded segments:
men: 105; women: 157.
Qualitative analysis enables us to identify
relevant nuances in the way teachers describe their relationship with GenAI
tools.
With regard to
word processing and content generation, among men, use is often associated with
technological exploration and process optimisation: “I use it to write texts
and generate content quickly”. Among women, use is more frequently described in
terms of specific pedagogical applications: “I use it for everything from course
design to assessment, including lesson preparation and providing feedback to
students”.
In the research category, men show a greater
focus on research efficiency and information analysis: “I have used them mainly
to speed up my research and literature review processes”. For women, research
appears to be more integrated into general teaching tasks: “I use it to search
for information and adapt content for my classes, always checking the sources
afterwards”.
When it comes to audiovisual resources, men tend
to view GenAI tools as part of a broader technological landscape: “I use
DALL·E, Stable Diffusion, Runway Gen-2…”, whereas women associate them more
with specific educational objectives: “I use Canva with AI to create more
visually engaging presentations for my students”.
In the ‘no use’ or ‘early-stage use’ category
among women, explicit references to a lack of training or confidence are more
evident: “I don’t use AI tools because I lack training […] because I don’t feel
sufficiently trained”. Among men, this is usually expressed in more neutral or
circumstantial terms: “I’ve carried out one-off trials, but I haven’t
integrated them systematically”.
3.3.2. Types of GenAI tools
A breakdown by type of tool reveals more marked
differences than those observed in the overall classification of uses (Table
5). The clearly dominant tool in both groups is ChatGPT (41.74% of men and
44.53% of women), confirming its central role in the teaching ecosystem.
However, significant differences emerge in
secondary tools. Canva AI is more prevalent among women (15.33%) than men
(4.96%), suggesting a greater focus on tools integrated into the design of
teaching materials and visual presentations. In contrast, tools such as Claude
(9.50% among men compared to 4.01% among women), Midjourney (6.20% compared to
1.82%) or Runway (3.31% compared to 0%) carry greater relative weight in male
discourse, indicating a more diversified exploration of the advanced generative
environment.
Table 5
Thematic
distribution by gender in relation to GenAI tools
|
Categories |
Men n |
% Men |
Women n |
% Women |
|
ChatGPT |
101 |
41,74 % |
122 |
44,53 % |
|
Canva IA |
12 |
4,96 % |
42 |
15,33 % |
|
Claude |
23 |
9,50 % |
11 |
4,01 % |
|
Copilot |
15 |
6,20 % |
14 |
5,11 % |
|
Gemini |
16 |
6,61 % |
14 |
5,11 % |
|
DALL·E |
18 |
7,44 % |
28 |
10,22 % |
|
Midjourney |
15 |
6,20 % |
5 |
1,82 % |
|
Perplexity |
11 |
4,55 % |
20 |
7,30 % |
|
Stable Diffusion |
8 |
3,31 % |
10 |
3,65 % |
|
Synthesia |
15 |
6,20 % |
8 |
2,92 % |
|
Runway |
8 |
3,31 % |
0 |
0,00 % |
Note. Multiple coding. Coded
segments: men: 242; women: 274.
The chi-square test confirms a statistically
significant association between gender and the type of tool mentioned (χ²
(10, N = 516) = 41.36, p < .001), with a small to medium effect size (V =
0.283). This result suggests that, although the general pattern of use does not
differ significantly by gender, there are differences in the choice and
diversification of specific tools. The standardised residual analysis showed that
the largest contributions to the χ² value come from Canva AI, followed by
Runway, Midjourney and Claude.
4. Discussion and conclusions
The results show that university lecturers
perceive AI as a technology that supports teaching practice. Both men and women
recognise its ability to personalise learning (Vorobyeva
et al., 2025) and to automate tasks related to teaching, learning and
assessment, always in accordance with the lecturer’s professional judgement (Bond
et al., 2024). Both men and women agree on its potential to drive
methodological changes, through its application in the design of activities and
in providing immediate feedback (Liu et al., 2025). In assessment, the use of
GenAI in the production of evidence of learning requires a shift from
traditional approaches towards more process-oriented and competence-based
models (Yan et al., 2025). Both groups agree on the importance of personalisation
and maintaining the teacher’s role in assessment processes, reinforcing the
idea that AI acts as a support under professional supervision (Liu et al.,
2025; Yan et al., 2025). University lecturers report a wide and varied use of
these tools, highlighting content generation, information synthesis and
resource creation (Sotelo, 2025; Sozon et al. 2025).
AI-assisted teaching is perceived as a tool with transformative potential,
although this is contingent upon teacher training to ensure the pedagogical and
ethical integration of this technology (Gallent-Torres
et al., 2023).
With regard to
gender differences, in line with other studies (Aksongur
& Bağrıacık, 2025; Blázquez et
al., 2025), there are subtle differences in the way the integration of GenAI
into teaching practice is conceptualised and justified.
In teaching and learning processes, women’s
discourse focuses on pedagogical and educational integration and on enhancing learning.
Female teachers emphasise the integration of GenAI through didactic reflection
and student support to ensure responsible use that preserves critical thinking
(Møgelvang et al., 2025). Men’s discourse pays greater attention to technical
and practical aspects, emphasising time optimisation, task automation and
improved efficiency in teaching planning and management (Mohammadi et al.,
2026). These findings also align with those of Bond et al. (2024), who found
that men maintain an instrumental and technological approach, based on
efficiency and the optimisation of teaching activities. For their part, women
focus more on a transformative approach centred on teacher training and
supporting students in their teaching-learning process. According to Møgelvang
et al. (2025), these gender differences in technology integration do not stem
from differing abilities, but from different ways of appropriating and valuing technology,
which are conditioned by social and professional factors.
About assessment, the discursive nuances
identified are consistent with the findings of Møgelvang et al. (2024) and
Mohammadi et al. (2026), who highlight differing approaches to the pedagogical
adoption of GenAI. However, women more frequently emphasise feedback and the
formative dimension of assessment, whilst men tend to focus on the structuring
of the assessment model, the assessment of competences, and the mechanisms for controlling
the process (Blázquez et al., 2025). These nuances are consistent with studies
indicating that female teachers tend to adopt more pedagogical and reflective
approaches, whilst male teachers more frequently highlight the technical and
organisational dimensions of integrating AI into educational practice (Miranda
& Chamorro-Mera, 2025; Mohammadi et al., 2026).
As regards tools, ChatGPT is the one most
widely used by teachers. In this regard, other studies indicate that this application
is the most widely used in the educational sector, compared to similar ones
such as Gemini, Claude or Copilot (García-Peñalvo et
al., 2024). Among men, the discourse centres on technological exploration and
the diversification of tools, with an emphasis on optimising research
processes. In line with these findings, the study by Mohammadi et al. (2026)
shows that male teachers use GenAI tools more frequently than women and report more
systematic use across different tasks, particularly those linked to research
productivity. For their part, women make more didactic use of the technology,
focusing on the creation of materials and feedback, accompanied by greater
concern for the critical and ethical use of this technology. Similar results
are presented in the study by Blázquez et al. (2025), which identifies that
female teachers express greater caution, ethical concern and critical reflection
regarding the integration of GenAI into teaching practice.
The findings highlight the need for higher
education institutions to develop teacher training programmes focused on the
pedagogical, critical and ethical integration of GenAI. In this regard,
incorporating a gender perspective involves designing training initiatives that
consider different ways of adopting and using the technology, promoting both
the development of technological skills and pedagogical reflection on its application
in teaching, learning and assessment processes. Furthermore, it is necessary
for university and educational policies to promote frameworks for the
responsible use of GenAI, reduce potential inequalities in access to and
training in technology, and foster the inclusive integration of these tools
into teaching practice.
This study has certain limitations that should
be borne in mind when interpreting the results. Firstly, the sample is confined
to the university setting, which limits the generalisability of the findings to
other educational levels or institutional contexts. Furthermore, although the
comparative analysis by gender enabled the identification of relevant
discursive nuances, the study did not consider other personal, professional or
contextual variables that might influence the perception and use of GenAI.
As for future lines of research, it would be
worthwhile to examine in greater depth how variables such as teaching experience,
subject area or level of digital competence influence the integration of GenAI
into educational practice. It would also be relevant to conduct longitudinal
studies to analyse how these perceptions and uses evolve as AI tools become
more established in the university context. Finally, future research could
explore the impact of gender-sensitive teacher training programmes aimed at
promoting a pedagogical, critical and inclusive integration of GenAI.
Contribution of authors
Conceptualisation, Author 1; data curation,
formal analysis, Author 1 and Author 2; securing funding, Author 2; research,
Author 1, Author 2, Author 3; Author 1; methodology, Author 1, Author 2;
project management, software, Author 2; supervision, Author 1, Author 2 and
Author 3; writing and preparation of the original draft: Author 1, Author 2,
Author 3; review and editing, Author 1, Author 2 and Author 3.
Funding
This study has been funded by MICIU/AEI /10.13039
/501100011033 (State Research Agency, Ministry of Science, Innovation and
Universities) and ERDF (European Regional Development Fund, European Union),
grant number PID2021-122206NB-I00.
Data availability
The data supporting this study are available
from the corresponding author upon reasonable request.
Ethics declarations
The study was conducted in accordance with the
Declaration of Helsinki, approved by the Research Ethics Committee of the University
of Seville (protocol code PEIBA 2083-N-23; date of approval: 05/03/2024).
Conflicts of interest
The authors declare that there is no financial,
personal, or other conflict of interest that could have influenced the results
of the study.
References
Aksongur, M., & Bağrıacık
Yılmaz, A., (2025). Generative artificial intelligence in education:
teachers’ attitudes and influencing variables. Journal of Information and
Communication Technologies, 7(2), 98-111. https://doi.org/10.53694/bited.1792948
Alissa, R.A., & Hamadneh,
M. (2023). The level of
science and mathematics teachers’ employment of artificial intelligence
applications in the educational process. International Journal of Education
in Mathematics, Science, and Technology (IJEMST), 11(6), 1597- 1608. https://doi.org/10.46328/ijemst.3806
Bannister,
P., Santamaría Urbieta, A., & Alcalde Peñalver, E.
(2023). Una revisión sistemática de la IA generativa y la educación superior
(en inglés como medio de instrucción). Aula Abierta, 52(4),
401–409. https://doi.org/10.17811/rifie.52.4.2023.401-409
Blázquez
Rodríguez M., Vartabedian J., Mancha Cáceres O. I. & Pichardo Galán J. I.
(2025). Percepciones y usos de la inteligencia artificial en el profesorado
universitario: un análisis desde la perspectiva de género. Investigaciones Feministas, 16(1), 71-91. https://doi.org/10.5209/infe.100625
Bond, M., Bedenlier,
S., Marín, V. I., & Händel, M. (2024). A meta systematic review of
artificial intelligence in higher education: A call for increased ethics,
collaboration, and rigour. International Journal of Educational Technology
in Higher Education, 21(4), 1-41. https://doi.org/10.1186/s41239-023-00436-z
Braun, V., & Clarke, V. (2006). Using
thematic analysis in psychology. Qualitative Research in Psychology, 3(2),
77–101. https://doi.org/10.1191/1478088706qp063oa
Braun, V., & Clarke, V. (2021). Thematic
analysis: A practical guide. SAGE Publications.
Creswell, J.W. & Plano Clark, V.L. (2011) Designing
and Conducting Mixed Methods Research. 2nd Edition, Sage Publications.
Comisión Europea. (2021). Brújula Digital 2030: el enfoque de Europa
para la Década Digital. https://digital-strategy.ec.europa.eu/
Cooper, G. (2023). Examining Science Education
in ChatGPT: An Exploratory Study of Generative Artificial Intelligence. Journal
of Science Education and Technology, 32, 444-452. https://doi.org/10.1007/s10956-023-10039-y
Cotton, D. R. E., Cotton, P. A., & Shipway,
J. R. (2023). Chatting and cheating: Ensuring academic integrity in the era of
ChatGPT. Innovations in Education and Teaching International, 61(2),
228-239. https://doi.org/10.1080/14703297.2023.2190148
Criado Pérez, C. (2019). Invisible Women.
Exposing Data Bias in a World Designed for Men. Abrams Press.
Díaz-Vera, J. P.,
Molina, R., Bayas, C. M. & Ruiz-Ramírez, A. K.
(2024). Asistencia de la inteligencia artificial generativa como herramienta
pedagógica en la educación superior. RITI Journal,
12(26), 61-76. https://doi.org/10.36825/RITI.12.26.006
Gallent-Torres, C., Zapata-González, A., & Ortego-Hernando,
J.L. (2023). El impacto de la inteligencia artificial generativa en educación
superior: una mirada desde la ética y la integridad académica. RELIEVE,
29(2), art.
M5. http://doi.org/10.30827/relieve.v29i2.29134
García Peñalvo, F. J., Llorens-Largo, F., &
Vidal, J. (2024). The new reality of education in the face of advances in
generative artificial intelligence. RIED-Revista
Iberoamericana de Educación a Distancia, 27(1),
9-39. 13 https://doi.org/10.5944/ried.27.1.37716
Holmes, W.,
Porayska-Pomsta, K., Holstein, K. et al. (2022).
Ethics of AI in Education: Towards a Community-Wide Framework. International
Journal of Artificial Intelligence in Education, 32, 504–526. https://doi.org/10.1007/s40593-021-00239-1
Kim, J.,
Klopfer, M., Grohs, J.R., Eldardiry, H., Weichert,
J., Cox, L. A. & Pike, D. (2025). Examining Faculty and Student
Perceptions of Generative AI in University Courses. Innovative Higher
Education, 50, 1281–1313. https://doi.org/10.1007/s10755-024-09774-w
Liu, Q., Hu, A., Gladman, T. & Gallagher, S.
(2025). Eight Months into Reality: A Scoping Review of the Application of
ChatGPT in Higher Education Teaching and Learning. Innovative Higher
Education, 50, 1677-1700. https://doi.org/10.1007/s10755-025-09790-4
Luckin, R., Cukurova,
M., Kent, C., & du Boulay, B. (2022). Empowering
educators to be AI-ready. Computers
and Education: Artificial Intelligence, 3, 100076. https://doi.org/10.1016/j.caeai.2022.100076
Melchior, C., & Farinosi,
M. (2026). From Support to Dependency: Exploring Student Perceptions of
Generative AI. Comunicar, 84(34), 215-227. https://doi.org/10.5281/zenodo.18115566
Miranda, F. J., &
Chamorro-Mera, A. (2025). The impact of gender and age
on hei teachers' intentions to use generative
artificial intelligence tools. Information Technologies and Learning Tools,
108(4), 112-128. https://doi.org/10.33407/itlt.v108i4.6046
Møgelvang, A., Bjelland, C., Grassini, S., & Ludvigsen, K. (2024) Gender
Differences in the Use of Generative Artificial Intelligence Chatbots in Higher
Education: Characteristics and Consequences. Education Sciences, 14(12),
1363. https://doi.org/10.3390/educsci14121363
Mohammadi, E., Thelwall, M., Cai, Y., Collier,
T., Tahamtan, I., & Eftekhar, A. (2026). Is generative AI reshaping academic practices worldwide? A
survey of adoption, benefits, and concerns. Information Processing &
Management, 63(1), 104350. https://doi.org/10.1016/j.ipm.2025.104350
Rebollo-Catalán, M. A., García-Pérez, R.,
Barragán-Sánchez, R., Buzón-García, O., &
Ruiz-Pinto, E. (2012). Tecnologías para la
coeducación y la igualdad: valoración del profesorado de una herramienta web. Educación XX1,
15(1), 87–111.
Saldaña, J. (2021). The
coding manual for qualitative researchers (4th ed.). Sage.
Sanders, J. (2006). Gender and technology: What
the research tells us. In C. Skelton, B. Francis & L. Smulyan (Eds.). Handbook
of gender in education (pp. 307-322). Sage.
Sánchez, M. M. (2023). La inteligencia
artificial como recurso docente: usos y posibilidades para el profesorado. Educar,
60(1) 33-47. https://doi.org/10.5565/rev/educar.1810
Sotelo, J.A. (2025). Utilización de la
Inteligencia Artificial en educación superior universitaria en España. Comunicar, 82,
152-166. https://doi.org/10.5281/zenodo.16126055
Sozon,
Md., Parnther, C., Wei, W., & Chowdhury, M. A.
(2025). Generative AI in higher education: navigating benefits and challenges
in the technological era. Journal of Applied Research in Higher Education,
1-16. https://doi.org/10.1108/JARHE-02-2025-0103
Vorobyeva, K., et al. (2025). Personalized
learning through AI: Pedagogical approaches and critical insights. Contemporary
Educational Technology, 17(2), ep574. https://doi.org/10.30935/cedtech/16108
Wajcman,
J. (2006). El tecnofeminismo. Ediciones
Cátedra.
Yan, J., Tian, H., Sun, X., & Song, L.
(2025) Role of artificial intelligence in enhancing competency assessment and transforming
curriculum in higher vocational education. Frontiers in Education, 10,
1551596. https://doi.org/10.3389/feduc.2025.1551596