A study from the University of La Laguna proposes an innovative didactic model to facilitate the development of critical thinking among university students. The proposal aims to guide the use of generative artificial intelligence (AI) tools, such as ChatGPT, Gemini, or Perplexity, in learning tasks.
According to a report by the Foundation for Knowledge and Development (2025), approximately 90% of Spanish university students already use generative AI in their academic work. Educators are increasingly finding it difficult to distinguish between AI-generated content and that produced by students, and traditional plagiarism detection tools show limitations with these texts.
In this context, research published in the scientific journal Pixel-Bit. Revista de Medios y Educación (Q1 in Scopus) presents a didactic model designed to plan and evaluate learning tasks that promote critical thinking through generative AI.
The research, conducted by Manuel Area-Moreira and Miriam González-González from the EDULLAB group at the ULL, is structured around six dimensions: inquiry, comparison, critical dialogue, verification, personal re-elaboration, and reflection. The goal is to teach students how to formulate questions, cross-reference information from various sources, validate data, reinterpret content, and reflect on the capabilities and limitations of AI.
The experience, carried out at the ULL during the 2023-2024 academic year in the subject “Information and Communication Technologies in Education,” involved 75 students from the Bachelor's Degree in Primary Education Teaching. Participants were required to use three different AI systems to complete an academic assignment, compare the responses, verify their validity, and submit a critical report.
Students highlighted the speed of information access, time savings, the potential for personalized learning, and support for idea generation as positive aspects of using AI. However, they also pointed out risks such as increased plagiarism, the presence of errors or incorrect information, and concerns regarding privacy and data protection. They expressed apprehension about the potential loss of skills like creativity, autonomy, and critical thinking.
Area-Moreira and González-González conclude that the main challenge is pedagogical: to educate reflective and responsible citizens in a digital society. The model promotes a balanced perspective, acknowledging both the opportunities and risks of AI in higher education, and advocates for its responsible integration as a supportive educational tool.
The researchers acknowledge that the model requires broader empirical validation. The EDULLAB group is developing new projects to apply it across different degree programs and university contexts. They are currently participating in the DIDACT.IA research project, funded by the National R&D&I Plan, which studies the impact of AI in university education.
