The rise of Large Language Models (LLMs) has fundamentally altered knowledge production in higher education, with over 80% of students now integrating AI into academic tasks (Xia et al., 2024; Fundación Ayuda en Acción, 2025). This widespread adoption, often lacking critical algorithmic or ethical awareness, necessitates a total reappraisal of traditional asynchronous written assessments  (Vieru & Petrea, 2015; Francis et al. 2025). While some institutions have responded reactively through bans or proctored exams, this catalyst demands a holistic, integrative approach (Chan, 2023). Aligning with the European Higher Education Area’s emphasis on transferable competencies, AI serves as an urgent driver for redesigning pedagogical methodologies and fostering critical AI literacy within the university space.

This paper presents the design, theoretical foundations, and methodological approach of an ongoing research project. The study aims to analyze the uses and ethical perceptions of generative AI among Education students to evaluate its impact on assessment processes and propose pedagogical strategies that foster critical literacy and institutional debate. The study is part of a funded project (Rethinking the quality of training for Education students in the era of generative artificial intelligence - IA-EduProf) by the National Distance Education University (UNED, Spain) within the call for Teaching Innovation Projects for Teaching Innovation Groups (GID2016-47). It is explicitly situated within higher education, aligning with formative assessment frameworks and internal quality assurance. It seeks to articulate adapted responses capable of combining digital literacy and ethical reflection through curricular redesign and the continuous improvement of training programs for education professionals.

The research is grounded in theoretical frameworks such as Selwyn (2021) and Watters (2021), who advocate for a critical educational technology perspective over technological determinism, allowing AI to be analyzed not as a neutral process but as an opportunity to enhance student agency. The study of uses and ethics is framed by the guidelines of UNESCO (Giannini, 2023) and Tai et al. (2023) regarding algorithmic literacy and educational justice, as well as the theory of assessment as a social practice (Bearman et al., 2020), seeking tasks that transcend the mere generation of automated products. Finally, the project aligns with Luckin’s (2018) vision of adaptive human intelligence, placing critical thinking and pedagogical design at the center of the curricular transformation demanded by the new landscape of quality higher education.

Methodologically, the project adopts a mixed-methods research approach, following the principles of Creswell and Creswell (2018). This allows for the integration of the precision of quantitative usage patterns with the depth of qualitative ethical perceptions (Teddlie & Tashakkori, 2009). This design is based on the concept of methodological triangulation and the pursuit of complementarity (Greene, 2007), ensuring a holistic understanding of AI's impact.

It is expected that this research facilitates a reconfiguration of evaluative praxis, providing clear criteria for transitioning toward authentic assessment models with high resistance to algorithmic replication. Beyond AI detection, the anticipated impact lies in strengthening student commitment and agency by proposing tasks that connect with their future professional identity and reduce the automated nature of assessment. Ultimately, the integration of critical literacy into the curriculum will enable future educators to develop strategic autonomy and sound ethical judgment, consolidating a pedagogical redesign proposal that rigorously addresses the demands of new digital scenarios.

The project is grounded in an action research framework, conceived as a cyclical process of systematic inquiry that links the diagnosis of AI usage with intervention and the improvement of teaching practice (Górriz, 1997; Vidal & Rivera, 2007). To provide this process with a holistic understanding, a mixed-methods research approach is adopted (Creswell & Creswell, 2018). This design allows for the integration of the precision of quantitative usage patterns with the depth of qualitative ethical perceptions, based on the principles of methodological triangulation and complementarity (Greene, 2007; Teddlie & Tashakkori, 2009). In this way, the analysis of the phenomenon serves as a basis for reflection-action, facilitating a pedagogical redesign of Continuous Assessment Tasks (PECs) and the establishment of institutional guidelines that respond in a situated and rigorous manner to the challenges of artificial intelligence. In the first phase, a questionnaire is applied to measure the use of artificial intelligence in university students, designed and validated by Trejo-Trejo & Gordillo-Espinoza (2026). This instrument evaluates dimensions such as information search and management, tutoring and academic assistance, content creation and editing, perceived self-efficacy, ethical use, accessibility and equity, environmental impact, and dependency or addiction. Open-ended questions have been added at the end of the questionnaire to explore ethical and metacognitive dimensions, analyzing the boundary between algorithmic support and personal authorship, the strategic autonomy of students, and their resistance to automation, thereby enhancing critical judgment and professional identity. The instrument is administered to students in Pedagogy, Early Childhood Education, and Social Education degree programs. In the second phase, online focus groups are conducted with students and professionals from the involved areas. A semi-structured guide will be used to delve deeper into aspects related to the participants' responses regarding the aforementioned dimensions. In the third phase, activities and assessment tasks (PECs) are designed in accordance with the results obtained. Quantitative data analysis is performed using descriptive and inferential techniques, while discourse analysis of the focus group transcripts is conducted with the support of Atlas.ti software. Finally, the results will be triangulated to seek complementarity (Greene, 2007), ensuring a holistic understanding of the impact of AI.

Although the research is currently in progress, its results are expected to provide a substantial contribution to the improvement of quality in higher education. First, the findings aim to inform a profound transformation of assessment processes by identifying tasks vulnerable to automation. Under the premise of assessment as social practice by Bearman et al. (2020), the study seeks to provide more meaningful and authentic evaluative designs that, being resistant to algorithmic replication, ensure training aligned with real professional competencies. Second, the results are intended to favor an increase in student participation and motivation. By leading to tasks that, as Selwyn (2021) and Watters (2021) argue, enhance agency against technological determinism, the research provides solutions that connect the curriculum with the identity challenges of future educators. Third, the study provides key insights to foster autonomous and collaborative learning through critical AI literacy. By integrating Luckin’s (2018) perspective on adaptive human intelligence, the results reinforce students' analytical capacity and ethical judgment, placing pedagogical reflection at the core of curricular transformation. Finally, this research offers a strategic contribution to the international and European debate. By converging with the UNESCO guidelines mentioned by Giannini (2023) and the holistic approach of Chan (2023), the work provides a transferable model for teacher training. In this way, the results contribute to redefining quality and teacher autonomy within the European Higher Education Area, responding rigorously to the demands of the new digital landscape.

Ayuda en Acción. (2025). El impacto de la inteligencia artificial en la educación superior en España. Fundación Ayuda en Acción. Bearman, M., Dawson, P., Ajjawi, R., Tai, J., y Boud, D. (2020). Re-imagining assessment in a digital world: Formative assessment for learning. Springer. Chan, C. K. Y. (2023). A comprehensive framework for AI literacy. Higher Education Research & Development. https://doi.org/10.1080/07294360.2023.2173521 Creswell, J. W., y Creswell, J. D. (2018). Research design: Qualitative, quantitative, and mixed methods approaches (5.ª ed.). SAGE. Francis N., Jones, M. & Smith D. (2025). Generative AI in Higher Education: Balancing Innovation and Integrity. Br. J. Biomed. Sci., Volume 81. https://doi.org/10.3389/bjbs.2024.14048 Giannini, S. (2023). Guidance for generative AI in education and research. UNESCO. Górriz, A. B. (1997). La investigación-acción como estrategia de formación permanente. Revista de Educación. Greene, J. C. (2007). Mixed methods in social inquiry. Jossey-Bass. Luckin, R. (2018). Machine learning and human intelligence: The future of education in the 21st century. UCL Press. Selwyn, N. (2020). ¿Deberían los robots sustituir al profesorado? La IA y el futuro de la educación. Morata. Selwyn, N. (2021). Education and technology: Key issues and debates (3.ª ed.). Bloomsbury Academic. Tai, J., Ajjawi, R., Bearman, M., y Dawson, P. (2023). Algorithmic literacy and the future of evaluative judgement. Higher Education. Teddlie, C., y Tashakkori, A. (2009). Foundations of mixed methods research: Integrating quantitative and qualitative approaches in the social and behavioral sciences. SAGE. Trejo-Trejo, A., y Gordillo-Espinoza, R. (2026). Diseño y validación de una escala para medir el uso de la inteligencia artificial en estudiantes universitarios. RevistaPixel-Bit, 75. Art. 7. https://doi.org/10.12795/pixelbit.1 Vidal, M., y Rivera, N. (2007). Investigación-acción. Educación Médica Superior, 21(4). Vieru, D., y Petrea, E. (2015). Ethical implications of algorithmic processes in academic settings. Journal of Academic Ethics. Watters, A. (2021). Teaching machines: The history of personalized learning. MIT Press. Xia, Q., Weng, X., Ouyang, F., Lin, T. J., & Chiu, T. K. (2024). A scoping review on how generative artificial intelligence transforms assessment in higher education. International Journal of Educational Technology in Higher Education, 21(1), 40. 10.1186/s41239-024-00468-z