International Journal of TESOL Studies (2024)

Vol. 6 (3) 36-57 https://doi.org/10.58304/ijts.20240304

Article

A Systematic Review of ChatGPT for English as a Foreign Language

Writing: Opportunities, Challenges, and Recommendations

Mark Feng Teng

Macao Polytechnic University, Macau SAR, China

Received: 1 June, 2024/Accepted: 27 June, 2024/Published: 1 July, 2024

Abstract
Since its release in November 2022, ChatGPT has become a focal point of discussion among
scholars and practitioners. This study investigates the role of ChatGPT in English as a Foreign
Language (EFL) writing through a systematic review of the 20 most pertinent articles. Inclusion
and exclusion of references were based on the Preferred Reporting Items for Systematic Reviews
and Meta-Analyses (PRISMA) guidelines. Findings suggest that the integration of artificial
intelligence (AI) into education is an evolving process that significantly impacts how EFL students
learn to write. This systematic review highlights both the opportunities and challenges associated
with adopting ChatGPT for EFL writing. On one hand, ChatGPT offers innovative ways to enhance
writing skills, providing instant feedback and diverse writing prompts. On the other hand,
challenges such as dependency on AI and the need for critical thinking skills remain. Based on the
synthesized literature, recommendations for EFL teaching were proposed, emphasizing a balanced
approach to integrating AI tools like ChatGPT into writing curricula and creating a community of
practice among teachers and students.

Keywords
Artificial intelligence, writing instruction, learning to write, teaching writing, digital education

1. Introduction

As AI technology advances, we can expect AI tools to become even more sophisticated,

assisting writers in generating unique content, improving language fluency, and adapting to
individual writing styles. Ultimately, AI tools have the power to revolutionize writing, making it
more efficient, accurate, and accessible for all (AIContentfy Team, 2023)
The above quote effectively communicates the supportive role of AI in writing. It highlights that AI is
meant to assist and improve the writing process. This perspective is important as it addresses potential
concerns about AI taking over educational roles and emphasizes the collaborative relationship between
technology and educators. The integration of ChatGPT into academia has become a highly debated topic

Email: [email protected]
Mark Feng Teng 37

in the writing domain, particularly due to its notable features that enhance academic writing in foreign
languages (Barrot, 2023). This article aims to explore the potential uses of ChatGPT in EFL writing
and the future of teaching writing skills for foreign language learners, with a focus on a collaborative
relationship between technology and educators.
ChatGPT transforms the education sector. Since its launch in November 2022, ChatGPT developed
by OpenAI has achieved remarkable success. By January 2023, it became one of the fastest-growing and
most widely accepted artificial intelligence (AI) tools, amassing over 100 million active users within just
two months. The underlying technology, Generative Pretrained Transformer (GPT), has revolutionized
traditional learning and writing practices. ChatGPT, an open-access and sophisticated chatbot powered
by GPT technology, has significantly transformed language teaching and learning (Kohnke et al., 2023).
The latest version, ChatGPT-4o, has further advanced these capabilities. ChatGPT processes language
and functions as a personal conversational interface, which warrants an examination of how it can
support instruction and enhance students’ AI literacy skills.
AI literacy, which requires new conceptualization, can be interpreted as the competencies needed
to interact effectively with both humans and machines, including the knowledge and skills that enable
humans to critically understand, evaluate, and use AI systems and tools to safely and ethically participate
in an increasingly digital world (Long & Magerko, 2020). In this new educational landscape, students
must continually strengthen their digital skills to keep pace with rapid technological advancements and
prepare for an unpredictable future. AI tools in 21st-century learning environments will equip students
to critically analyze AI-generated outputs and become technologically literate in an increasingly tech-
centric world. AI can thus enhance professional development by offering real-time feedback and data-
driven insights, empowering educators to continually improve their pedagogical practices in teaching
writing in a foreign language.
Familiarizing students with AI tools like ChatGPT has significant real-world relevance beyond
the classroom. ChatGPT has the potential to transform the preparation of writing teachers, enabling
them to effectively address the evolving needs of 21st-century learners in building digital literacy and
writing skills in a target foreign language with limited language input. However, integrating ChatGPT
into writing instruction presents challenges, particularly regarding robust data privacy. The collection
and analysis of data by AI tools, including ChatGPT, pose risks, and over-reliance on technology may
also diminish the importance of human interaction and mentorship in learning to write (Barrot, 2023).
Additionally, the digital divide and unequal access to technology can exacerbate educational disparities
(Akgun & Greenhow, 2022).
Celik et al. (2022) conducted a comprehensive analysis to investigate the potential and challenges of
AI in teaching practices from the perspectives of teachers and their roles in AI-based education research.
Their findings highlighted AI’s potential to enhance teaching planning, implementation, and assessment
while also emphasizing challenges related to technical capacity, reliability, and the necessity for teacher
involvement in AI development. In the context of English language teaching, Kostka and Toncelli (2023)
explored the opportunities, such as the accelerated innovation in English teaching due to the emergence
of ChatGPT, the challenges, including concerns over potential student misuse for academic work, and
recommendations like fostering greater collaboration among teachers, staff, and stakeholders, associated
with AI integration. While ChatGPT offers ample opportunities for innovation, widespread concerns
about its use in educational settings persist, particularly regarding academic integrity (Teng, 2023).
The present study aims to contribute to the role of ChatGPT in teaching and learning English writing
by presenting a comprehensive overview of opportunities, challenges, and recommendations. It will
also examine gaps and unexplored areas, such as the adaptation of ChatGPT as a writing assistant for
learners and teachers, assessment scenarios, pedagogical roles, and ethical considerations. This article
aims to provide an overview of the current state of scholarship in teaching EFL writing with ChatGPT.
It highlights opportunities, challenges, and recommendations of using ChatGPT for teaching writing,
contributing to the ongoing discourse on AI in foreign language writing instruction.
38 International Journal of TESOL Studies 6 (3)

2. Conceptualizations of AI Literacy and Digitalized Writing

2.1 A conceptualization of AI literacy

In the past five years, there has been a growing need to define AI literacy due to the increasingly
pervasive presence of AI technologies in education. Long and Magerko (2020) defined AI literacy
as a set of competencies that enable individuals to critically evaluate AI technologies, communicate
and collaborate effectively with AI, and use AI as a tool online, at home, and in the workplace. Their
framework categorizes AI literacy into five main components: understanding what AI is, recognizing
what AI can do, comprehending how AI works, determining how AI should be used, and understanding
how people perceive AI.
Ridley and Pawlick-Potts (2021) offered another definition focused on algorithmic literacy, which
includes the skills, expertise, and awareness needed to understand and reason about algorithms, recognize
and interpret their use in systems, create and apply algorithmic techniques to various problems, and
assess the influence of algorithms in social, cultural, economic, and political contexts. This definition
positions individuals as active participants in algorithmic decision-making processes.
The capabilities of ChatGPT and other generative AI tools, which can generate significant bodies
of content from short prompts, have shifted concepts of what AI literacy refers to. Using Long and
Magerko’s (2020) framework, Zhao et al. (2024) updated the notion of AI literacy to include generative
AI literacy, which might be defined through several dimensions: pragmatic understanding, safety
understanding, reflective understanding, socio-ethical understanding, and contextual understanding.
Based on these evolving understandings, there is a need to update the notion of AI literacy specifically
for writing. This updated literacy should enable individuals to effectively interpret information, pick the
right tool for the task, use the chosen tool effectively for specific tasks, and make informed decisions
about where in the writing process to use AI tools (e.g., for search, brainstorming, structuring text,
prompt engineering). It should also encompass defining what sort of answer is required, rephrasing
questions, asking for sources used, iterating and synthesizing results, and updating knowledge as tools
develop rapidly. Moreover, critical thinking on generative AI outputs is essential to ensure information
accuracy, currency, citeability, and to mitigate privacy risks. Social and cultural awareness of AI must
also be integrated into this updated literacy to address the broader implications of AI technologies. By
developing a comprehensive understanding of AI literacy in the context of generative AI, individuals can
navigate the complexities of digitalized writing more effectively and responsibly.

2.2 A conceptualization of digitalized writing

The advent of generative AI tools, such as ChatGPT, has significantly transformed the landscape of
digitalized writing. Digitalized writing represents a major shift in the process of writing, which remains
a fundamental mode of expression and learning (Strobl et al., 2019). Despite its evolution, digitalized
writing retains the cultural value and status of conventional writing, albeit with notable differences.
In the context of EFL teaching, digitalized writing is increasingly characterized by the integration of
multimedia elements and interactive platforms. This trend facilitates a more engaging and dynamic
learning experience for students. Moreover, digitalized writing tools enable real-time feedback and
collaborative writing practices, allowing students to receive immediate corrections and suggestions,
which can significantly improve their writing proficiency.
Traditional writing processes, which consist of stages such as prewriting, planning, drafting, revising,
and editing (Strobl et al., 2019), still apply to digitalized writing. However, digitalized writing is different
from traditional writing. Digitalized writing leverages advanced technologies to offer real-time feedback,
multimedia integration, and collaborative features, making it more interactive and efficient compared to
Mark Feng Teng 39

traditional writing. Traditional writing, on the other hand, relies on physical materials, manual editing,
and lacks the dynamic capabilities of digital tools. The process of writing has been subject to innovation,
evolving from the introduction of word processors to spelling, grammar, and style checkers, connectivity,
and now, generative AI tools. This digitization process has profound effects on writing, which are
often difficult to pinpoint because they are primarily mental rather than directly visible (Kruse &
Rapp, 2019). The acceleration of this digitization process is evident with the emergence of AI-powered
writing assistants like Grammarly and Automatic Writing Evaluation (AWE) tools, as well as other
applications such as QuillBot. Additionally, tools for search, text summarization (e.g., Scholarcy, Iris.ai,
SummarizeBot), and literature reviewing (e.g., ResearchRabbit, Gecko, Connected Papers) have become
integral to the writing process. Tools for referencing, which link search and reading to producing a final
reference list for completed assignments (e.g., EndNote, RefWorks, Zotero, and Mendeley), further
illustrate this integration. Tools like AWE and QuillBot are predominantly used in the stages of revising
and editing. Rephrasing tools, such as Wordtune, are employed both to improve text during the revision
stage and to overcome mental blocks in the early stages of writing or even during planning (Zhao et al.,
2024). Translation tools may also be utilized at various stages, including processing reading materials
and drafting text.
Writing, whether traditional or digitalized, remains a complex and iterative process. The complexity
of digitalized writing is further compounded by the use of multiple digital tools at different stages of
the writing process. The dramatic debut of ChatGPT has added a new dimension to this landscape,
making generative AI an integral component of digitalized writing. The complexity of digitalized writing
is evident in the challenges it presents, such as the need for technical proficiency and digital literacy,
concerns over data privacy and security, the risk of over-reliance on technology, and the digital divide
that can exacerbate educational disparities.
However, these challenges are accompanied by significant opportunities. Digital tools enhance
writing efficiency by automating tasks like grammar checking and reference management, improve
writing quality through real-time feedback, and facilitate the writing process by assisting with planning,
drafting, revising, and editing. The integration of generative AI encourages innovation in writing
practices, allowing for experimentation with different styles and formats. Looking to the future,
digitalized writing will likely be an important element of AI literacy.

3. A Systematic Review of ChatGPT for EFL Writing: Methods and Findings

Despite the growing interest in ChatGPT for EFL writing and the call for a synthesis of the existing
scholarship in order to move the field forward, there is not yet a review paper that brings together the
findings of extant investigations and discusses their implications. To fill this gap, the current study aims
to synthesize research to understand EFL learners’ perceptions on ChatGPT in writing and the role of
ChatGPT in EFL writing, guided by the following research questions:
1. How do EFL learners perceive ChatGPT in writing?
2. What is the role of ChatGPT in EFL writing?

3.1 Methods

The review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses
(PRISMA) guidelines (Figure 1). Following the PRISMA guidelines, which are widely used in
educational research projects (Imran & Almusharraf, 2023), the review process was structured to ensure
comprehensiveness and reliability.
40 International Journal of TESOL Studies 6 (3)

Figure 1
PRISMA Flow Diagram for the Systematic Review

The first step, identification, involved a comprehensive search to locate all pertinent studies on
ChatGPT in EFL writing that were published between 2020 and 2024. To obtain relevant studies on this
topic, three academic databases that are commonly used in applied linguistics were consulted: Web of
Science, Scopus, and Google Scholar. Combinations of the following keywords were utilized: ChatGPT
and AI in EFL writing, ChatGPT as a writing assistant, ChatGPT and the future of writing, and AI and
scientific writing. The keywords were developed in line with the topic and research questions related
to AI literature. The initial search resulted in 343 publications. To mitigate the risk of overlooking
significant works, a manual search was also performed by examining the references of the identified
studies. After removing duplicate records, the total number of studies for potential inclusion was
narrowed down to 180.
The second step, screening, was to formulate the inclusion and exclusion criteria to screen out the
located publications. Five criteria must be met for a study to be included: (a) the study is informed by
the role of ChatGPT in EFL writing, (b) the study contains empirical evidence based on original data
collection, (c)) the study is conducted in EFL writing contexts, (d) the journals are included in Scopus or
ESCI, and (e) the study is written in English. A study was excluded if it (a) does not contain empirical
data (e.g., conceptual papers), (b) does not investigate EFL student writers, (c) is not conducted in EFL
writing contexts, (d) not included in Scopus or ESCI, or (e) is not written in English.
The third step, eligibility, was to screen out studies that did not meet the eligibility criteria. After a
careful examination of the abstract, literature review, and method section of the articles identified, a total
of 40 primary studies were found to be eligible for current review as per our selection criteria. Language
or education related journals are the main source journals included in this review.
The fourth step, inclusion, involved the inclusion of the selected documents in the systematic review.
To reflect the latest findings, only 20 of the most relevant documents published in the years of 2023 and
Mark Feng Teng 41

2024 were selected, while others were excluded. The next step was to code the studies selected for analysis.
Based on the research questions, relevant information about the studies including identification (i.e.,
authors, year, publishing venues), research contexts, participants, and research design were extracted.
Overall, following the PRISMA framework, this systematic review ensured that the research was
conducted in a structured and methodical manner. The selected studies offer a comprehensive overview
of the current state of ChatGPT in writing, highlighting their opportunities and challenges. This review
underscores the importance of effectively leveraging ChatGPT for EFL writing.

3.2 Findings

This section examines the background information of these selected documents, including the distribution
of countries based on the origin of the first author (Figure 2) and journals and citations (Table 1). Figure
2 reveals that scholars in China produced more articles than those in other countries on the topic of
ChatGPT in EFL writing. Japan and the USA also significantly contributed to research on ChatGPT as
a tool for EFL writing. Authors from various countries highlighted that ChatGPT’s role as a writing tool
for EFL students has garnered global attention.

Figure 2
Distribution of Selected Articles Based on Their Origin of the First Author

The given collection of articles in Table 1 offers a comprehensive overview of recent research on the use
of ChatGPT in EFL writing instruction. The studies span a range of topics, from automated essay scoring
to learner perceptions and the impact on writing skills and writing efficacy. Notably, Mizumoto and
Eguchi’s (2023) and Yan’s (2023) work on automated essay scoring has garnered significant attention,
with 140 and 189 citations, highlighting its influence in the field. Even more recent studies, particularly
those published in 2024 (e.g., Steiss et al., 2024), have accumulate citations, reflecting the attention to
this topic of ChatGPT in EFL writing.
Several studies explore practical applications and the perceived benefits of ChatGPT. For instance,
Allen and Mizumoto (2024) investigated Japanese EFL learners’ preferences for editing strategies, while
Teng (2024) examined learners’ perceptions of using ChatGPT for feedback, both indicating a trend
towards integrating AI tools for EFL writing. The potential for ChatGPT to enhance writing instruction is
a recurring theme. Ghafouri et al. (2024) and Song and Song (2023) discuss its role in improving writing
skills and motivation, suggesting positive outcomes for both teachers and learners. However, some
studies also address challenges and limitations. Tsai et al. (2024) highlight pitfalls and feasibility issues,
underscoring the need for cautious implementation.
42 International Journal of TESOL Studies 6 (3)

Table 1
Documents Reviewed Based on Journals, Titles, Citations and the Main Findings
Author(s) and Article title Journal Citations Main focus Main findings
year (up to
June
2024)
1. Allen, T., & ChatGPT RELC 0 Japanese Students mostly preferred
Mizumoto, A. over my EFL students’using AI technology for
(2024). friends: experiences editing and proofreading,
Japanese and as it provided effective
EFL learners’ perceptions feedback that improved
preferences in using the clarity and cohesion
for editing ChatGPT for of their writing. Students
and writing also expressed a preference
proofreading for specific prompts to
strategies enhance their writing.
ChatGPT served a dual role
by offering authoritative
insights into students’
writing and delivering
feedback that learners could
comprehend, similar to
the peer review process in
writing groups.
2. Al-Garaady, ChatGPT’s Arab World 16 Potential ChatGPT successfully
J., & Mahyoob, capabilities English benefits and identified most surface-
M. (2023) in spotting Journal challenges of level errors but could not
and analyzing integrating detect writing errors related
writing errors Artificial to deep structure and
experienced Intelligence pragmatics. Conversely,
by EFL (AI) into human teachers could
learners. EFL writing spot most of these issues.
education ChatGPT cannot replace
human instructors’ expertise
and nuanced understanding
in detecting errors related to
the more complex aspects
of writing.
3. Boudouaia, A study on Journal of 3 The use and The results showed that
A., Mouas, S., ChatGPT-4 as Educational acceptance of the experimental group
& Kouider, B. an innovative Computing ChatGPT-4 in (EG) of using ChatGPT
(2024) approach to Research students’ EFL outperformed the control
enhancing writing. group. The findings also
English as revealed substantial
a foreign improvements in the
language EG’s views of perceived
writing usefulness, perceived
learning. ease of use, attitudes, and
behavioral intention.
Mark Feng Teng 43

4. Bucol, J. L., Exploring Innovations 0 How well ChatGPT displays

& Sangkawong, ChatGPT in Education ChatGPT substantial promise as an
N. (2024). as a writing and Teaching can assess AWE tool, offering distinct
assessment International students’ features such as human-
tool writing using like interface, consistency,
prompts and efficiency, and scalability.
pre-defined
rubrics
compared to
human raters.
5. Ghafouri, M., From virtual Language 0 Using Results showed a significant
Hassaskhah, assistant Teaching ChatGPT effect of ChatGPT for
J., & Mahdavi- to writing Research. for a 10- teacher self-efficacy.
Zafarghandi, A. mentor: week writing In addition, learners
(2024) Exploring instruction demonstrated significant
the impact of program. improvement in writing
a ChatGPT- skills.
based writing
instruction
protocol on
EFL teachers’
self-efficacy
and learners’
writing skill.
6. Gozali et al. ChatGPT as JALT CALL 0 Compare The AWE tools
(2024). an Automated ChatGPT and complemented each other
Writing Grammarly in supporting
Evaluation and Quillbot almost all aspects of
(AWE) Tool: for EFL students’ feedback literacy,
Feedback feedback with “feedback processing”
Literacy literacy being the aspect that
Development ChatGPT could potentially
and AWE enhance or diminish,
Tools’ contingent upon students’
Integration feedback-seeking behavior.
Framework..
7. Guo, K., To resist it or Education and 42 Examining the ChatGPT produced a
& Wang, D. to embrace Information possible role significantly larger amount
(2024). it? Examining Technologies of ChatGPT of feedback than teachers
ChatGPT’s in facilitating and that compared with
potential the teaching teacher feedback, which
to support and learning mainly focused on content-
teacher of writing related and language-related
feedback in English as issues, ChatGPT distributed
EFL writing a Foreign its attention relatively
Language equally among the three
(EFL) feedback foci (i.e., content,
organisation, and language).
44 International Journal of TESOL Studies 6 (3)

8. Mahapatra, S. Impact of Smart 7 Applying The affordance and

(2024) ChatGPT on Learning ChatGPT’s potential applicability of the
ESL students’ Environments text generation tool in L2 writing pedagogy.
academic feature in a Additionally, the tool also
writing skills: one-week showcased an automatic
a mixed L2 writing workflow that could
methods practicum. maximize the efficiency in
intervention composing writing.
study.
9. Mizumoto, Exploring Research 140 The reliabilityAES using GPT
A., & Eguchi, the potential Methods and accuracy demonstrated notable
M. (2023) of using an in Applied of automated accuracy and reliability.
AI language Linguistics essay scoring Leveraging linguistic
model for (AES) for features could further
automated writing enhance scoring accuracy.
essay scoring. assessment ChatGPT has the capability
to provide automated
corrective feedback, making
it a valuable resource for
classroom writing teachers.
10. Mizumoto, Testing the Research 0 ChatGPT’s There is a strong correlation
A., Shintani, viability of Methods viability as a between assessment by
N., Sasaki, M., ChatGPT as in Applied companion ChatGPT and human
& Teng, M. F. a companion Linguistics. for L2 writing accuracy ratings, with
(2024). in L2 writing accuracy a correlation coefficient
accuracy assessment. (ρ) of 0.79, surpassing
assessment. Grammarly’s correlation of
0.69. This strong correlation
affirms the potential of
AI in accurately reflecting
linguistic accuracy
assessment. ChatGPT
is precise in automated
assessments, making it a
suitable companion for L2
writing.
11. Pfau, A., Exploring Research 6 ChatGPT’s There is a strong correlation
Polio, C., & Xu, the potential Methods potential for (ρ = 0.97 using one
Y. (2023). of ChatGPT in Applied measuring method and .94 using
in assessing Linguistics. linguistic another method) between
L2 writing accuracy ChatGPT’s error detection
accuracy in second and human coding. This
for research language correlation diminishes with
purposes. writing lower proficiency levels
Mark Feng Teng 45

12. Punar Cultivating Smart 3 The impact of Students found ChatGPT

Özçelik, N., & writing skills: Learning ChatGPT on beneficial for acquiring
Yangın Ekşi, G. the role of Environments the acquisition
formal register knowledge
(2024) ChatGPT as of register but perceived it as
a learning knowledge unnecessary for informal
assistant—a across variouswriting. Additionally, the
case study. writing tasks.effectiveness of ChatGPT
in teaching neutral register
was questioned by the
participants
13. Shin, D., & Exploratory Education and 0 The potential There was a strong
Lee, J. H. (2024) study on the Information of ChatGPT as similarity between human
potential of Technologies an automated rater and ChatGPT scores.
ChatGPT writing However, ChatGPT showed
as a rater evaluation a slightly greater deviation
of second tool for L2 from the model than its
language assessment human counterparts.
writing.
14. Song, C., & Enhancing Frontiers in 34 The role of There were significant
Song, Y. (2023) academic Psychology ChatGPT in improvements in both
writing enhancing writing skills and
skills and writing for motivation among students
motivation: EFL students who received AI-assisted
assessing the instruction compared
efficacy of to the control group.
ChatGPT in The experimental group
AI-assisted demonstrated enhanced
language proficiency in various
learning for aspects of writing, including
EFL students organization, coherence,
grammar, and vocabulary.
15. Steiss et al. Comparing Learning and 16 The ability of Human raters were better
(2024) the quality of Instruction generative AI at providing high-quality
human and (ChatGPT) feedback to students in
ChatGPT to provide all categories other than
feedback formative criteria-based. AI and
of students’ feedback. humans showed differences
writing in feedback quality based
on essay quality.
16. Su, Y., Lin, Collaborating Assessing 107 Integrating ChatGPT is helpful for
Y., & Lai, C. with Writing ChatGPT into tasks such as outline
(2023) ChatGPT in argumentative preparation, content
argumentative writing revision, proofreading, and
writing classrooms as post-writing reflection.
classrooms a promising ChatGPT supported pre-
solution. writing, during-writing and
post-writing.
46 International Journal of TESOL Studies 6 (3)

17. Teng, M. F. “ChatGPT Computers 0 Chinese AI assistance had a

(2024) is the & Education: students’ significant effect on
companion, Artificial perceptions writing, including
not enemies”: Intelligence and writing motivation, self-
EFL learners’ experiences efficacy, engagement,
perceptions in utilizing and collaborative writing
and ChatGPT for tendency
experiences their writing
in using process
ChatGPT for
feedback in
writing
18. Tsai, C. Impacts of Education and 0 The impacts ChatGPT-assisted revised
Y., Lin, Y. T., ChatGPT- Information of using essays shifted the score
& Brown, I. K. assisted Technologies ChatGPT to curve from a normal
(2024). writing for assist English distribution to a skewed
EFL English as a foreign distribution towards higher
majors: language grades, with the greatest
Feasibility (EFL) increase in revision scores
and students’ seen among students who
challenges. writing had lower original scores.
19. Woo, D. J., Teaching Education and 1 EFL students’ ChatGPT has a potential
Wang, D., Guo, EFL students Information experiences to engage EFL students in
K., & Susanto, to write with Technologies and the writing classroom, but
H. (2024). ChatGPT: perceptions its use can impose heavy
Students’ as they learn cognitive demands.
motivation to write a
to learn, composition
cognitive with
load, and ChatGPT’s
satisfaction support
with the
learning
process
20. Yan, D. Impact of Education and 189 The impact of ChatGPT has a significant
(2023) ChatGPT on Information ChatGPT on effect on students’ academic
learners in a Technologies writing writing skills, and students’
L2 writing perceptions of the impact
practicum: were also overwhelmingly
An positive.
exploratory
investigation.

Table 1 also highlights the diverse applications and impacts of ChatGPT in EFL writing instruction. The
studies consistently demonstrate the potential of ChatGPT to enhance writing skills, provide accurate
assessments, and offer valuable feedback. For instance, Mizumoto and Eguchi (2023) found that
automated essay scoring (AES) using GPT is both accurate and reliable, while Mizumoto et al. (2024)
revealed a strong correlation between ChatGPT’s assessments and human ratings, indicating ChatGPT’s
precision in linguistic accuracy assessment.
Mark Feng Teng 47

Several studies focus on learners’ preferences and perceptions. Allen and Mizumoto (2024) noted
that Japanese EFL learners preferred using ChatGPT for editing and proofreading, as it improved clarity
and cohesion in their writing. Similarly, Teng (2024) highlighted that AI assistance significantly boosted
writing motivation and self-efficacy among Chinese students.
The impact of ChatGPT on teaching efficacy and learner outcomes is also evident. Ghafouri et
al. (2024) showed that a 10-week ChatGPT-based writing instruction program significantly enhanced
teacher self-efficacy and student writing skills. Gozali et al. (2024) emphasized the importance of
ChatGPT in EFL students’ feedback behaviors.
Other studies explored the broader educational implications of ChatGPT. Yan (2023) and Guo and
Wang (2024) examined ChatGPT’s role in providing extensive and balanced feedback, surpassing human
feedback in some areas. However, some challenges remain, such as ChatGPT’s limitations in detecting
deep structure and pragmatic errors (Al-Garaady & Mahyoob, 2023) and its potential to impose heavy
cognitive demands on learners (Woo et al., 2024).
There were four main themes when comes to an analysis of the included literature.

Theme 1: EFL Students’ Positive Preferences and Experiences with ChatGPT

The integration of ChatGPT into EFL writing has garnered positive feedback from students, who
appreciate its effectiveness in enhancing their writing skills.
Allen and Mizumoto (2024) conducted a study focusing on Japanese EFL students’ preferences for
using ChatGPT for editing and proofreading their written work. The research revealed that students
highly favored ChatGPT due to its ability to provide effective feedback that significantly improved the
clarity and cohesion of their writing. The AI offered authoritative insights and feedback that were easy
for learners to understand, mirroring the benefits of peer review processes. This ease of comprehension
and the quality of feedback made ChatGPT a preferred tool among students for refining their writing.
In a similar vein, Boudouaia, Mouas, and Kouider (2024) investigated the use and acceptance of
ChatGPT-4 in EFL writing classes. Their study included an experimental group that utilized ChatGPT
and a control group that did not. The findings were striking: the experimental group outperformed the
control group, demonstrating significant improvements in several areas. These included perceived
usefulness, ease of use, attitudes towards the technology, and behavioral intention to continue using it.
The results suggest that ChatGPT not only enhances writing performance but also positively influences
students’ attitudes and intentions regarding its use.
Teng (2024) further explored the impact of ChatGPT on learners’ writing perceptions, uncovering
several key benefits. The study highlighted that ChatGPT had a significant effect on learners’ writing
motivation, self-efficacy, engagement, and collaborative writing tendencies. Students reported feeling
more motivated to write and more confident in their writing abilities when using ChatGPT. Additionally,
the AI tool fostered greater engagement with writing tasks and encouraged collaborative efforts, as
students were more inclined to share and discuss their work with peers.
Overall, EFL students have shown a strong preference for using ChatGPT in their writing practices,
driven by the effective feedback and user-friendly interface it provides. The studies by Allen and
Mizumoto (2024), Boudouaia, Mouas, and Kouider (2024), and Teng (2024) collectively underscore the
positive impact of ChatGPT on EFL students’ writing skills and attitudes. By improving clarity, cohesion,
motivation, and engagement, ChatGPT is a valuable tool in the EFL writing landscape, enhancing both
individual and collaborative writing experiences.

Theme 2: ChatGPT: An Innovative Tool for Writing Assessment

The advent of ChatGPT has introduced a new paradigm in the field of writing assessment,
showcasing its potential as a powerful tool for evaluating and improving written work. Al-Garaady and
Mahyoob (2023) conducted an in-depth analysis of ChatGPT’s ability to spot and analyze writing errors.
48 International Journal of TESOL Studies 6 (3)

Their research revealed that ChatGPT excels at identifying surface-level errors such as grammatical
mistakes, spelling errors, and punctuation issues. However, the AI struggled with more complex errors
involving deep structure and pragmatic aspects of writing, areas where human instructors still hold a
significant advantage. This indicates that while ChatGPT can serve as a helpful preliminary tool for error
detection, it may need to be complemented by human oversight for more nuanced writing issues.
In another study, Bucol and Sangkawong (2024) explored the potential of ChatGPT as a writing
assessment tool. They highlighted its promise as an Automated Writing Evaluation (AWE) system,
emphasizing its human-like interface, consistency, efficiency, and scalability. These characteristics
make ChatGPT an attractive option for large-scale writing assessments, where human resources may
be limited. The AI’s ability to provide consistent feedback quickly and at scale positions it as a valuable
asset in educational settings, particularly for formative assessments.
Mizumoto and Eguchi (2023) focused on the reliability and accuracy of automated essay scoring
(AES) using GPT technology. Their findings demonstrated that ChatGPT could provide highly accurate
and reliable automated corrective feedback. This suggests that GPT-based systems can be effectively
integrated into educational environments to assist with essay scoring, reducing the workload on educators
and providing students with timely feedback to improve their writing skills.
Expanding on the application of AI in writing assessment, Mizumoto, Shintani, Sasaki, and Teng
(2024) investigated ChatGPT’s viability as a companion for assessing linguistic accuracy in second
language (L2) writing. Their research affirmed the potential of AI in this area, indicating that ChatGPT
could serve as a valuable tool for enhancing L2 writing accuracy. The AI’s ability to provide detailed
feedback on linguistic errors helps learners refine their language skills more effectively.
Pfau, Polio, and Xu (2023) explored ChatGPT’s potential for measuring linguistic accuracy in second
language writing for research purposes. Their study highlighted the effectiveness of ChatGPT in this role,
noting its capability to provide reliable assessments of linguistic accuracy. However, they also observed
that the correlation between ChatGPT’s assessments and human evaluations diminished with lower
proficiency levels. This suggests that while ChatGPT is a useful tool for linguistic accuracy assessment,
its performance may vary depending on the proficiency level of the writers, and it may be more effective
for intermediate to advanced learners.
ChatGPT represents an innovative tool for writing assessment, offering significant benefits in terms
of efficiency, scalability, and consistency. While it excels at identifying surface-level errors and providing
reliable automated feedback, its limitations in detecting deep structural and pragmatic errors highlight
the continued importance of human oversight. As research continues to explore and refine its capabilities,
ChatGPT is poised to become an integral part of the writing assessment landscape, particularly in
educational settings where it can complement traditional assessment methods.

Theme 3: ChatGPT: A Need for Collaboration between Teachers and Learners

ChatGPT has been found to be particularly helpful for editing and proofreading when systematically
implemented and supervised by teachers. It offers valuable insights into word choice, coherence,
structure, language style, and grammar. For instance, Ghafouri, Hassaskhah, and Mahdavi-Zafarghandi
(2024) explored the impact of a ChatGPT-based writing instruction protocol on EFL teachers’
self-efficacy and learners’ writing skills over a 10-week program. Their study showed significant
improvements in both teacher self-efficacy and learners’ writing skills.
Additionally, Guo and Wang (2024) examined ChatGPT’s potential to support teacher feedback in
EFL writing. They found that ChatGPT produced a significantly larger amount of feedback compared
to teachers, effectively addressing content, organization, and language issues. This highlights the tool’s
capacity to enhance the feedback process, ensuring that learners receive comprehensive and detailed
guidance.
Mark Feng Teng 49

Furthermore, Mahapatra (2024) investigated the impact of ChatGPT on ESL students’ academic
writing skills during a one-week L2 writing practicum. The study underscored ChatGPT’s affordance and
potential applicability in L2 writing pedagogy, demonstrating its ability to support learners in improving
their academic writing skills.
These findings collectively emphasize the importance of a collaborative approach where teachers
systematically integrate and supervise the use of ChatGPT. Such collaboration ensures that learners can
fully benefit from the tool’s capabilities while receiving the necessary guidance and support from their
instructors.

Theme 4: ChatGPT on Writing Improvement

Punar Özçelik and Yangın Ekşi (2024) conducted a case study examining the role of ChatGPT in
cultivating writing skills, particularly in acquiring register knowledge across various writing tasks.
Students found ChatGPT beneficial for formal register knowledge but questioned its effectiveness
for informal and neutral registers. Gozali et al. (2024) revealed that ChatGPT and other AI tools
complemented each other in supporting almost all aspects of students’ feedback literacy. In particular,
ChatGPT could potentially enhance EFL students’ feedback processing.
Overall, the different themes highlight the multifaceted role of ChatGPT in EFL writing education,
from enhancing student preferences and experiences to improving error detection and writing assessment.
They also underscore the pedagogical implications of integrating ChatGPT into writing instruction and
its specific contributions to various aspects of writing improvement. ChatGPT holds significant promise
as a tool for enhancing EFL writing instruction. Its ability to provide effective feedback, support error
detection and writing assessment, and offer valuable insights into writing improvement makes it a
valuable addition to language learning environments. However, the limitations of ChatGPT, particularly
in detecting deep structure and pragmatic errors and its varying effectiveness across different writing
registers, underscore the need for a balanced approach that integrates AI tools with human instruction.
Thoughtful and systematic implementation of ChatGPT, supervised by teachers, can maximize its
benefits while addressing its limitations, ultimately enhancing both teaching and learning experiences in
EFL writing contexts.

4. ChatGPT and Future of Writing

Through examining the literature on the opportunities and challenges posed by advanced AI technologies
and their extensive use in writing, the findings from the systematically reviewed literature underscore that
the development of AI technologies, like ChatGPT, has the potential to significantly impact the future
of writing. ChatGPT is an invaluable assistant for researchers, teachers, and students. It can be utilized
as a collaborative tool to provide personalized feedback, tailor search results, offer writing suggestions,
and enhance overall writing quality. Much like human raters, ChatGPT can identify advanced grammar
and syntax issues, suggest improvements for writing structure, and enhance the coherence of prompts
(Mizutomo et al., 2024). The literature highlights ChatGPT’s ability to significantly aid in various writing
tasks. It excels in generating text, creating initial drafts, brainstorming ideas, and summarizing literature.
These capabilities can greatly assist writers in the early stages of the writing process, facilitating the
development and organization of their thoughts (Boudouaia et al., 2024). By handling repetitive and
time-consuming tasks, ChatGPT allows writers to focus on more creative and complex aspects of their
work. The use of ChatGPT can streamline the writing process, enabling users to produce content more
quickly and efficiently. This is particularly beneficial in EFL settings, for which students lack self-
efficacy (Teng, 2024). By automating certain aspects of writing, ChatGPT helps to increase productivity,
allowing individuals to manage their time more effectively.
50 International Journal of TESOL Studies 6 (3)

Despite its advantages, ChatGPT raises significant concerns regarding ethical use and academic
integrity. There is a risk of over-reliance on AI-generated content, which could undermine the
development of critical thinking and writing skills (Barrot, 2023). Ensuring that AI tools are used
responsibly and ethically is crucial to maintaining the integrity of academic and professional writing.
While ChatGPT can assist with various writing tasks, it cannot fully replace the nuanced skills and
knowledge of a human writer. AI lacks the ability to deeply understand context, interpret complex
emotions, and generate truly original ideas, which are essential components of high-quality writing.
Human writers bring a level of creativity, intuition, and cultural understanding that AI currently cannot
replicate. The ease of generating content with ChatGPT could lead to misuse, such as creating misleading
or plagiarized content, including false citations (Day, 2023). Establishing guidelines and best practices
for the ethical use of AI in writing is necessary to mitigate these risks. It is important to educate users on
the responsible use of AI tools to prevent potential negative consequences.
In this scientifically advanced era, the importance of AI and its technologies is undeniable, as they are
making significant strides in various fields, including writing. The integration of AI tools like ChatGPT
into the EFL writing process presents both opportunities and challenges. On the positive side, AI can
revolutionize writing techniques by enhancing productivity, providing valuable feedback, and supporting
the creation of high-quality content. On the negative side, it poses risks related to ethical considerations,
academic integrity, and the potential erosion of human creativity and critical thinking skills.

4.1 Opportunities

The development of artificial intelligence (AI) has been underway for decades, but it has gained
significant momentum in recent years, as evidenced by the doubling of publications on the topic (Maslej
et al., 2023). The emergence of AI tools like ChatGPT has accelerated innovation in English language
teaching, particularly in EFL contexts. Despite the vast majority of educational systems retaining a
basic structure, the process of innovating English teaching and learning to meet 21st-century needs has
been slow (Kostka & Toncelli, 2023). However, the rapid advancements in EFL teaching and learning
driven by AI tools such as ChatGPT present a critical opportunity for significant shifts in educational
philosophies and practices.
The rapid changes necessitated by AI advancements require educators to re-evaluate and redefine
their roles in EFL writing instruction (Guo & Wang, 2024). AI tools like ChatGPT are prompting much-
needed innovation in assessment and teaching methods (Rudolph et al., 2023). For instance, promising
uses of ChatGPT in writing assessment have been documented (e.g., Bucol & Sangkawong, 2024;
Mizumoto & Eguchi, 2023). ChatGPT can serve as a personal mentor, answering student questions
and summarizing key information for writing (Ghafouri et al., 2024). This allows class time to move
beyond mere knowledge acquisition, focusing instead on applying knowledge to enhance writing
through collaborative and critical engagement (Mahapatra, 2024). AI-generated text can be leveraged as
a starting point for projects or as a comparative tool alongside student-generated work. This approach
encourages writing and strengthens EFL students’ self-efficacy (Song & Song, 2023; Teng, 2024).
Allen and Mizumoto (2024) noted that the new approaches to fostering EFL students’ writing strategies
inspired by ChatGPT may also create space for realigning prior knowledge with current knowledge,
thereby benefiting writing skills. This aligns with the recommendation by Su et al. (2023) to help learners
collaborate with AI using theoretically grounded principles for argumentative writing classrooms.
Mitchell (2019) identified creativity, curiosity, and critical engagement as essential components
of education. If used with critical intentionality, caution, and care, ChatGPT may facilitate innovative
teaching practices in EFL writing contexts. The integration of AI tools like ChatGPT into educational
settings holds the potential to revolutionize traditional writing teaching methods, making them more
interactive, personalized, and effective.
Mark Feng Teng 51

4.2 Challenges

While ChatGPT offers numerous opportunities for innovation in EFL writing instruction, it also brings
widespread concerns and challenges. Since the release of ChatGPT, issues surrounding academic
integrity have been a significant point of contention. Stewart (2023) coined the term “high-tech
plagiarism” from Chomsky to describe the potential misuse of AI in academic settings, while Weismann
(2023) went as far as to suggest that such technologies could signal the end of the traditional teaching
profession. One of the primary fears among educators is that students will rely on ChatGPT to complete
their writing assignments, thereby undermining the process of learning to write. Instances have been
reported where students’ articles were essentially drafted by ChatGPT, raising serious concerns about
the authenticity of student work. Teng (2023) noted that ChatGPT adds a new layer of risk for cheating,
exacerbating existing concerns about academic integrity.
In response to these risks, some institutions initially took drastic measures. For example, the
University of Hong Kong banned the use of ChatGPT when it was first released. Although such
restrictions were later relaxed, the episode underscored the urgent need for new approaches to uphold
academic integrity in the age of AI. Guo and Wang (2024) highlighted the critical question facing
educators: should we resist or embrace ChatGPT for EFL writing instruction?
The debate over ChatGPT is also reflected in media coverage and public discourse. Tlili et al. (2023)
analyzed tweets and interviews with stakeholders, revealing mixed reactions to the potential uses and
misuses of these technologies. Similarly, Song and Song (2023) explored the motivations behind student
cheating, suggesting that generative AI might tempt students who feel they lack control over the writing
process. They conclude that revolutionizing assessments and reflecting on the learning process could
be key to addressing these challenges. García-Peñalvo (2023) observed that hysteria and worry are
typical reactions to disruptive changes. The diverse range of reactions to generative AI may be “a true
reflection of general behavior towards innovation” (p. 2). This mixed response to ChatGPT highlights the
immediate challenges of adapting to technological advancements, yet it also underscores the need for a
balanced approach that considers both the risks and pedagogical benefits.
The push and pull between concerns about cheating and the opportunities presented by ChatGPT
is a central theme in the current discourse. While the potential for misuse is significant, so too are the
educational benefits. ChatGPT can serve as a powerful tool for enhancing student engagement, providing
personalized feedback, and supporting the development of writing skills (Punar Özçelik & Yangın Ekşi,
2024). However, these benefits must be weighed against the risks to integrity in writing.

4.3 Recommendations

The integration of ChatGPT into undergraduate and graduate writing classes, as outlined in Table
2, serves as an example for recommendations to address both promising opportunities and notable
challenges.

Table 2
Recommendations for Writing Classes in a Foreign Language Context
Course title Undergraduate: Persuasive Writing Graduate: Advanced Academic Writing
Lesson Constructing Persuasive Arguments Academic Presentation Skills in Writing
Prior learning - Aristotle’s conceptualization of - Formal writing skills such as structuring
ethos, pathos, and logos in persuasive arguments, incorporating visual aids, writing
argumentation effective introductions and conclusions, and
ensuring transitions and cohesion
52 International Journal of TESOL Studies 6 (3)

Lesson - Analyze ChatGPT-generated - Evaluate ChatGPT output on presentation

Objectives persuasive texts for the presence and skills in academic writing
quality of ethos, pathos, and logos - Compare and contrast ChatGPT output
- Construct and revise persuasive with students’ prior learning in academic
arguments using student-improved writing
texts
Materials - ChatGPT - ChatGPT
- Shared Google - Laptops
- Laptops - Hard copies of Venn diagrams
- Instructor-created prompt questions to
input into ChatGPT
In-class activity - In groups of 3, students craft - In groups of 2-3, students engage in retrieval
ChatGPT prompts to generate either an practice by listing what they remember about
affirmative or negative persuasive text one of the five major presentation skill topics
on a topic selected by the instructor given by the instructor
- Students confirm prompts with the - The instructor gives each group a question
instructor and generate texts, which are related to their topic (e.g., How can I
then pasted into a shared Google Doc engage my audience during an academic
- Groups analyze their texts for ethos, presentation?)
pathos, and logos by adding comments - Students input their question into ChatGPT
to the Google Doc and read its response
- Groups conduct research to improve - Students create a Venn diagram to compare
their texts, including finding credible their knowledge with ChatGPT output,
sources to enhance ethos and adding focusing on similarities and differences
personal stories to enhance pathos
- Groups peer-review each other’s
revised texts
Assessment - Groups exchange texts and provide - Students submit a comparative analysis of
peer feedback, focusing on the their knowledge and ChatGPT’s responses,
effective use of ethos, pathos, and highlighting areas of agreement and
logos in the revised texts divergence, and reflecting on how the activity
- Students write a short reflection on enhanced their academic writing skills
whether the peer review process helped
them improve their persuasive writing

As demonstrated in the above sections, incorporating ChatGPT into English writing classes presents
both opportunities and challenges. Students have shown considerable interest in the capabilities of
ChatGPT and its potential role in enhancing their writing skills. The growing body of literature provides
valuable recommendations for utilizing ChatGPT to support English writing instruction, emphasizing the
importance of collaboration among educators and students.
In the above recommendations for sample writing classes in Table 2, students collaborated to
brainstorm ideas on leveraging ChatGPT, reviewed each other’s lessons and materials, and discussed
preliminary insights from classroom experiences. Such collaborative efforts align with the concept of
collegial inquiry, where teachers and students work together to explore and develop more complex
perspectives, thereby enhancing engagement in writing (Guo & Wang, 2024; Woo et al., 2024). This
collaboration is crucial, particularly in the context of the evolving nature of AI tools like ChatGPT,
Mark Feng Teng 53

which necessitates continuous dialogue and shared learning among educators and students. Teachers and
students need to foster an awareness of ChatGPT’s functionalities for writing class (Table 3).

Table 3
ChatGPT Functionalities for Writing Class
ChatGPT Functionalities Examples Reflection
Content Development - Suggestions for potential topics, Can aid in early writing stages; risk of
outlines, or resources over-reliance for ideation.
- Collection and summarization Helpful for gathering ideas; ensures
of related information critical engagement.
- Facilitation of brainstorming Stimulates thought processes; may
through conversation hinder original idea generation.
Corpus Search - Retrieval of example sentences Enhances practical understanding; must
using the given vocabulary or ensure critical engagement.
grammar
Text Modification - Adjustment of difficulty level Useful for tailoring content; requires
oversight for nuanced judgement.
- Adjustment of genre, tone, Aids contextual appropriateness; needs
voice, or formality careful review.
Feedback and Revision - Editing and proofreading with Accelerates learning with detailed
explanations feedback; risk of surface-level
understanding.
- Revision for specific aspects of Supports targeted improvement;
writing (content, organization, students must understand revisions.
clarity)
- Analysis of learner errors and Identifies common issues; must foster
mistakes deep comprehension of errors.
Response to Questions - Providing answers to specific Clarifies doubts quickly; accuracy and
questions comprehensibility are crucial.

- Appropriateness of word usage, Ensures correct usage; requires

confusing synonyms accurate and clear explanations.
Support for Sentence - Translation or articulation of a Overcomes writer’s block; sentences
Generation given sentence need contextual review.
- Generation of templates or Provides structure; students must learn
sentence starters to adapt templates meaningfully.
Vocabulary/Grammar - Providing definitions or Expands knowledge; explanations
Support explanations for unknown should be clear and contextualized.
vocabulary or grammar items
Paraphrasing/Summary - Paraphrasing and summarizing Teaches critical writing skills; outputs
to meet specific requirements need evaluation to meet academic
standards.
54 International Journal of TESOL Studies 6 (3)

The development of writing lessons within the EFL curriculum reflects the AI literacy framework, which
comprises five essential elements: understanding, accessing, prompting, corroborating, and incorporating
(Tseng & Warschauer, 2023). According to sociocultural theory, the significance of tools lies not in
their abstract characteristics but in how they transform human activity (Vygotsky, 1981). From this
perspective, incorporating tools like AI does not merely facilitate actions that could occur without them;
instead, their inclusion fundamentally alters the entire flow and structure of cognitive functions. Writing
instructors need to recognize that AI is not just a medium for teaching the same content differently;
it is reshaping the entire landscape of language use and learning. The ability to write effectively is,
to some extent, dependent on the tools available. Writing well with AI and writing well without it
are distinct, though overlapping, competencies. This distinction is evident in the analysis of various
writing instruction examples. Recognizing how AI is transforming language use and writing is crucial.
Integrating AI into writing instruction in ways that align with students’ learning goals and proficiency
levels ensures that students become adept at writing both with and without AI. By applying an AI
literacy framework, writing instructors can help EFL students effectively harness the power of AI tools to
enhance their English writing skills.
Guidelines for the use of AI in writing courses can be co-created with students at the beginning of the
course and refined throughout the semester as needed. This iterative process ensures that the guidelines
remain relevant and effective. Over time, teachers and students may develop a community of practice,
a concept introduced by Lave and Wenger (1991), where they co-construct knowledge about ChatGPT
and stay abreast of the latest advancements. Such communities of practice are beneficial for professional
development, as they facilitate knowledge sharing, learning, and adaptation to change (Wenger &
Snyder, 2000). A community of practice centered on the exploration of ChatGPT can help educators and
students feel confident in experimenting with AI applications and addressing associated challenges. This
collaborative environment not only supports the process of learning to write but also raises awareness
of digital literacy among teachers. As Wenger (1998) notes, developing mutual understanding within a
community of practice is essential for fostering a supportive and dynamic learning environment.

5. Concluding Remarks

This systematic review has illuminated the pivotal role of ChatGPT as a tool for enhancing EFL
writing. The introduction of ChatGPT marks the beginning of a new era characterized by unprecedented
advancements in science, technology, and education. The majority of the studies reviewed support
the notion that ChatGPT, as an advanced AI generative model, can significantly aid EFL students in
improving their writing skills.
Given the rapid expansion of AI technology, scholars worldwide are exploring various aspects of
its application and utility. This study provides a thorough overview of literature published during 2023
and 2024, focusing on the role of ChatGPT in EFL writing. The data was systematically extracted and
presented in figures and tables to elucidate its impact on teaching EFL students how to write effectively.
By offering detailed findings and discussions, this study addresses the existing gap in the literature
concerning the opportunities, challenges, and recommendation associated with ChatGPT as a facilitator
and assistant in EFL writing. It underscores ChatGPT’s potential to revolutionize current and future EFL
writing instruction, highlighting both its benefits and areas for further exploration.

References

(* references included for systematic review)

Akgun, S., & Greenhow, C. (2022). Artificial intelligence in education: Addressing ethical challenges in
K-12 settings. AI and Ethics, 2, 431–440. https://doi.org/10.1007/s43681-021-00096-7
Mark Feng Teng 55

*Al-Garaady, J., & Mahyoob, M. (2023). ChatGPT’s capabilities in spotting and analyzing writing
errors experienced by EFL learners. Arab World English Journal, 9, 3-17. DOI: https://dx.doi.
org/10.24093/awej/call9.1
*Allen, T., & Mizumoto, A. (2024). ChatGPT over my friends: Japanese EFL learners’ preferences for
editing and proofreading strategies. RELC Journal.
Barrot, J. S. (2023). Using ChatGPT for second language writing: Pitfalls and potentials. Assessing
Writing, 57, 100745. https://doi.org/10.1016/j.asw.2023.100745
*Boudouaia, A., Mouas, S., & Kouider, B. (2024). A study on ChatGPT-4 as an innovative approach
to enhancing English as a foreign language writing learning. Journal of Educational Computing
Research. https://doi.org/10.1177/0735633124124746
*Bucol, J. L., & Sangkawong, N. (2024). Exploring ChatGPT as a writing assessment tool. Innovations
in Education and Teaching International. https://doi.org/10.1080/14703297.2024.2363901
Celik, I., Dindar, M., Muukkonen, H., & Järvelä, S. (2022). The promises and challenges of artificial
intelligence for teachers: A systematic review of research. Tech Trends, 66(4), 616–630. https://doi.
org/10.1007/s11528-022-00715-y
Day, T. (2023). A preliminary investigation of fake peer-reviewed citations and references generated by
ChatGPT. The Professional Geographer, 75(6), 1024-1027. https://doi.org/10.1080/00330124.2023.
2190373
*Gozali, I., Wijaya, A. R. T., Lie, A., Cahyono, B. Y., & Suryati, N. (2024). ChatGPT as an Automated
Writing Evaluation (AWE) Tool: Feedback Literacy Development and AWE Tools’ Integration
Framework. JALT CALL Journal, 20(1), n1. DOI: https://doi.org/10.29140/jaltcall.v20n1.1200
García-Peñalvo, F.J. (2023). The perception of Artificial Intelligence in educational contexts after the
launch of ChatGPT: Disruption or panic? Education in the Knowledge Society, 24, 1-9. https://doi.
org/10.14201/eks.31279
*Ghafouri, M., Hassaskhah, J., & Mahdavi-Zafarghandi, A. (2024). From virtual assistant to
writing mentor: Exploring the impact of a ChatGPT-based writing instruction protocol on EFL
teachers’ self-efficacy and learners’ writing skill. Language Teaching Research. https://doi.
org/10.1177/1362168824123976
*Guo, K., & Wang, D. (2024). To resist it or to embrace it? Examining ChatGPT’s potential to support
teacher feedback in EFL writing. Education and Information Technologies, 29, 8435–8463. https://
doi.org/10.1007/s10639-023-12146-0
Kohnke, L., Moorhouse, B. L., & Zou, D. (2023). ChatGPT for language teaching and learning. RELC
Journal. https://doi.org/10.1177/00336882231162868
Kostka, I., & Toncelli, R. (2023). Exploring applications of ChatGPT to English language teaching:
Opportunities, challenges, and recommendations. TESL-EJ, 27(3), n3. https://doi.org/10.55593/
ej.27107int
Kruse, O., & Rapp, C. (2019). Seamless writing: How the digitisation of writing transforms thinking,
communication, and student learning. In Looi CK, Wong LH, Glahn C, Cai, S. (eds), Seamless
learning. Lecture notes in educational technology (pp. 191-208). Springer, Singapore. https://doi.
org/10.1007/978-981-13-3071-1_10
Imran, M., & Almusharraf, N. (2023). Analyzing the role of ChatGPT as a writing assistant at higher
education level: A systematic review of the literature. Contemporary Educational Technology,
15(4), ep464. https://doi.org/10.30935/cedtech/13605
Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge
University Press.
Long, D., & Magerko, B. (2020). What is AI literacy? Competencies and design considerations. In
Bernhaupt R, Mueller F, Verweij D, Andres J (eds). Proceedings of the 2020 CHI conference on
human factors in computing systems (pp. 1-16). Association for Computing Machinery.
56 International Journal of TESOL Studies 6 (3)

*Mahapatra, S. (2024). Impact of ChatGPT on ESL students’ academic writing skills: a mixed methods
intervention study. Smart Learning Environments, 11. doi.org/10.1186/s40561-024-00295-9
Maslej, N., Fattorini, L., Brynjolfsson, E., Etchemendy, J., Ligett, K., Lyons, T., Manyika, J., Ngo, H.,
Niebles, J.C., Parli, V., Shoham, Y., Wald, R., Clark, J., & Perrault, R. (2023). The AI index 2023
annual report. Stanford University Institute for Human-Centered AI. https://aiindex.stanford.edu/
wp-content/uploads/2023/04/HAI_AI-Index-Report_2023-pdf
Mitchell, M. (2019). Artificial intelligence: A guide for thinking humans. Farrar, Straus, and Giroux.
*Mizumoto, A., & Eguchi, M. (2023). Exploring the potential of using an AI language model for
automated essay scoring. Research Methods in Applied Linguistics, 2(2), 100050. https://doi.
org/10.1016/j.rmal.2023.100050
*Mizumoto, A., Shintani, N., Sasaki, M., & Teng, M. F. (2024). Testing the viability of ChatGPT as a
companion in L2 writing accuracy assessment. Research Methods in Applied Linguistics. https://
doi.org/10.1016/j.rmal.2024.100116
*Pfau, A., Polio, C., & Xu, Y. (2023). Exploring the potential of ChatGPT in assessing L2 writing
accuracy for research purposes. Research Methods in Applied Linguistics, 2(3), 100083. https://doi.
org/10.1016/j.rmal.2023.100083
*Punar Özçelik, N., & Yangın Ekşi, G. (2024). Cultivating writing skills: the role of ChatGPT as a
learning assistant—a case study. Smart Learning Environments, 11(1), 10. https://doi.org/10.1186/
s40561-024-00296-8
Ridley, M., & Pawlick-Potts, D. (2021). Algorithmic literacy and the role for libraries. Information
Technology Libraries, 40(2). https://doi.org/10.6017/ital.v40i2.12963
Rudolph, J., Tan, S., & Tan, S. (2023). ChatGPT: Bullshit spewer or the end of traditional assessments in
higher education? Journal of Applied Teaching & Learning, 6(1), 342-362. https://doi.org/10.37074/
jalt.2023.6.1.9
*Shin, D., & Lee, J. H. (2024). Exploratory study on the potential of ChatGPT as a rater of second
language writing. Education and Information Technologies. https://doi.org/10.1007/s10639-024-
12817-6
*Song, C., & Song, Y. (2023). Enhancing academic writing skills and motivation: assessing the efficacy
of ChatGPT in AI-assisted language learning for EFL students. Frontiers in Psychology, 14,
1260843. https://doi.org/10.3389/fpsyg.2023.1260843
*Steiss, J., Tate, T., Graham, S., Cruz, J., Hebert, M., Wang, J., ... & Olson, C. B. (2024). Comparing
the quality of human and ChatGPT feedback of students’ writing. Learning and Instruction, 91,
101894. https://doi.org/10.1016/j.learninstruc.2024.101894
Stewart, J. (2023, February 17). Noam Chomsky says ChatGPT is a form of “high-tech plagiarism”. My
Modern Met. https://mymodernmet.com/noam-chomsky-chatgpt/#:~:text=ChatGPT%20is%20
basically%20high%2Dtech,to%20avoid%20doing%20-the%20work
Strobl, C, Ailhaud, E, Benetos, K, Devitt, A, Kruse, O, Proske, A, & Rapp, C. (2019). Digital support for
academic writing: a review of technologies and pedagogies. Computers & Education, 131, 33–48.
https://doi.org/10.1016/j.compedu.2018.12.005
*Su, Y., Lin, Y., & Lai, C. (2023). Collaborating with ChatGPT in argumentative writing classrooms.
Assessing Writing, 57, 100752. https://doi.org/10.1016/j.asw.2023.100752
Teng, M. F. (2023). Scientific writing, reviewing, and editing for open-access TESOL journals: The
role of ChatGPT. International Journal of TESOL Studies, 5, 87-91. https://doi.org/10.58304/
ijts.20230107
*Teng, M. F. (2024). “ChatGPT is the companion, not enemies”: EFL learners’. perceptions and
experiences in using ChatGPT for feedback in writing. Computers & Education: Artificial
Intelligence.
Mark Feng Teng 57

Tlili, A., Shehata, B., Adarkwah, M.A., Bozkurt, A., Hickey, D.T., Huang, R., & Agyemang, B. (2023).
What is the devil is my guardian angel: ChatGPT as a case study of using chatbots in education.
Smart Learning Environments, 10, 1-24. https://doi.org/10.1186/s40561-023-00237-x
*Tsai, C. Y., Lin, Y. T., & Brown, I. K. (2024). Impacts of ChatGPT-assisted writing for EFL English
majors: Feasibility and challenges. Education and Information Technologies. https://doi.org/10.1007/
s10639-024-12722-y
Tseng, W., & Warschauer, M. (2023). AI-writing tools in education: If you can’t beat them, join them.
Journal of China Computer-Assisted Language Learning. 10.1515/jccall-2023-0008
Vygotsky, L. S. (1981). The genesis of higher mental functions. In J.V. Wertsch (Ed.), The concept of
activity in Soviet psychology (pp. 144-188). M.E. Sharpe.
Weissman, J. (2023, February 8). ChatGPT is a plague upon education (opinion). Inside Higher Ed.
https://www.insidehighered.com/views/2023/02/09/chatgpt-plague-uponeducation-opinion
Wenger, E. (1998). Communities of practice: Learning as a social system. Systems Thinker, 9(5), 2-3.
Wenger, E. C., & Snyder, W. M. (2000). Communities of practice: The organizational frontier. Harvard
Business Review, 78(1), 139-45.
*Woo, D. J., Wang, D., Guo, K., & Susanto, H. (2024). Teaching EFL students to write with ChatGPT:
Students’ motivation to learn, cognitive load, and satisfaction with the learning process. Education
and Information Technologies. doi.org/10.1007/s10639-024-12819-4
*Yan, D. (2023). Impact of ChatGPT on learners in a L2 writing practicum: An. exploratory investigation.
Education and Information Technologies, 28(11), 13943-13967. https://doi.org/10.1007/s10639-
023-11742-4
Zhao, X., Cox, A., & Cai, L. (2024). ChatGPT and the digitisation of writing. Humanities and Social
Sciences Communications, 11, 482. https://doi.org/10.1057/s41599-024-02904-x

Mark Feng Teng, Ph.D., is Associate Professor at Macao Polytechnic University. He was the recipient
of the 2017 Best Paper Award from the Hong Kong Association for Applied Linguistics (HAAL),
2023 Best Paper Award in social sciences from Education Ministry in China. His research portfolio
mainly focuses on computer-assisted L2 vocabulary acquisition and L2 writing. His publications have
appeared in international journals, including Applied Linguistics, TESOL Quarterly, Language Teaching
Research, System, Applied Linguistics Review, Computer Assisted Language Learning, Computers &
Education, Foreign Language Annals, and IRAL, among others. His recent monographs were published
by Routledge, Springer, and Bloomsbury. He also edited and co-edited special issues for international
journals, including Journal of Writing Research, Studies in Second Language Learning and Teaching,
and TESOL Journal, etc. Currently he serves as the Chief Editor for International Journal of TESOL
Studies (IJTS).