RISKS OF INFORMATION LEAKAGE USING ONLINE MACHINE TRANSLATION SERVICES
DOI:
https://doi.org/10.28925/2663-4023.2025.27.730Keywords:
machine translation, online services, information leakage, cybersecurity, data confidentialityAbstract
The article is devoted to the analysis of the risks of information leakage when using online machine translation services and suggests methods for minimizing these risks. In connection with the development of technologies and the growing popularity of online services for text translation, the issues of ensuring confidentiality and data security are becoming particularly relevant. Online machine translation services, such as Google Translate, Microsoft Translator, DeepL and others, are convenient tools for processing large volumes of texts, however, they pose significant threats associated with the leakage of sensitive information, which can have serious legal, financial and reputational consequences. The article examines the main factors contributing to information leakage when using online translation services, in particular, data storage on the servers of third-party service providers, imperfect privacy policies and technical vulnerabilities. In particular, the problem of data security when transmitting sensitive information through online translation services is emphasized. Examples of specific cases of data leakage are given, emphasizing the need for increased attention to security aspects when using these services in professional activities.
Particular attention is paid to data protection methods, such as encryption, depersonalization and anonymization, which reduce the risks of leakage of sensitive information. At the same time, attention is focused on the need to improve existing standards and security policies in online machine translation services, which will ensure proper protection of confidential information, in particular in such areas as legal, medical, financial and government activities.
The article also compares the level of security of various online machine translation services, in particular their ability to ensure proper information protection through guaranteed data deletion after translation. The prospect of using local translation models as an alternative to cloud services, which reduces the risks of data leakage, is separately considered.
The main areas of further research in the field of information security of machine translation are highlighted, in particular regarding new data protection technologies and increasing user awareness of potential threats.
Downloads
References
Aletras, N., Gkotsis, G., & Baldwin, T. (2021). Neural machine translation and data privacy: Challenges and solutions. Computational Linguistics Journal, 47(3), 521–540.
Bajpai, S., & Sharma, R. (2023). Cybersecurity threats in online translation services. Journal of Information Security, 12(1), 67–84.
Council of Europe. (2022). GDPR and machine translation: Legal implications of cross-border data processing. European Data Protection Review, 5(2), 201–215.
OpenAI Research. (2023). Privacy issues in large-scale AI translation models. AI & Society, 38(4), 299–315.
Smith, J., Tan, H., & Williams, K. (2020). Cloud-based translation services: Vulnerabilities and risk mitigation. Cybersecurity Review, 45(2), 102–118.
Vashee, K. (2022). Ethical challenges in AI-driven machine translation. Journal of Artificial Intelligence Ethics, 7(1), 33–48.
Belinkov, Y., & Bisk, Y. (2021). Robustness of neural machine translation to input perturbations. Transactions of the Association for Computational Linguistics, 9, 1169–1185.
Koehn, P. (2020). Neural machine translation. Cambridge University Press.
Sun, Z., Wang, J., & Li, H. (2023). Privacy-preserving approaches in neural machine translation: A survey. IEEE Transactions on Artificial Intelligence, 4(2), 145–162.
Kocmi, T., & Bojar, O. (2022). Security risks of online translation platforms: A case study on data leakage. Cybersecurity & AI Research Journal, 5(1), 21–39.
Ruder, S., Vulić, I., & Søgaard, A. (2021). A survey of cross-lingual word embedding models. Journal of Natural Language Processing, 58(3), 215–238.
Zoph, B., Yuret, D., & Knight, K. (2022). Data security in neural machine translation: Threats and countermeasures. Transactions of the Association for Computational Linguistics, 10, 231–250.
Green, S., & Manning, C. D. (2023). Confidentiality in AI-based translation: Evaluating risks and solutions. Journal of Machine Learning & Security, 6(2), 101–123.
Klymchuk, O. (2023). Threats of data leakage when using machine translation in government agencies. Scientific Notes of the Institute of Information Security, 4(1), 45–58.
Sydorenko, L. (2022). Cryptographic protection of textual information in automated translation systems. Cybersecurity: theory and practice, 8(3), 102–118.
Zakharchenko, M., & Polishchuk, V. (2023). Local machine translation models as an alternative to cloud services. Analysis of efficiency and security. Ukrainian Journal of Computer Linguistics, 6(2), 77–90.
Serhieieva, T. I. (2020). Fundamentals of information security in information systems. Kyiv: Naukova dumka.
Taran, O. V. (2019). Data protection methods in online translation services. Kharkiv: KhNU.
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Леся Матвієнко, Любов Хоменко
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
