Artificial Intelligence and Machine Translation: ChatGPT, Google Translate, DeepL, Human-AI Collaboration, Post-Editing, and Translation Quality
DOI:
https://doi.org/10.62497/irjai.275Keywords:
artificial intelligence, machine translation, ChatGPT, Google Translate, DeepL, large language models, post-editing, human–AI collaboration, translation qualityAbstract
Introduction: Artificial intelligence has transformed machine translation from rule-based and statistical systems to neural and large language model systems that can produce fluent multilingual text at unprecedented speed. ChatGPT, Google Translate, and DeepL represent different manifestations of this transformation, while human–AI workflows increasingly combine automated generation with expert revision.
Objective: This paper examines the development of AI-based machine translation and compares general-purpose large language models, translation-oriented platforms, human translation, and hybrid workflows. It focuses on post-editing, quality evaluation, hallucination, contextual accuracy, terminology, and the conditions under which AI output can be considered fit for purpose.
Method: An integrative literature review was conducted using foundational studies on neural machine translation and Transformer architectures, empirical studies of multilingual large language models, research on post-editing productivity, automatic and human evaluation, hallucination, and professional standards. Contemporary research from 2024–2025 was emphasized where available.
Results: The synthesis indicates that AI translation provides major gains in speed, scalability, consistency, and drafting capacity, but quality varies by language pair, domain, text type, and system. Large language models provide unusually strong instruction-following and interactive revision, while specialized translation systems remain valuable for high-volume and repeatable tasks. Human review is essential when semantic precision, terminology, cultural appropriateness, or accountability is important.
Conclusion: AI is best understood as a translation partner rather than a universal substitute for professional translators. The most reliable workflow uses AI for candidate generation, comparison, terminology support, and repetitive work while assigning humans responsibility for interpretation, error detection, cultural judgment, and final approval.
References
Bahdanau D, Cho K, Bengio Y. Neural machine translation by jointly learning to align and translate. In: Proceedings of the 3rd International Conference on Learning Representations. San Diego: ICLR; 2015.
Wu Y, Schuster M, Chen Z, Le QV, Norouzi M, Macherey W, et al. Google's neural machine translation system: bridging the gap between human and machine translation. arXiv. 2016;1609.08144.
Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, et al. Attention is all you need. In: Advances in Neural Information Processing Systems 30. Red Hook (NY): Curran Associates; 2017. p. 5998-6008.
OpenAI. GPT-4 technical report. arXiv. 2023;2303.08774.
Zhu W, Liu H, Dong Q, Xu J, Huang S, Kong L, et al. Multilingual machine translation with large language models: empirical results and analysis. Findings of the Association for Computational Linguistics: NAACL 2024. Mexico City: Association for Computational Linguistics; 2024. p. 2765-2781. doi:10.18653/v1/2024.findings-naacl.176.
Li J, Zhou H, Huang S, Cheng S, Chen J. Eliciting the translation ability of large language models via multilingual finetuning with translation instructions. Trans Assoc Comput Linguist. 2024;12:576-592. doi:10.1162/tacl_a_00655.
Papineni K, Roukos S, Ward T, Zhu WJ. BLEU: a method for automatic evaluation of machine translation. In: Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics. Philadelphia: Association for Computational Linguistics; 2002. p. 311-318. doi:10.3115/1073083.1073135.
Rei R, Stewart C, Farinha AC, Lavie A. COMET: a neural framework for MT evaluation. In: Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing. Online: Association for Computational Linguistics; 2020. p. 2685-2702. doi:10.18653/v1/2020.emnlp-main.213.
Freitag M, Foster G, Grangier D, Ratnakar V, Tan Q, Macherey W. Experts, errors, and context: a large-scale study of human evaluation for machine translation. Trans Assoc Comput Linguist. 2021;9:1460-1474. doi:10.1162/tacl_a_00437.
Specia L, Farzindar A. Estimating machine translation post-editing effort with HTER. In: Proceedings of the Second Joint EM+/CNGL Workshop: Bringing MT to the User. Denver: Association for Machine Translation in the Americas; 2010. p. 33-43.
Gaspari F, Toral A, Naskar SK, Groves D, Way A. Perception vs. reality: measuring machine translation post-editing productivity. In: Proceedings of the 11th Conference of the Association for Machine Translation in the Americas. Vancouver: Association for Machine Translation in the Americas; 2014. p. 60-72.
Sanchez-Torron M, Koehn P. Machine translation quality and post-editor productivity. In: Proceedings of the AMTA 2016 MT Researchers' Track. Austin: Association for Machine Translation in the Americas; 2016. p. 16-26.
Guerreiro NM, Alves DM, Waldendorf J, Haddow B, Birch A, Colombo P, et al. Hallucinations in large multilingual translation models. Trans Assoc Comput Linguist. 2023;11:1500-1517. doi:10.1162/tacl_a_00615.
International Organization for Standardization. ISO 18587:2017. Translation services—post-editing of machine translation output—requirements. Geneva: ISO; 2017.
Finkelstein M, Kovacs G, Caswell I, Domhan T, Peter JT, Juraska J, Freitag M, Vilar D. Google Translate's research submission to WMT2025. In: Proceedings of the Tenth Conference on Machine Translation. Suzhou: Association for Computational Linguistics; 2025. p. 723-731. doi:10.18653/v1/2025.wmt-1.48.
Kocmi T, Artemova E, Avramidis E, Bawden R, Bojar O, Dranch K, et al. Findings of the WMT25 general machine translation shared task: time to stop evaluating on easy test sets. In: Proceedings of the Tenth Conference on Machine Translation. Suzhou: Association for Computational Linguistics; 2025.
Lavie A, Hanneman G, Agrawal S, Kanojia D, Lo CK, Zouhar V, et al. Findings of the WMT25 shared task on automated translation evaluation systems: linguistic diversity is challenging and references still help. In: Proceedings of the Tenth Conference on Machine Translation. Suzhou: Association for Computational Linguistics; 2025. p. 436-483.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 ESRAA HASSAN, Bushra Naseer (Author)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
