Generative AI, Linguistic Hierarchies and Unequal Access in Under‐Resourced Educational Contexts
Abstract
Despite growing enthusiasm around generative AI (GenAI) as a transformative force in education, access to these technologies remains profoundly unequal. This qualitative study examines how educators and students in under‐resourced Chinese educational contexts experience unequal access to GenAI and interpret its consequences for language learning, teaching and local knowledge production. Informed by critical sociolinguistics and political economy, the study draws on ten semi‐structured interviews with six teachers/educators and four students from local universities, rural communities and linguistically minoritised settings in China. The analysis shows that GenAI access is rarely a binary matter of presence or absence, but a layered set of hurdles that participants had to navigate actively: tools are visible through media, training sessions and peer networks, but unstable connectivity, outdated devices, paywalls, institutional permissions and limited local integration meant that turning visibility into sustained use required ongoing effort, workaround strategies and situated forms of agency. Participants also reported English‐centred performance, weak support for local dialects and minority languages and culturally generic or stereotyped outputs. Although empirically situated in China, the study uses this contextually bounded dataset to contribute analytically to wider debates on global digital inequality, linguistic hierarchy and algorithmic colonialingualism.