AI-SUPPORTED DIGITAL INCLUSION FOR OLDER ADULTS: DESIGNING AND EVALUATING TAILORED LEARNING MATERIALS AND A RAG-BASED EDUCATIONAL ASSISTANT (GRANS ACTIUS PROJECT)
Universitat de les Illes Balears (SPAIN)
About this paper:
Conference name: 18th International Conference on Education and New Learning Technologies
Dates: 29 June-1 July, 2026
Location: Palma, Spain
Abstract:
Digital exclusion remains a significant barrier for adults aged 55 and over, limiting their access to essential services, lifelong learning opportunities, and social participation. Addressing this challenge requires not only accessible training programmes but also pedagogically adapted materials and continuous support mechanisms aligned with the needs of older learners.
This study presents the design, implementation, and data-driven evaluation of an innovative digital inclusion initiative developed within the frameworks of the “Grans Actius” and “Grans Proactius” projects (https://gransactius.org/es).
The initiative combines tailored educational materials specifically designed for adults aged 55+ with the integration of an AI-based educational assistant built on a Retrieval-Augmented Generation (RAG) model. This assistant enables learners to resolve doubts in real time through natural language interaction, enhancing autonomy, confidence, and continuity in the learning process.
The pedagogical design is grounded in accessibility, simplicity, and user-centred learning principles, ensuring alignment with the cognitive and motivational characteristics of older adults. In addition to the instructional materials, usage data and interaction metrics from the AI assistant have been analysed to assess engagement, patterns of use, and perceived usefulness.
Preliminary results indicate high levels of engagement and satisfaction, with learners particularly valuing the clarity of the materials and the immediacy, relevance, and personalization of the AI-generated responses. The availability of on-demand support through the RAG-based assistant significantly reduces dependency on instructors and contributes to sustained learning beyond formal training sessions.
The findings highlight the potential of combining tailored instructional design with AI-supported learning tools to foster digital competence and promote lifelong learning among older adults. This work proposes a scalable and transferable model for inclusive digital education, contributing to ongoing efforts to bridge the digital divide in ageing societies. This project is supported by the Government of the Balearic Islands and funded by the European Union – NextGenerationEU programme.Keywords:
Digital inclusion, older adults, lifelong learning, AI-supported learning, RAG-based systems, digital literacy.