Abstract:

The use of Generative Artificial Intelligence (GenAI) in educational contexts has increasingly established itself as a promising tool for supporting pedagogical differentiation, enabling teachers to access suggestions for adapting content to students’ profiles and needs. However, despite this potential, little is known about how GenAI can be concretely used to adapt teaching materials for students receiving selective support measures, in line with the principles of Decree-Law No. 54/2018. It is within this context that the present study examines the potential of GenAI in the development of personalised instructional materials for these students.

The research was conducted with three students benefiting from selective support measures, integrated into an 8th-grade class, and was guided by the research question: “How can GenAI be used to adapt teaching materials for students with selective support measures in Portuguese language education?”. To address this question, four objectives were established:
i) to analyse the clarity, usefulness, and pedagogical suitability of GenAI-generated suggestions;
ii) to evaluate the effectiveness of the adapted materials based on their use in the classroom and observation of the students;
iii) to compare materials produced with GenAI support with traditional materials; and
iv) to examine the broader potential of GenAI in adapting pedagogical resources for students with selective support measures.

A qualitative approach was adopted, grounded in an interpretative case study design, drawing on participant observation, documentary analysis of the instructional materials produced, and content analysis of the responses generated by ChatGPT.

The findings provide evidence, in a real classroom context, of GenAI’s contribution to pedagogical differentiation, through the production of more accessible and linguistically adjusted materials tailored to these students’ needs. Furthermore, the results reinforce the central role of the teacher in interpreting, validating, and pedagogically contextualising GenAI-generated suggestions, emphasising that GenAI should be understood as a support tool for the adaptation process rather than a replacement for professional decision-making.

These conclusions imply:
a) evidence of GenAI’s potential as a complementary resource for inclusive practice;
b) the need for teacher training that integrates technological competence with pedagogical knowledge; and
c) the opening of new avenues for research into AI-mediated adaptation processes in language education, including long-term effects on students’ performance and development.

Keywords:

Generative Artificial Intelligence, Materials Adaptation, Inclusive Education, Portuguese Language Teaching.