Innovation in Language Learning

Generative artificial intelligence is beginning to reshape needs analysis, the methodological cornerstone of Language for Specific Purposes (LSP) course design, offering new capacity to synthesise occupational discourse, simulate stakeholder input and generate discipline-specific task inventories at a scale traditional target-situation surveys and interviews cannot match. This paper presents a systematic literature review, conducted in accordance with the PRISMA protocol, of studies published between 2020 and 2026 that examine the application of large language models to needs analysis and course design in LSP and English for Specific Purposes contexts. A structured search of major bibliographic databases identified a corpus of studies screened against predefined inclusion and exclusion criteria and appraised for methodological transparency and empirical grounding. Thematic synthesis of the included studies identifies three emerging applications of generative AI within the needs analysis process: automated analysis of authentic occupational and academic discourse, AI-simulated stakeholder consultation, and generation of draft syllabi and task banks subsequently validated by human experts. The review also identifies recurring concerns regarding data currency, disciplinary accuracy, and the continued necessity of expert and stakeholder validation of AI-generated outputs. Drawing on these findings, the paper proposes a revised, hybrid needs analysis model that repositions generative AI as a preparatory tool within, rather than a replacement for, established LSP course design methodology, offering course designers a methodologically sound and immediately applicable framework, and outlining priority directions for future research and practice. 

 

Keywords: generative AI; needs analysis; Language for Specific Purposes; large language models; course design; systematic review

 

References

 

[1] Yan, H. (2025). Trends in empirical research in English for Specific Purposes: A systematic review of SSCI-indexed journal articles (2014-2023). SAGE Open, 15(1). https://doi.org/10.1177/21582440251328460

[2] Shi, Y., Yu, K., Dong, Y., & Chen, F. (2025). Large language models in education: A systematic review of empirical applications, benefits, and challenges. Computers and Education: Artificial Intelligence, 10, Article 100529. https://doi.org/10.1016/j.caeai.2025.100529

[3] Warden, C. A., Stanworth, J. O., & Chen, J. F. (2025). Generative AI and the emergence of hybrid authorship in ESP writing: A corpus-based analysis (2015-2024). Journal of English for Academic Purposes, 78, Article 101578. https://doi.org/10.1016/j.jeap.2025.101578