Academic Performance and Inequality: Can Generative AI Widen the Digital Divide in Higher Education?
Journal
EAI/Springer Innovations in Communication and Computing
Emerging Technologies in Applied Engineering and Education
ISSN
2522-8595
2522-8609
Date Issued
2026
Type
Resource Types::text::conference output::conference proceedings::conference paper
Abstract
Higher educationHigher education has undergone a rapid transformation driven by digitalization and the rise of generative artificial intelligenceGenerative artificial intelligence tools, such as Chat Generative Pre-trained Transformer and Copilot. While these technologies offer significant opportunities to support learning processes, they also risk deepening existing inequalities if not implemented under equitable conditions. Many students, particularly those in rural areas or with limited resources, continue to face barriers related to device availability, Internet connectivity, and digital training, which restricts the effective use of generative AIArtificial intelligence and negatively affects academic performance. To analyze this gap and identify possible solutions, a critical narrative review of eighteen academic and technical sources published between 2022 and 2024 was conducted. The review focused on five key areas: access, digital literacy, academic impact, structural inequality, and inclusive strategies. The findings reveal sharp disparities: only 27% of rural students have access to devices compatible with generative AI, compared to 70% in urban settings, and students with strong digital competencies obtain up to 60% greater academic benefit from these tools. These results underscore the urgent need for targeted interventions to avoid exacerbating the digital divide. In response, the study proposes five strategies: ensuring equitable institutional access, strengthening critical digital literacyCritical digital literacy, adopting hybrid pedagogical models, designing inclusive curricula, and promoting dedicated research lines on public policy. The conclusion is that although generative artificial intelligenceGenerative artificial intelligence holds considerable educational potential, its benefits will only be realized if embedded in inclusive ecosystems that guarantee equal conditions, teacher support, and the ethical development of learning. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.