Out-of-the-Box GPTs for lexicography
| dc.contributor.author | De Schryver, Gilles-Maurice | |
| dc.date.accessioned | 2026-02-18T05:43:44Z | |
| dc.date.available | 2026-02-18T05:43:44Z | |
| dc.date.issued | 2025 | |
| dc.description.abstract | ABSTRACT : A detailed analysis is offered of how ChatGPT-3.5 has treated the world's languages for lexicographic purposes, starting with hyper-visible English, moving to semi-visible Portuguese, and ending with low-visible Bantu languages. Case studies for the latter group are region-visible Swahili, nation-visible Xhosa and local-visible Lusoga. This impact-based scale is reflected in the uptake of GenAI technology for lexicography. The results of using out-of-the-box ChatGPT-3.5 are fantastic for English, fake for Portuguese, and simply fail for the Bantu languages. But there is a way out, and the seeds for that are sown in this article. Keywords: generative artificial intelligence (GenAI), large language models (LLMs), generative pre-trained transformer (GPT), out-of-the-box GPTs, lexicography, dictionaries, lexicographers, English, Portuguese, Bantu, Swahili, Xhosa, Lusoga | |
| dc.description.abstract | OBUFUNZE : Okutangaazibwa ku ngeli ChatGPT-3.5 bw'ekozeseibwa mu nnimi dh'ensi yoonayoona mu kugezaaku okutuukiliza ebigendelelwa by'okuwandiika amawanika, kuweebwa mu lupapula luno. Okutangaaza kuno kutandiikila ku nkozesa eyolekebwa ku lulimi oluganzi mu nsi yoonayoona olw'Olungeleza, kw'amala kwaizibwa ku nnimi engazimuuku akatono ng'Olupootigo, kw'asembyayo enkozesa eyeeyolekela mu nnimi edhikozesebwa abantu abatono einho ng'edho edhigwa mu luse lwa Bantu. Ennimi dha Bantu edhiliku enkozesa entono einho edhigelaagela-nhizibwa ku y'edho emba oba eŋŋanziku n'omuli, olulimi lw'Oluswayili, Oluxhosa n'Olusoga. Ennimi dhino dhonsatule n'edhisinziilwaku okulaga tekinologiya wa GenAI bwe yeeyolekelamu ng'akozese-bwa mu milimu gy'okuwandiika amawanika. Kizuuse nti enkozesa ya tekinologiya eno nnambulukufu inho mu milimu egikola ku mawanika g'Olungeleza, aye, ya kibogwe ku nnimi eŋŋanziku akatono ng'Olupootigo. Bwe kituuka ku nnimi dh'omu luse lwa Bantu kabili enkola eno eililila inho y'aba oti kumpi tekolela ilala. Aye, wakaaliwooku eisuubi eliva mu byandhulwa mu kitundu ky'olupapula luno eky'okubili. Mu lupapula olw'okubili n'omuli ebyava mu kwongela okunoonheleza ku ngeli eyinza okukozesebwa okuziba olutanda oluli wagati w'ennimi dh'ensi yoonayoona eŋŋanzi, n'edhitali, aye nga dhoonadhoona dhiluubilila okukoseza GenAI okuwandiika amawanika. | |
| dc.description.department | African Languages | |
| dc.description.librarian | ay2026 | |
| dc.description.sdg | SDG-04: Quality education | |
| dc.description.sdg | SDG-09: Industry, innovation and infrastructure | |
| dc.description.uri | https://journals.co.za/journal/lexikos | |
| dc.identifier.citation | De Schryver, G.-M. 2025, 'Out-of-the-Box GPTs for lexicography', Lexikos, vol. 35, no. 2, pp. 362-396. https://doi.org/10.5788/35-2-2103. | |
| dc.identifier.issn | 684-4904 (pint) | |
| dc.identifier.issn | 2224-0039 (online) | |
| dc.identifier.other | 10.5788/35-2-2103 | |
| dc.identifier.uri | http://hdl.handle.net/2263/108351 | |
| dc.language.iso | en | |
| dc.publisher | Bureau of the WAT | |
| dc.rights | © 2025 Bureau of the WAT. | |
| dc.subject | Generative artificial intelligence (GENAI) | |
| dc.subject | Large language models (LLMs) | |
| dc.subject | Generative pre-trained transformer (GPT) | |
| dc.subject | Out-of-the-box GPTs | |
| dc.subject | Lexicography | |
| dc.subject | Dictionaries | |
| dc.subject | Lexicographers | |
| dc.subject | English | |
| dc.subject | Portuguese | |
| dc.subject | Bantu | |
| dc.subject | Swahili | |
| dc.subject | Xhosa | |
| dc.subject | Lusoga | |
| dc.title | Out-of-the-Box GPTs for lexicography | |
| dc.title.alternative | GPT ey'omutindo eya GenAI mw'isomo ly'okuwandiika ama-wanika | |
| dc.type | Article |
