





































Education, Language and Sociology Research 

ISSN 2690-3644 (Print) ISSN 2690-3652 (Online) 

Vol. 6, No. 4, 2025 

www.scholink.org/ojs/index.php/elsr 

37 
 

Original Paper 

Integrating AI-Based Machine Translation in Translation 

Pedagogy: Evidence from a Mixed-Methods Study in Vietnam 

Nguyen Thi Viet Phuong1*, Pham Thi Huong2 & Pham Phuong Lan2 

1, 2 Lecturers, Department of Foreign Languages, Trade Union University, Vietnam 

* Corresponding Author: Nguyen Thi Viet Phuong, Email: phuongntv@dhcd.edu.vn 

 

Received: November 1, 2025   Accepted: November 15, 2025  Online Published: November 17, 2025 

doi:10.22158/elsr.v6n4p37        URL: http://dx.doi.org/10.22158/elsr.v6n4p37 

 

Abstract 

The rise of artificial intelligence (AI), neural machine translation (NMT), and large language models 

(LLMs) such as GPT, Gemini, and Claude has transformed translation pedagogy worldwide. However, 

empirical research on how AI affects translator training in Vietnam remains scarce. 

This study adopts a mixed-methods design to examine AI adoption in translation classrooms, combining 

survey data from 93 students and five lecturers with semi-structured interviews. Quantitative data were 

analyzed through descriptive and comparative statistics, while qualitative findings were thematically 

coded. 

Results indicate high readiness and positive attitudes among both students and teachers, yet AI 

applications remain fragmented and unsystematic. Tools such as ChatGPT, Google Translate, and 

DeepL enhance translation speed, vocabulary, and critical reflection but also pose risks of over-reliance, 

diminished linguistic sensitivity, and ethical issues. Four key determinants-digital competence, 

infrastructure, institutional policy, and academic integrity culture-shape effective implementation. 

The study recommends a Pedagogically Guided AI Integration (PGAI) model to ensure responsible, 

sustainable adoption. It advances the concept of AI Literacy for Translators and contributes to the 

emerging framework of Translation Pedagogy 4.0 in the context of digital transformation. 

Keywords 

Artificial Intelligence, Neural Machine Translation, Translation Pedagogy, AI Literacy, Post-Editing, 

Responsible AI, Vietnam 

 

1. Introduction 

The rapid advancement of artificial intelligence (AI), particularly neural machine translation (NMT) and 

large language models (LLMs) such as GPT, Gemini, and Claude-has profoundly reshaped translation 



www.scholink.org/ojs/index.php/elsr              Education, Language and Sociology Research              Vol. 6, No. 4, 2025 

38 
Published by SCHOLINK INC. 

pedagogy worldwide. Once confined to rule-based and statistical systems, machine translation now relies 

on deep-learning architectures capable of capturing complex semantic and contextual relationships across 

languages (Vaswani et al., 2017; Jurafsky & Martin, 2023). As a result, AI has evolved from being a 

purely technical aid to becoming a pedagogical agent that enhances teachers’ ability to personalize 

instruction while offering learners immediate, data-driven feedback (Bozkurt et al., 2020; Krüger, 2024). 

Within the broader educational paradigm shift from teacher-centered to AI-enhanced learner-centered 

approaches, translation education has emerged as a pivotal domain for experimentation and reform. The 

integration of AI tools enables students not only to generate translations but also to analyze, critique, and 

post - edit machine - produced texts. Through this process, learners develop higher-order competences 

such as critical thinking, metacognitive awareness, and ethical reflection, abilities that align with twenty-

first-century translator-training objectives (Liu & Liang, 2024; Morentsova, 2022). Nevertheless, this 

technological transformation brings with it pedagogical and ethical challenges, including over-reliance 

on automation, diminished linguistic creativity, and the risk of plagiarism (Kanglang, 2021). 

In Vietnam, these global shifts coincide with national strategies that place AI at the heart of educational 

innovation. The National Digital Transformation Strategy (2020-2030) identifies AI as a key driver for 

innovation and productivity in higher education, while the Higher-Education Development Strategy 

(Ministry of Education and Training, 2021) highlights AI integration and digital pedagogy as pillars of 

modernization. Against this backdrop, exploring AI-supported translation pedagogy is both timely and 

essential for aligning translator-training programs in Vietnam with international standards and labor-

market expectations. 

Despite the proliferation of AI translation tools, their pedagogical application in Vietnam remains 

fragmented. Most translation courses continue to rely on conventional teacher-centered practices, with 

little methodological guidance for embedding AI into curriculum design, assessment, or reflective 

learning. The sporadic adoption of tools such as ChatGPT, DeepL, or Trados often occurs informally 

and without alignment to clear learning outcomes. Moreover, there is limited empirical evidence 

regarding how AI influences students’ translation competence, post-editing literacy, or ethical awareness. 

While international research documents both positive effects-such as enhanced fluency and reduced 

anxiety-and potential risks, including automation dependency and academic dishonesty (Klimova et al., 

2022; Krüger, 2024), comparable studies within Vietnamese universities remain scarce. The country’s 

distinct infrastructural and cultural conditions necessitate localized investigation rather than the uncritical 

transfer of Western pedagogical models. 

Existing literature on AI-assisted translation pedagogy in Vietnam remains largely exploratory, 

characterized by conceptual reflections or small-scale classroom observations. To date, no 

comprehensive mixed-methods study has systematically examined teachers’ and students’ readiness, 

perceptions, and practices in AI-enhanced translation training. Furthermore, there is no context-specific 

pedagogical model designed to integrate AI technologies into English-language programs in a sustainable 

and ethically responsible way. International frameworks such as the European Master’s in Translation 



www.scholink.org/ojs/index.php/elsr              Education, Language and Sociology Research              Vol. 6, No. 4, 2025 

39 
Published by SCHOLINK INC. 

(EMT, 2017) and the PACTE Group’s Translation-Competence Model (2003) underscore the importance 

of developing technological-AI competence and critical-ethical competence as integral components of 

contemporary translator education. However, localized adaptations of these frameworks to Vietnam’s 

higher-education context are virtually absent. Addressing this gap requires robust empirical evidence and 

theoretically grounded design principles for guided AI integration that are both pedagogically sound and 

culturally sensitive. 

Against this background, the present study pursues three interrelated objectives. First, it seeks to evaluate 

the current state of AI application among translation students and lecturers in Vietnamese higher 

education. Second, it aims to identify key opportunities and challenges emerging from AI-supported 

teaching and learning practices. Third, it endeavors to propose a sustainable, pedagogically guided model 

for integrating AI into translation curricula that balances technological innovation with academic 

integrity and human creativity. Correspondingly, the study addresses four central research questions: (1) 

What are the levels of awareness and actual use of AI technologies among translation students and 

instructors? (2) How does AI affect translation learning, post-editing performance, and critical-thinking 

development? (3) What technical, pedagogical, and ethical barriers constrain effective AI integration? 

and (4) Which pedagogical framework can best support responsible and sustainable AI adoption in 

translation training? 

By addressing these questions, this paper contributes both empirical evidence and conceptual insights to 

the evolving discourse on AI-assisted translation pedagogy. It situates Vietnam’s translator education 

within global debates on Translation Pedagogy 4.0 and AI Literacy, providing a theoretically informed 

and contextually grounded foundation for future policy development, curriculum innovation, and 

responsible AI use in higher education. 

 

2. Literature Review 

2.1 AI in Language and Translation Education 

The rise of neural machine translation (NMT), natural language processing (NLP), and large language 

models (LLMs) has revolutionized how languages and translation are taught and learned. Since the 

introduction of attention-based deep-learning architectures (Vaswani et al., 2017; Jurafsky & Martin, 

2023), NMT systems have evolved from statistical phrase-based models to self-learning systems capable 

of modeling contextual and semantic nuances across languages (Floridi & Chiriatti, 2020; Russell & 

Norvig, 2021). In the educational domain, these technologies have become essential components of the 

digital-learning ecosystem, enabling authentic, data-driven translation tasks and automated feedback that 

complement human instruction (Klimova et al., 2022; Duan et al., 2025). 

Within this transformation, AI functions as a pedagogical agent rather than a mere computational 

instrument. Bozkurt et al. (2020) and Krüger (2024) describe AI as an intelligent co-participant that 

supports adaptive learning, scaffolding, and personalized error correction. Studies such as Yang (2022) 

and Ulitkin and Ivanova (2024) demonstrate that AI-driven translation environments foster greater 



www.scholink.org/ojs/index.php/elsr              Education, Language and Sociology Research              Vol. 6, No. 4, 2025 

40 
Published by SCHOLINK INC. 

learner autonomy and engagement when combined with guided human oversight. Thus, AI is reshaping 

translator education through AI-enhanced experiential learning, where machine-generated outputs 

become stimuli for reflection, critique, and skill development rather than final products. 

2.2 Translation Competence and AI Literacy 

Modern translation studies conceptualize competence as a multidimensional construct encompassing 

linguistic, cultural, strategic, and technological components. The PACTE Model (2003) and the 

European Master’s in Translation (EMT) framework (2017) remain foundational, identifying translation 

competence as the coordinated integration of linguistic-cultural, instrumental, and psycho-physiological 

sub-competences. Building upon these, Krüger (2024) introduces the concept of AI Literacy for 

Translators, which extends technological competence to include critical awareness of algorithmic 

mediation, data bias, and ethical implications. 

In the digital-age context, translation competence now comprises at least three key dimensions: (1) 

linguistic-cultural competence, ensuring contextual accuracy and intercultural sensitivity (Byram, 1997; 

Spencer-Oatey, 2008); (2) technological-AI competence, reflecting the ability to use and evaluate NMT 

and computer-assisted translation (CAT) systems effectively (Mishra & Koehler, 2006; Ulitkin & 

Ivanova, 2024); and (3) critical-ethical competence, which equips translators to maintain human agency, 

quality control, and academic integrity in AI-mediated workflows. These evolving competences illustrate 

that translation pedagogy must move beyond tool training toward ethical, reflective, and hybrid human-

AI collaboration. 

2.3 AI-Assisted Translation Pedagogy 

The integration of AI into translator education has given rise to diverse blended-learning and AI-

enhanced classroom models. Empirical research demonstrates that combining AI tools with collaborative, 

constructivist learning strategies can enhance both performance and motivation (Kruk & Kałużna, 2024; 

González Davies, 2017). Within such environments, post-editing becomes a central pedagogical practice: 

students engage in error detection, style refinement, and cross-lingual comparison, thereby strengthening 

cognitive control and critical-thinking skills (Morentsova, 2022; Omar & Salih, 2024). 

However, these benefits coexist with significant risks. Over-reliance on machine output can erode 

linguistic creativity, reduce contextual awareness, and encourage digital plagiarism or passive learning 

habits (Rybina et al., 2025). Kanglang (2021) warns that unregulated AI use may lead to a decline in 

translators’ interpretive abilities and ethical accountability. Consequently, effective AI pedagogy 

requires structured guidance, balancing automation with reflective human evaluation-a principle often 

framed as Pedagogically Guided AI Integration (PGAI). 

2.4 Legal and Ethical Guidelines 

The global discourse on AI ethics underscores the need for responsible, human -centered innovation in 

education. UNESCO’s Recommendation on the Ethics of Artificial Intelligence (2021) articulates 

universal principles of transparency, fairness, accountability, and human dignity in AI deployment across 

sectors, including education. Similarly, the OECD (2022) report Artificial Intelligence in Education 



www.scholink.org/ojs/index.php/elsr              Education, Language and Sociology Research              Vol. 6, No. 4, 2025 

41 
Published by SCHOLINK INC. 

advocates for policy frameworks that foster responsible innovation while safeguarding learners’ privacy 

and academic integrity. The European Union’s AI Act (2023)-the world’s first comprehensive legal 

framework for AI-adopts a risk-based approach, classifying AI applications by their potential harms and 

mandating transparency, data protection, and traceability. 

Together, these instruments establish the ethical and regulatory foundation for AI-assisted translation 

pedagogy. They emphasize that integrating AI into translator training must not only advance 

technological proficiency but also uphold principles of equity, responsibility, and educational integrity. 

For developing countries like Vietnam, these global frameworks offer critical reference points for 

designing localized standards of Responsible AI in language and translation education. 

 

3. Methodology 

3.1 Research Design 

This study adopted a mixed-methods research design combining quantitative and qualitative approaches 

to gain a comprehensive understanding of how artificial intelligence (AI) tools are being used and 

perceived in translation pedagogy within Vietnamese higher education. The quantitative component 

consisted of a structured survey administered to translation students and lecturers, while the qualitative 

component involved semi-structured interviews designed to explore participants’ lived experiences and 

pedagogical reflections in greater depth. 

The rationale for this design lies in the complementarity between numerical and narrative data, enabling 

both breadth and depth of insight. Quantitative analysis captures general trends in AI awareness, usage, 

and perceived impact, whereas qualitative insights illuminate contextualized meanings, attitudes, and 

ethical concerns that are not easily measurable. To enhance the validity and reliability of findings, 

methodological triangulation was applied-cross-verifying data across instruments, sources, and analytical 

methods (Creswell & Plano Clark, 2018). This integrative design ensures that observed patterns are both 

statistically supported and theoretically meaningful, reflecting the complex intersection of technology, 

pedagogy, and human agency in translation education. 

3.2 Participants 

The research sample comprised 93 undergraduate students majoring in English Language Studies (Years 

3 and 4) and five lecturers specializing in translation, interpreting, and translation technology. The 

student group was selected to represent learners with sufficient exposure to both translation practice and 

AI-based tools such as Google Translate, DeepL, and ChatGPT. The lecturers were chosen from three 

universities offering translator - training programs in Northern Vietnam. 

Participants reflected a diversity of demographic and academic backgrounds, including differences in 

gender, academic standing, and prior experience with digital technologies. This heterogeneity provided 

a representative snapshot of the evolving translation -learning ecosystem in Vietnamese higher education.  

 

 



www.scholink.org/ojs/index.php/elsr              Education, Language and Sociology Research              Vol. 6, No. 4, 2025 

42 
Published by SCHOLINK INC. 

3.3 Instruments 

Data were collected through two main instruments developed and validated through a multi-stage process. 

The first instrument was a 22-item questionnaire based on a five-point Likert scale (ranging from 1 = 

Strongly Disagree to 5 = Strongly Agree), measuring participants’ (i) AI awareness and literacy, (ii) 

frequency and purpose of AI use, (iii) perceived pedagogical benefits, and (iv) perceived ethical or 

technical risks. The questionnaire was piloted with 20 students to ensure linguistic clarity and contextual 

appropriateness, achieving a Cronbach’s Alpha reliability coefficient of ≥ 0.7, indicating strong internal 

consistency (Hair et al., 2019). 

The second instrument consisted of semi-structured interview protocols comprising six thematic domains: 

(1) perceptions of AI in translation learning, (2) classroom integration practices, (3) post-editing 

strategies, (4) ethical and academic integrity issues, (5) perceived institutional or infrastructural barriers, 

and (6) recommendations for sustainable AI adoption. Each interview lasted 30-45 minutes, conducted 

in Vietnamese and subsequently translated into English for analysis. The dual-instrument design allowed 

for both quantifiable comparisons and in-depth exploration of pedagogical dynamics. 

3.4 Data Analysis 

Quantitative data were analyzed using SPSS (version 26.0) through descriptive and inferential statistics. 

Descriptive analysis (means, standard deviations, and frequency distributions) was employed to profile 

AI usage trends, while independent-samples t-tests and ANOVA were used to examine group differences 

by academic level, gender, and AI familiarity. 

Qualitative data from the semi-structured interviews were transcribed verbatim and analyzed using 

thematic analysis (Braun & Clarke, 2019). A coding framework (T1-T4) was developed to identify 

recurring patterns related to: 

T1: awareness and readiness, 

T2: application and learning practices, 

T3: perceived risks and ethical dilemmas, and 

T4: proposed solutions and pedagogical innovations. 

Cross-validation between quantitative and qualitative datasets enabled a multi-dimensional interpretation 

of the findings, strengthening the robustness of the conclusions drawn. 

3.5 Ethical Considerations 

The study adhered to the ethical standards of educational and social research, following the principles 

outlined in UNESCO’s Recommendation on the Ethics of Artificial Intelligence (2021) and the OECD 

(2022) guidelines on AI in education. All participants were informed of the research purpose, procedures, 

and their right to withdraw at any stage without consequence. Participation was entirely voluntary and 

anonymous, with all personal identifiers removed during data processing. 

Informed consent was obtained from all respondents, and digital data were stored securely in compliance 

with institutional and international data-protection protocols. The ethical framework thus ensured 



www.scholink.org/ojs/index.php/elsr              Education, Language and Sociology Research              Vol. 6, No. 4, 2025 

43 
Published by SCHOLINK INC. 

transparency, respect for participants’ autonomy, and the protection of sensitive educational information 

in accordance with global norms of Responsible AI and research integrity. 

 

4. Results 

4.1 Students’ AI Readiness and Usage 

Survey results reveal a high level of AI readiness and frequent usage among translation students. The 

mean frequency score for AI use reached M = 3.56/5, indicating that most participants engage with AI 

tools on a regular basis to support translation learning and related linguistic tasks. Among the 

technologies used, ChatGPT emerged as the most popular tool, reported by 72% of respondents, followed 

by Google Translate (65%) and DeepL Translator (38%). 

While these tools were primarily employed for preliminary translation and vocabulary enhancement, 

qualitative interviews show that students also utilized AI for post-editing exercises, paraphrasing, and 

cross-checking terminology accuracy. Comparative analysis between third - and fourth-year students 

revealed statistically significant differences in technological self-efficacy and autonomy, with senior 

students demonstrating higher proficiency in evaluating AI output and managing post-editing workflows. 

The findings suggest a gradual transition from dependency to critical engagement, highlighting the role 

of classroom scaffolding and cumulative digital exposure in shaping AI literacy. 

4.2 Teachers’ AI Integration Practices 

Lecturers’ responses demonstrated a cautious yet growing interest in leveraging AI to enhance translation 

pedagogy. The qualitative data indicated that instructors integrated AI primarily into learning-material 

design, corpus simulation, and post-editing exercises, aligning with blended-learning and task-based 

approaches. Teachers reported using AI tools to generate sample translations for class discussions, to 

illustrate machine-human translation discrepancies, and to develop interactive tasks that promote critical 

error analysis. 

However, the integration process remains fragmented and uneven, hindered by several contextual 

constraints. Key barriers include limited technological infrastructure, inconsistent access to premium AI 

tools, and the absence of a clear pedagogical framework guiding ethical and effective AI usage. 

Additionally, the digital competence gap between instructors of different age groups or professional 

backgrounds further restricts consistent adoption. Despite these challenges, most teachers expressed a 

positive attitude toward AI, acknowledging its potential to foster learner autonomy and engagement when 

accompanied by appropriate pedagogical oversight. 

4.3 Perceived Benefits and Risks 

Both student and teacher participants identified a balanced mix of benefits and risks associated with AI-

assisted translation learning. 

From a pedagogical perspective, the benefits were considerable. Participants agreed that AI significantly 

accelerates linguistic processing, allowing users to handle large volumes of text more efficiently. Many 

reported that AI tools contribute to vocabulary expansion and semantic awareness, as students actively 



www.scholink.org/ojs/index.php/elsr              Education, Language and Sociology Research              Vol. 6, No. 4, 2025 

44 
Published by SCHOLINK INC. 

compare and refine AI-generated outputs against human translations. Furthermore, exposure to imperfect 

machine translations encouraged critical thinking, as learners were prompted to question linguistic 

accuracy, stylistic choices, and cultural equivalence. This reflective engagement aligns with the 

objectives of higher-order learning and metacognitive development in translator education. 

Nevertheless, participants also voiced concerns about academic and cognitive risks. Over-reliance on AI 

outputs may diminish students’ attention to grammar, stylistic nuance, and contextual appropriateness, 

leading to mechanical reproduction of AI text without critical mediation. Teachers cited instances of 

digital plagiarism, where students submitted AI-generated translations with minimal revision-and noted 

a decline in effort toward manual translation and editing skills. Additionally, the opacity of AI algorithms 

raised ethical concerns about data transparency, authorship accountability, and the erosion of academic 

integrity. These findings underscore the necessity of cultivating responsible AI use through explicit 

instruction in post-editing, citation ethics, and reflective evaluation. 

4.4 Key Influencing Factors 

The synthesis of quantitative and qualitative data identified four major factors shaping the effectiveness 

of AI integration in translation pedagogy. 

First, digital competence and user attitudes emerged as decisive variables. Students and teachers 

possessing higher digital literacy and more positive dispositions toward technology demonstrated more 

innovative and self-regulated use of AI tools. Second, technological infrastructure and software access 

strongly influenced adoption levels, as many institutions lacked stable internet connectivity or 

institutional subscriptions to advanced AI platforms. 

Third, the absence of a pedagogical and policy framework governing AI use in translation education 

created uncertainty among educators regarding ethical limits, data protection, and evaluation criteria. 

Finally, the broader culture of AI use and academic integrity played a crucial role. Where institutional 

cultures emphasized transparency, originality, and reflective learning, AI was more likely to be integrated 

responsibly. Conversely, environments lacking clear norms were associated with uncritical imitation and 

dependence. 

Together, these factors reveal that the success of AI-assisted translation pedagogy depends not only on 

technological availability but also on pedagogical readiness, institutional policy, and ethical orientation. 

A structured, pedagogically guided approach to AI integration is therefore imperative for achieving 

sustainable, responsible innovation in translator education. 

 

5. Discussion 

5.1 AI as a Catalyst for Pedagogical Shift 

The findings of this study reaffirm that artificial intelligence (AI) is not merely a technological 

supplement but a catalyst for deep pedagogical transformation in translator education. Consistent with 

the constructivist and experiential paradigms outlined by Kolb (1984) and Vygotsky (1978), AI redefines 

the classroom as a dynamic, data-driven environment where learners interact critically with machine 



www.scholink.org/ojs/index.php/elsr              Education, Language and Sociology Research              Vol. 6, No. 4, 2025 

45 
Published by SCHOLINK INC. 

outputs. As emphasized by Muñoz-Basols et al. (2023) and Ulitkin & Ivanova (2024), this environment 

shifts the lecturer’s function from that of a traditional content transmitter to a pedagogical designer, 

cognitive coach, and ethical advisor, a professional who orchestrates human-machine interaction while 

safeguarding linguistic and moral integrity. 

In this reconfigured ecosystem, data-driven learning (DDL) becomes a cornerstone of translator training. 

Students exploit AI’s vast linguistic corpora to observe real-time language patterns, test hypotheses, and 

refine stylistic choices. Such interaction cultivates critical reading, post-editing, and reflective judgment-

skills that bridge theory and practice. This shift resonates with the current vision of Translation Pedagogy 

4.0, where AI functions as both collaborator and mirror, reflecting the learner’s evolving competence 

rather than replacing human cognition. 

5.2 Risks of Over-Reliance 

Despite these pedagogical advantages, the research also validates concerns raised by Kanglang (2021) 

and Krüger (2024) regarding the risks of uncritical dependence on AI technologies. Students’ tendency 

to accept AI output without rigorous evaluation illustrates what Kanglang terms automation 

complacency-a phenomenon that erodes grammatical precision, stylistic sensitivity, and intercultural 

awareness. Krüger’s AI-literacy model similarly warns that without guided reflection, users may develop 

“surface efficiency” at the expense of deeper linguistic reasoning. 

To mitigate such risks, this study proposes graduated AI-use thresholds aligned with task complexity. In 

introductory courses, AI may serve as a stimulus for terminology exploration and comparative error 

analysis; in intermediate courses, it should be restricted to supervised post-editing; and in advanced 

professional-practice modules, AI engagement ought to emphasize ethical evaluation and creative 

augmentation, not substitution. This scaffolding ensures that learners’ agency and accountability grow 

in proportion to technological autonomy, maintaining equilibrium between automation and human 

expertise. 

5.3 The Need for Structured, Guided AI Integration 

The results underscore an urgent need for a structured and pedagogically guided model of AI integration 

in translator education. The absence of institutional frameworks and ethical guidelines has produced 

fragmented practices, echoing the global call for Responsible AI articulated by UNESCO (2021) and the 

OECD (2022). Within translation pedagogy, this translates into the imperative to operationalize AI 

Literacy for Translators, a construct defined by Krüger (2024) as the interplay of technological 

proficiency, critical awareness, and ethical discernment in AI-mediated communication. 

Embedding AI literacy within existing competence frameworks such as the European Master’s in 

Translation (EMT, 2017) and PACTE (2003) models requires the evolution toward an EMT-Hybrid 

Competence Framework. This hybrid model integrates (1) linguistic-cultural competence for contextual 

accuracy, (2) technological-AI competence for intelligent tool use and evaluation, and (3) critical-ethical 

competence for responsible decision-making. The study’s evidence confirms that without such explicit 

structuring, AI implementation risks remaining incidental and ethically ambiguous. 



www.scholink.org/ojs/index.php/elsr              Education, Language and Sociology Research              Vol. 6, No. 4, 2025 

46 
Published by SCHOLINK INC. 

Ultimately, sustainable innovation in translation education will depend on the coexistence of humanistic 

pedagogy and algorithmic intelligence. Guided AI integration-anchored in reflective practice, 

professional ethics, and curricular coherence-can transform translator training from a reactive adaptation 

to a proactive re-envisioning of what it means to learn, teach, and translate in the digital age. 

 

6. Conclusion 

6.1 Key Findings 

This mixed-method study provides empirical insights into the early-stage integration of artificial 

intelligence (AI) in translation education within the Vietnamese higher education context. The findings 

indicate that both students and lecturers demonstrate a high level of readiness and positive disposition 

toward AI, yet their applications remain fragmented and inconsistent across courses and institutions. 

AI-based tools such as ChatGPT, Google Translate, and DeepL are widely used to enhance vocabulary 

learning, post-editing, and comparative translation analysis. However, the pedagogical use of AI still 

lacks systematic guidance, resulting in uneven outcomes. The research highlights that while AI presents 

significant opportunities for innovation, it also introduces academic and ethical risks, including over-

reliance, reduced linguistic sensitivity, and emerging challenges of digital plagiarism and transparency. 

Therefore, the study concludes that a controlled, pedagogically guided approach to AI integration is 

essential. AI should not substitute human cognition but rather function as a co-agent that stimulates 

critical reflection, creativity, and ethical awareness in translation learning. The findings emphasize that 

sustainable innovation in translator education depends on balancing technological capability with 

humanistic pedagogy and institutional responsibility. 

6.2 Theoretical Contributions 

Theoretically, this research extends existing literature by offering context-specific empirical evidence 

from Vietnam, an underrepresented region in global studies of AI-assisted translation pedagogy. It 

validates and contextualizes the conceptual frameworks proposed by Krüger (2024), Ulitkin and Ivanova 

(2024), and Muñoz-Basols et al. (2023), demonstrating how AI literacy can be cultivated within 

developing educational ecosystems undergoing digital transformation. 

By articulating the notion of “AI Literacy for Translators” as a multidimensional construct-encompassing 

technological, critical, and ethical competences-this study contributes to the advancement of Translation 

Pedagogy 4.0. Furthermore, it supports the evolution toward an EMT-Hybrid Competence Model, where 

linguistic-cultural competence is dynamically linked with technological-AI and ethical decision-making 

capacities. 

Collectively, these contributions enrich the global discourse on translator education, bridging the gap 

between technological innovation and pedagogical sustainability. They position Vietnam as a meaningful 

case study in the worldwide movement toward responsible, human-centered AI integration in translation 

training, a cornerstone for future research and curriculum design. 

 



www.scholink.org/ojs/index.php/elsr              Education, Language and Sociology Research              Vol. 6, No. 4, 2025 

47 
Published by SCHOLINK INC. 

References 

International Sources 

Bozkurt, A., Jung, I., Xiao, J., Vladimirschi, V., Schuwer, R., Egorov, G., et al. (2020). A global outlook 

to the interruption of education due to COVID-19 pandemic: Navigating in a time of uncertainty 

and crisis. Asian Journal of Distance Education, 15(1), 1-126. 

https://doi.org/10.5281/zenodo.3878572 

Byram, M. (1997). Teaching and assessing intercultural communicative competence. Multilingual 

Matters. 

Duan, H., Gao, X., & Zhang, Y. (2025). The application of AI translation tools in improving students’ 

translation fidelity and accuracy. Arab World English Journal. https://doi.org/10.24093/awej/AI.16 

European Commission. (2017). European Master’s in Translation (EMT) competence framework 2017. 

Directorate-General for Translation. 

European Union. (2023). Artificial Intelligence Act. Official Journal of the European Union. 

Floridi, L., & Chiriatti, M. (2020). GPT-3: Its nature, scope, limits, and consequences. Minds and 

Machines, 30(4), 681-694. https://doi.org/10.1007/s11023-020-09548-1 

González Davies, M. (2017). Collaborative learning in translation education. Routledge. 

https://doi.org/10.4324/9781315623139-9 

Hair, J. F., Hult, G. T. M., Ringle, C. M., & Sarstedt, M. (2019). A primer on partial least squares 

structural equation modeling (PLS-SEM) (2nd ed.). SAGE Publications. 

https://doi.org/10.3926/oss.37 

Hutchins, J. (2005). The history of machine translation in a nutshell. International Journal of Translation, 

17(1), 5-20. 

Jurafsky, D., & Martin, J. H. (2023). Speech and language processing (3rd ed.). Prentice Hall. 

Kanglang, L. (2021). Artificial intelligence (AI) and translation teaching: A critical perspective on the 

transformation of education. International Journal of Educational Sciences, 33(2), 115-129. 

https://doi.org/10.31901/24566322.2021/33.1-3.1159 

Kelly, D. (2005). A handbook for translator trainers. St. Jerome Publishing. 

Kenny, D. (2017). Machine translation for human translators. Routledge. 

https://doi.org/10.4324/9781315678481-27 

Klimova, B., Pikhart, M., Benites, A., Lehr, C., & Sanchez-Stockhammer, C. (2022). Neural machine 

translation in foreign language teaching and learning: A systematic review. Education and 

Information Technologies, 28, 663-682. https://doi.org/10.1007/s10639-022-11194-2 

Kolb, D. A. (1984). Experiential learning: Experience as the source of learning and development. 

Prentice Hall. 

Krüger, R. (2024). Outline of an artificial intelligence literacy framework for translation, interpreting and 

specialised communication. Lublin Studies in Modern Languages and Literature, 48(3), 11-23. 

https://doi.org/10.17951/lsmll.2024.48.3.11-23 



www.scholink.org/ojs/index.php/elsr              Education, Language and Sociology Research              Vol. 6, No. 4, 2025 

48 
Published by SCHOLINK INC. 

Kruk, M., & Kałużna, A. (2024). Investigating the role of AI tools in enhancing translation skills, 

emotional experiences, and motivation in L2 learning. European Journal of Education. 

https://doi.org/10.1111/ejed.12859 

Liu, K. (2020). Artificial intelligence (AI) and translation teaching. International Journal of Education 

and Science, 3(2), 55-63.  

Liu, Y., & Liang, J. (2024). Multidimensional comparison of Chinese-English interpreting outputs from 

human and machine: Implications for interpreting education. Linguistics and Education, 84, 101132. 

https://doi.org/10.1016/j.linged.2024.101273 

Mishra, P., & Koehler, M. J. (2006). Technological pedagogical content knowledge: A framework for 

teacher knowledge. Teachers College Record, 108(6), 1017-1054. https://doi.org/10.1111/j.1467-

9620.2006.00684.x 

Morentsova, A. (2022). Learning neural machine translation tools as a component of training professional 

translators. Innovate Pedagogy Journal, 14(3), 97-109. https://doi.org/10.32843/2663-

6085/2022/47.36 

Muñoz-Basols, J., Martín, M., & Calvi, M. V. (2023). Teaching translation and interpreting in the digital 

age: Innovation and ethics in language education. Routledge. 

OECD. (2022). Artificial intelligence in education: Promises and implications for teaching and learning. 

OECD Publishing. 

Omar, L., & Salih, A. (2024). Systematic review of English/Arabic machine translation post-editing. 

Informatics, 11(2), 23. https://doi.org/10.3390/informatics11020023 

PACTE Group. (2003). Building a translation competence model. In F. Alves (Ed.), Triangulating 

translation (pp. 43-66). John Benjamins. https://doi.org/10.1075/btl.45.06pac 

Pym, A. (2011). Training translators: Programmes, curricula, practices. John Benjamins. 

https://doi.org/10.1093/oxfordhb/9780199239306.013.0032 

Russell, S., & Norvig, P. (2021). Artificial intelligence: A modern approach (4th ed.). Pearson. 

Rybina, L., Koshil, O., & Hyryla, K. (2025). Risks of artificial intelligence overuse in translator training. 

Journal of Language and Education, 11(2), 85-97. 

Schäffner, C., & Krüger, R. (2016). Translation competence: New conceptualizations and assessments. 

The Interpreter and Translator Trainer, 10(1), 1-14. 

Spencer-Oatey, H. (2008). Culturally speaking: Culture, communication and politeness theory (2nd ed.). 

Continuum. 

Tannen, D. (1984). Conversational style: Analyzing talk among friends. Oxford University Press. 

Ulitkin, A., & Ivanova, E. (2024). Digital transformation in translator and interpreter training: AI tools 

and pedagogical implications. Translation and Interpreting Studies, 19(2), 45-68. 

UNESCO. (2021). Recommendation on the ethics of artificial intelligence. UNESCO Publishing. 



www.scholink.org/ojs/index.php/elsr              Education, Language and Sociology Research              Vol. 6, No. 4, 2025 

49 
Published by SCHOLINK INC. 

Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., & Polosukhin, 

I. (2017). Attention is all you need. Advances in Neural Information Processing Systems, 30, 5998-

6008. 

Vygotsky, L. S. (1978). Mind in society: The development of higher psychological processes. Harvard 

University Press. 

Yang, Y. (2022). Teaching translation in a digital environment: A three-phase AI-enhanced pedagogy. 

Journal of Language Teaching and Research, 13(6), 1211-1220. 

Yuxiu, Y. (2024). Application of translation technology based on AI in translation teaching. Systems and 

Soft Computing, 9(4), 201-215. https://doi.org/10.1016/j.sasc.2024.200072 

Vietnamese Sources 

Bộ Giáo dục và Đào tạo. (2021). Chiến lược phát triển giáo dục đại học giai đoạn 2021-2030. Hà Nội: 

Nhà xuất bản Giáo dục Việt Nam. 

Chính phủ Việt Nam. (2020). Chiến lược chuyển đổi số quốc gia đến năm 2030, tầm nhìn 2045. 

Chính phủ Việt Nam. (2022, 28/01). Quyết định số 146/QĐ-TTg phê duyệt Đề án “Nâng cao nhận thức, 

phổ cập kỹ năng và phát triển nguồn nhân lực chuyển đổi số quốc gia đến năm 2025, định hướng 

đến năm 2030.” 

Nguyễn Thị Thu Hiền, Trần Huệ Minh, & Nguyễn Tuấn Linh. (2025). Applying ChatGPT in education: 

Opportunities, challenges and recommendations for lecturers in the context of digital transformation. 

VNU Journal of Science: Education Research, 41(2). https://doi.org/10.25073/2588-

1159/vnuer.5066 

Nguyễn Thị Kim Sơn, Bùi Thị Thanh Hương, Chu Cẩm Thơ, Phạm Tuấn Anh, & Nguyễn Quốc Trí. 

(2021). The applications of machine learning in education science research. VNU Journal of Science: 

Education Research, 37(4), 19-26. https://doi.org/10.25073/2588-1159/vnuer.4562 

Phan Hoàng Diệu Linh. (2022). The effects of translation as a pedagogical tool in teaching non-English-

majored students. VNU Journal of Science: Education Research, 38(4), 20-30. 

Phạm Vĩnh Khang, & Nguyễn Hồng Bửu Long. (2022). Hướng đến tiền huấn luyện cross-attention trong 

dịch máy bằng nơ-ron. Tạp chí Khoa học Trường Đại học Sư phạm TP.HCM, 19(10), 1749-1760.  

  


