Corpus data in the age of generative artificial intelligence: continuity of digital methods in foreign language education
Year: 2026
Type: article
Abstract: Importance . Generative artificial intelligence in language education renews the question of sufficient grounds for language decisions because chat systems often do not disclose the origin of examples or the procedure behind conclusions. Research purpose – to test the hypothesis that corpus data retain methodological value by making language hypotheses open to observable and reproducible checks. Materials and Methods . A targeted integrative review covered 27 scholarly publications and policy documents: foundational and review studies of data-driven learning, research on generative artificial intelligence, Russian-language evidence, and early hybrid projects. Twenty-four sources were examined in full text; three were used at the abstract and metadata level. Results and Discussion. Corpus-based language pedagogy has a more mature evidence base for immediate vocabulary and lexicogrammar outcomes. Research on generative artificial intelligence remains limited by short interventions, university samples, writing tasks, and self-reports. The synthesis produced six stages: task definition, hypothesis generation, corpus search, interpretation, application, and transfer assessment. Conclusion . The hypothesis received partial support: the literature substantiates the checking function of corpora, but the proposed sequence has not been tested comparatively. The findings can inform the design and evaluation of foreign language courses that combine a generative system with corpus data.
Author: I. D. Kachurin
Institution: Lomonosov Moscow State University
Language: English
Subfield: Developmental and Educational Psychology
Field: Psychology
Domain: Social Sciences
SDG: Quality education
Open Access status: diamond
OpenAlex ID: W7223360438