Local contexts in artificial intelligence adaptation: human and machine understanding of Kazakhstani news

Section: New Media

The ability of generative language models to correctly adapt texts and recognize deep cultural and historical local contexts remains poorly studied. Despite the high quality of generating and translating texts, and analytical outputs, the models still demonstrate limited capabilities for interpreting socially significant meanings and values. These limits are clearly visible in current news items, where the model can produce distorted or downright erroneous interpretations. The purpose of this study is to diagnose the practical applicability of the model in adapting Kazakhstani news for an English-speaking audience unfamiliar with local realities, as well as to classify errors and inaccuracies. During the two-level analysis, the generated ChatGPT texts were compared with experts’ interpretations. In addition to assessing the adapted surface meanings, the tasks included a two-stage analysis of the current deep contexts of Kazakhstani public discourse. Based on the results of comparative analysis, the ChatGPT output data revealed a number of persistent errors, classified into a typology of interpretative distortion vectors. At the same time, authors have recorded unique interpretations, which allows us to assume the prospect of practical interaction with generative models as a tool for an alternative view of phenomena or a source of non-obvious ideas.

Abisheva C., Koldasbaeva Z., Nossiyeva N., Irgebaeva N., Aipova А., Doldinova S., Smoilov S., Aykenova R., Umirzakova L., Idrissova M. (2024) Formation of Ethical Competences for AI Use in English Foreign Language Teachings. Qubahan Academic Journal 4 (4): 191–205. DOI: 10.48161/qaj.v4n4a1256

Abylkhozhin Zh. B., Krupko I. V. (2024) Ideologicheskie diskussii i poisk istoricheskoy subyektnosti kazakhstanskogo obshchestva v XX veke [Ideological discussions and the search for historical subjectivity of Kazakhstani society in the twentieth century]. Novye issledovaniya Tuvy 3: 265–278. DOI: 10.25178/nit.2024.3.15. (In Russian)

Amanzholova D. B., Iskendir A. A., Shaymerdenova N. Zh. (2024) Kommunikativnyy i lingvodidakticheskiy potentsial epistolyarnykh tekstov: prikladnoy aspekt (na materiale pochtovoy korrespondentsii D.A. Mongusha – A. S. Amanzholovu) [Communicative and linguodidactic potential of epistolary texts: applied aspect (based on postal correspondence between D.A. Mongush and A.S. Amanzholov)]. Novye issledovaniya Tuvy 3: 240–264. DOI: 10.25178/nit.2024.3.14. (In Russian) 

Broussard M., Diakopoulos N., Guzman A. L., Abebe R., Dupagne M., Chuan C.-H. (2019) Artificial Intelligence and Journalism. Journalism & Mass Communication Quarterly 96 (3): 673–695. DOI: 10.1177/1077699019859901 

Chang C. T., Farah H., Gui H., Rezaei Sh. J. et al. (2025) Red teaming ChatGPT in medicine to yield real-world insights on model behavior. Digital Medicine 8: 149. DOI: 10.1038/s41746-025-01542-0

Davydov S. G., Zamkov A. V., Krasheninnikova M. A., Lukina M. M. (2023) Ispol'zovanie tekhnologiĭ iskusstvennogo intellekta v rossiĭskikh media i zhurnalistike [Use of artificial intelligence technologies in Russian media and journalism]. Vestnik Moskovskogo universiteta. Seriya 10: Zhurnalistika 5: 3–21. DOI: 10.30547/vestnik.journ.5.2023.321. (In Russian) 

Dergaa I., Chamari K., Zmijewski P., Ben Saad H. (2023) From human writing to artificial intelligence generated text: examining the prospects and potential threats of ChatGPT in academic writing. Biology of Sport 40 (2): 615–622. DOI: 10.5114/biolsport.2023.125623

Duan J. (2024) Identification and influence of tourism consumption behavior based on artificial intelligence. Informatica (Ljubljana) 48 (15): 135–150. DOI: 10.31449/inf.v48i15.6203

Dugin E. Y. (2024) Transformatsiya mediakommunikatsii pod vozdeĭstviem tsifrovykh tekhnologii: teoretiko-metodologicheskii aspect [Transformation of media communication under the influence of digital technologies: theoretical and methodological aspect]. Vestnik Moskovskogo universiteta. Seriya 10. Zhurnalistika 5: 140–151. DOI: 10.55959/msu.vestnik.journ.5.2024.140151 (In Russian) 

El-Tigani A. (2024) The Reality of Using ChatGPT in Media Studies for Students, Scholars, and Practitioners. Herald of Journalism 74 (4): 4–18. DOI: 10.26577/HJ.2024.v74.i4.1

Hassouni A., Mellor N. (2025) AI in the United Arab Emirates’ Media Sector: Balancing Efficiency and Cultural Integrity. Journalism and Media 6 (1). DOI: 10.3390/journalmedia6010031

Izadi S., Forouzanfar M. (2024) Error Correction and Adaptation in Conversational AI: A Review of Techniques and Applications in Chatbots. AI 5: 803–841. DOI: 10.3390/ai5020041

Kazun A. D. (2024) Perspektivnye napravleniya issledovaniĭ potrebleniya novosteĭ v kontekste mediasredy s bol'shim vyborom [Promising Directions for Research on News Consumption in the Context of a Media Environment with a Large Choice]. Vestnik Moskovskogo universiteta. Seriya 10: Zhurnalistika 2: 3–47. DOI: 10.30547/vestnik.journ.2.2024.347. (In Russian)

Kalla D., Smith N., Samaah F., Kuraku S. (2023) Study and Analysis of Chat GPT and its Impact on Different Fields of Study. International Journal of Innovative Science and Research Technology 8 (3). 

Lee M. A. (2023) Mathematical Interpretation of Autoregressive Generative Pre-Trained Transformer and Self-Supervised Learning. Mathematics 11 (11): 2451. DOI: 10.3390/math11112451

Mambetova G. T., Shingareva M. Y., Dmitryuk N. V., Stycheva O. A. (2023) Sopostavitel'nyy analiz frazeologicheskikh edinits s somatizmom ruka kak sposob formirovaniya lingvokul'turnoy kompetentsii (na materiale russkogo, kazakhskogo, tuvinskogo i angliyskogo yazykov) [Comparative analysis of phraseological units with the somatism hand as a way of forming linguocultural competence (based on the Russian, Kazakh, Tuvan and English languages)]. Novye issledovaniya Tuvy 4: 109–123. DOI: 10.25178/nit.2023.4.8. (In Russian) 

Novikova M. L., Novikov F. N. (2024) Ispolzovanie iskusstvennogo intellekta dlya sozdaniya sistemy mashinnogo perevoda i obrazovatelnykh resursov na tuvinskom yazyke [Using Artificial Intelligence to Create a Machine Translation System and Educational Resources in Tuvan]. Novye issledovaniya Tuvy 1: 6–17. DOI: 10.25178/nit.2024.1.1. (In Russian) 

Nurash A., Rozhkov A. (2024) Thematic programs (television) and periodicals on the culture of the period of independence of Kazakhstan (content, methods of their formation)]. Herald of Journalism 73 (3): 132–141. DOI: 10.26577/HJ.2024.v73.i3.11

Ogunde F. (2025) Legal large language models (LLMs): legal dynamos or “fancifully packaged ChatGPT”? Discover Artificial Intelligence 5 (21). DOI: 10.1007/s44163-024-00167-8

Oprea S. V., Bâra A. (2025) Interpreting text corpora from androids-related stories using large language models: “Machines like me” by Ian McEwan in generative AI. Humanities and Social Sciences Communication 12. DOI:10.1057/s41599-025-04633-1

Petruccio P., Neilson T., Stöcker C. (2025) ‘A Part of Our Work Disappeared’: AI Automated Publishing in Social Media Journalism. Journalism and Media 6 (1). DOI: 10.3390/journalmedia6010030

Upadhyaya N. (2024) AI-Augmented Dynamic Language Adaptation for Low-Resource Languages: A Transfer Learning Solution for Optimized NLP Performance. International Research Journal of Advanced Engineering and Science 9 (4): 89–94.

Zelianskaia N. L., Belousov K. I., Sychov O. A., Galinskaia T. N. (2023) Semantic structure of clickbait headlines and internet user activities. Media Linguistics 10 (1): 66–86. DOI: 10.21638/spbu22.2023.104

To cite this article: Akhmetov S. S., Akhmetova L. S. (2025) Lokal’nye konteksty v adaptatsii iskusstvennogo intellekta: chelovecheskoe i mashinnoe ponimanie kazakhstanskikh novostey [Local contexts in artificial intelligence adaptation: human and machine understanding of Kazakhstani news]. Vestnik Moskovskogo Universiteta. Seriya 10. Zhurnalistika 4: 24–49. DOI: 10.55959/msu.vestnik.journ.4.2025.2449