Between Algorithm and Agency: Designing Inclusion With Universal Design for Learning in the Age of Generative Artificial Intelligence
Year: 2026
Type: article
Abstract: In the transition from digital competence to Digital Affective Literacy, instructional design emerges as a space for negotiation between the human and the artificial. This paper presents a quasi-experimental study with repeated measures and a crossover design involving 100 TFA special education trainees who were tasked with designing lessons based on Universal Design for Learning under three conditions: no AI, use of a generalist chatbot, and use of an MCP server designed according to inclusive principles. The lesson plans were evaluated by experts using a structured UDL rubric, while reflective logbooks were analysed qualitatively to investigate perceived agency, uncertainty, delegation and awareness during the design process. The results show that the UDL-MCP server supports usability and provides clearer guidance in translating theoretical knowledge of UDL into concrete inclusive design choices. The integration of assessment data and reflective narratives suggests that AI can act as a symbolic mediator when it preserves teacher agency and strengthens inclusive intentionality in hybrid educational contexts.
Institution: Pegaso University
Language: English
Subfield: Education
Field: Social Sciences
Domain: Social Sciences
SDG: Quality education
Open Access status: diamond
OpenAlex ID: W7225560367