JEELS
(Journal of English Education and Linguistics Studie.
P-ISSN: 2407-2575 E-ISSN: 2503-2194 https://jurnalfaktarbiyah.
id/index.
php/jeels ARTIFICIAL INTELLIGENCE IN ENGLISH WRITING: ETHICAL INTEGRATION AND EDUCATORSAo PERSPECTIVES *Ratna Yunita1.
Nurma Fitrianna2.
Nashrul Ingqidam Maulana3 1,2,3English Education Department.
Institut Agama Islam Negeri (IAIN) Ponorogo.
Indonesia *ratnayunita@iainponorogo.
nurma@iainponorogo.
nashrulingqidammaulana@gmail.
(*) Corresponding Author Abstract: This study examines the ethical implications of the AI-assisted writing toolsAo integration, such as Grammarly, in English language instruction, based on the perspectives of 30 English lecturers in Indonesia.
The research aims to understand lecturers' views regarding the application of AI in writing courses, especially concerning academic integrity, student autonomy, and fairness.
Using a qualitative approach, open-ended questionnaires and analyzed thematically with the help of NVivo software.
Descriptive statistics such as percentages were used to support the presentation of qualitative themes.
The results suggest that numerous lecturers recognize the advances of AI toolsAisuch as real-time feedback, improved writing accuracy, and increased efficiencyAi concerns persist about excessive dependence on AI, erosion of critical thinking skills, and algorithmic bias.
The study also reveals a lack of Citation in APA style:
Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
DOI: 10.
30762/jeels.
Submission: April 2025.
Revision: May 2025.
Publication: June 2025 Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
institutional guidelines to support ethical AI integration.
Lecturers emphasized the need for balanced use, combining AI tools with traditional teaching methods, alongside training and clear policies to guide ethical These revelations help create practical rules for the ethical incorporation of AI in writing settings in higher education.
Keywords: AI-assisted learning.
Artificial Intelligence.
English writing instruction, ethical considerations, higher
INTRODUCTION
A paradigm shift in teaching and learning is being brought by Artificial Intelligence (AI) integration in educational environments, particularly in English language writing classes.
The introduction of AI technologies for English writing tasks has prompted substantial ethical questions as these systems become increasingly sophisticated.
It is generally agreed that artificial intelligence has the potential to improve learning outcomes, particularly because of its ability to offer feedback in real time and to promote tailored learning practices (Luckin & Holmes, 2.
Numerous studies have highlighted how AI-enhanced platforms contribute significantly to improving writing proficiency, offering personalized learning experiences and scaffolding support (Alharbi & Al-Hoorie, 2020.
Hooda et al.
, 2.
These benefits are echoed in classroom experiences globally, where AI is often seen as a powerful supplement to traditional writing However, ethical challenges such as plagiarism, authorship ambiguity, and the dilution of critical thinking have also emerged as pressing concerns (Baker & Smith, 2.
, reinforcing the need for responsible AI integration in writing classrooms.
Concerns are amplified by the increasing adoption of generative AI tools like ChatGPT.
QuillBot, and Grammarly, which can support but also potentially undermine students' independent writing skills (Fontanilla et al.
, 2.
As Selwyn .
argues, the rise of AI in educational systems must be matched with critical reflection on its Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
societal and institutional implications.
This is of particular importance in Indonesia, since the application of AI in educational environments is still in its infant stages at this time.
While tools like Grammarly and Turnitin are beginning to be adopted, institutional policies and ethical frameworks for AI use remain unclear.
At institutions like Institut Agama Islam Negeri (IAIN) Ponorogo, for example, there is still no consistent or formal guidance on evaluating AI-assisted student writing.
contrast, institutions like the Open University in Indonesia have begun to establish more structured protocols, though ambiguity remains around AIAos role in potential plagiarism and authorship attribution (Jamil et al.
, 2.
These patterns mirror a broader global universitiesAiparticularly research-intensive institutionsAiare actively developing GenAI-related policies and classroom guidelines, often positioning GenAI tools as both a challenge and an opportunity for innovation in pedagogy (McDonald et al.
, 2.
Recent literature emphasizes that AI can both enhance and complicate writing instruction.
Studies have explored the dual nature of AI tools: while they support content generation, feedback, and language accuracy, they may also reduce critical thinking and foster dependency (Rentier, 2024.
Gustilo et al.
, 2024.
Hossain, 2.
For example.
Han and Li .
argue that AI tools can democratize education by supporting students who struggle with conventional writing approaches, while Lee .
highlights their utility in demonstrating grammatical conventions and structure.
However, as Ersyz and Engin .
and Rentier .
point out, the use of generative AI raises concerns around authorship, data accuracy, privacy, and the ethical responsibility to acknowledge machinegenerated content.
EducatorsAo perceptions are especially important, as they are the mediators between AI technologies and pedagogical Research shows that educators tend to value the potential of AI tools but remain cautious due to limited training, unclear institutional policies, and ethical grey areas (Gustilo et al.
Fontanilla et al.
, 2023.
Hossain, 2.
These perceptions are also Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
shaped by teacher identity, experience, and access to professional development, particularly among senior educators (Jamil et al.
, 2.
Tools like Padlet and Slack have also been explored in Indonesia as collaborative aids in writing instruction, but their integration has not been linked directly to broader AI ethics frameworks (Rinanda et al.
Even though research on artificial intelligence in education is growing, there remains a significant gap in understanding how university lecturers specifically perceive and respond to the ethical dimensions of AI-assisted writing tools across diverse cultural and institutional contexts.
This study addresses that gap by focusing on lecturers at IAIN Ponorogo and other Indonesian institutions, where discussions surrounding AI in academic writing are still in the early While prior studies have examined educators' perceptions more broadly, this research narrows its focus to university lecturers, whose views are crucial in shaping the ethical and pedagogical integration of AI in higher education writing instruction.
concentrating on this group, the study captures in-depth insights from those directly responsible for developing studentsAo academic writing competencies in tertiary education.
In summary, this study seeks to address two key objectives:
To reveal how lecturers feel about the use of AI-assisted writing software in English classes.
To investigate the ethical factors that drive these perceptions.
METHOD
This study employed a qualitative research design to explore university lecturersAo perspectives on the integration of AI-assisted writing tools in English language instruction.
A qualitative approach was selected to provide a rich and in-depth understanding of the participantsAo experiences, ethical concerns, and classroom practices.
Although some numerical summaries are presented .
, these serve only to enhance the presentation of recurring themes and do not constitute a separate quantitative Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
Participants This study involved 30 English lecturers from a range of higher education institutions across Indonesia.
They were selected based on their experience teaching English writing or other relevant English language courses, as outlined in the research invitation.
The aim was to ensure a diverse sample in terms of geographic location, institutional type .
ublic and privat.
, and teaching experience, particularly those who are actively involved or potentially affected by the integration of artificial intelligence in writing instruction.
Table 1.
Participant Demographics and Institutional Affiliation Participant ID
Gender
P10
P11
P12
P13
P14
P15
P16
P17
P18
P19
P20
P21
P22
P23
P24
P25
P26
P27
P28
P29
P30
Male Female Female Female Female Female Male Female Female Male Female Female Female Female Female Female Female Female Female Male Female Female Female Male Female Female Female Female Female Male Estimated Age Years of Experience University University A University B University C University D University E University F University G University G University H University I University H University H University I University J University K University L University M University N University H University O University P University Q University H University R University S University T University U University V University H University H Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
In terms of geographic representation, 28 lecturers were from universities located on the island of Java.
These institutions were spread across several provinces, including East Java.
Central Java.
Yogyakarta.
Jakarta, and West Java.
Each province was represented by at least one university, and a mix of both public and private institutions were included.
To ensure wider representation beyond Java, 2 lecturers were selected from universities located outside Java.
This was intended to capture potential regional variations in perspective and access to technological infrastructure.
Participant selection was done purposively.
The inclusion criteria were: .
active teaching experience in English writing or related courses such as academic writing or English for Academic Purposes, .
at least three years of teaching experience, and .
willingness to participate voluntarily.
The goal was to include at least one representative from each selected region in Java and from institutions outside Java, while ensuring that the selected individuals had sufficient experience to provide informed responses.
The participantsAo teaching experience ranged from 3 to 17 years, with an average of approximately 8.
6 years.
This range reflects a good balance between early-career and more experienced lecturers.
Most participants were estimated to be in their early to mid-30s, based on their reported years of teaching, assuming a typical university teaching career begins in the mid-20s.
Regarding gender distribution, 24 of the participants were female and 6 were male, reflecting the gender composition commonly found in language education programs in Indonesia.
The demographic information and institutional affiliations of all participants are summarized in Table 1.
This table includes each participantAos ID, gender, estimated age, years of experience, and university affiliation.
The diversity in institutions and geographic locations provided a robust foundation for exploring how lecturers across Indonesia perceive and experience the integration of AI in writing instruction.
Data Collection Data in this study were collected using an open-ended Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
electronic questionnaire designed with Google Forms.
This method was selected for its practicality and wide reach, enabling efficient data collection from participants located across different provinces in Indonesia.
The questionnaire link was distributed through personalized WhatsApp messages, allowing for direct communication and improving the likelihood of response.
The use of WhatsApp was deemed appropriate due to its ubiquity and frequent use among university lecturers in Indonesia, particularly for academic coordination and professional communication.
The contact information of participants was obtained through the researcher's existing professional network.
Specifically, the researcher utilized established academic and collegial connections with English lecturers from various institutions.
These connections had been developed through prior collaborations, academic forums, teacher associations, and inter-university communication.
While the sampling was not random, it followed purposive sampling principles, targeting individuals who met specific inclusion criteria relevant to the study's focus.
These criteria included current engagement in teaching English writing or related subjects and a minimum of three years of teaching experience.
All participants were informed about the voluntary nature of their involvement and were assured of the confidentiality and anonymity of their responses.
An introductory section at the beginning of the questionnaire provided clear information about the studyAos objectives, ethical considerations, and the right to withdraw participation at any stage.
The instrument consisted of 25 open-ended items grouped into four thematic categories: .
familiarity and experience with AI tools in teaching.
perceived benefits and challenges of AI-assisted writing.
ethical concerns, including plagiarism, authorship, and academic integrity.
institutional readiness and policies regarding AI integration in the curriculum.
This structure facilitated an in-depth exploration of participantsAo views while allowing them to elaborate based on their unique institutional and pedagogical contexts.
Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
Data Analysis The data gathered from the open-ended questionnaire were analyzed using NVivo, a qualitative data analysis software, following established procedures for thematic analysis.
The process was conducted in several systematic steps.
First, the responses were read repeatedly to achieve familiarity with the data.
Then, initial codes were generated inductively, focusing on recurring words, phrases, and concepts that reflected participants' perspectives.
This stage was guided by Braun and ClarkeAos .
six-phase framework for thematic analysis, which includes: familiarization with data, generating initial codes, searching for themes, reviewing themes, defining and naming themes, and producing the report.
The codes were then organized into broader categories to identify primary themes and sub-themes, allowing for a detailed interpretation of lecturersAo views on the ethical, pedagogical, and institutional dimensions of AI in writing instruction.
In addition, descriptive quantification was used to report the frequency and percentage of recurring themes, providing a clearer sense of the prevalence of certain viewpoints across the sample.
However, this numerical summary remained supportive in nature and did not alter the qualitative orientation of the study.
This approach also aligns with CreswellAos .
emphasis on exploring participants' meanings through emergent, inductive data analysis while ensuring thematic saturation and coherence across the dataset.
FINDINGS
This section presents the findings of the study, structured thematically in alignment with the research objectives and the categories derived from the open-ended questionnaire.
Drawing from the responses of 30 English lecturers from various Indonesian universities, the findings highlight a diverse range of perspectives on the integration of AI-assisted writing tools in English language Thematic analysis yielded several recurring themes that reflect both pedagogical opportunities and ethical concerns associated with AI use in academic writing contexts.
Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
The presentation of findings is organized into two main areas:
lecturersAo perceptions of AI-assisted writingAiencompassing both perceived benefits and challengesAiand .
the ethical considerations that influence those perceptions, such as academic integrity, bias, and Within each area, quantitative indicators .
percentages of respondent.
are included to support the qualitative insights and to illustrate the prevalence of key themes.
Each theme is accompanied by representative participant quotes to capture the depth and nuance of their responses.
Gauging Lecturers' Perceptions of AI-Assisted Writing in English Courses This section addresses the first objective of the study by exploring lecturersAo perceptions of AI-assisted writing in English The analysis draws from open-ended questionnaire responses provided by lecturers across multiple institutions .
dentified as University A to University V).
The data reveal a range of perspectivesAifrom enthusiastic acceptance to cautious scepticismAi highlighting both the benefits and the challenges of integrating AI tools into English writing instruction.
Overall Perspectives Towards AI-Assisted Writing The findings disclose that 40% of lecturers .
out of .
viewed AI positively, citing its potential to enhance the learning process, especially through real-time feedback and assistance with grammar and structure.
Lecturers like Participant 6 emphasized that "AI can be an excellent supplementary tool for students, helping them with areas like grammar and sentence structure while allowing them to focus on content development.
On the other hand, fifty percent of the participants, or fifteen out of thirty, voiced their reservations about the excessive reliance on AI, expressing their fear that it could hinder the development of independent writing skills.
Participant 4 pointed out, "While AI tools are useful, there is a risk that students may rely too much on them and not develop their critical thinking or creativity.
" This sentiment underscores the scepticism many lecturers feel about how AI might affect students' Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
writing autonomy.
Perceived Benefits of AI Tools In their writing assignments, participants highlighted several advantages gained from utilizing AI tools.
A significant benefit recognized by 75% of lecturers was real-time feedback, which allows students to immediately correct their errors.
One lecturer (P.
AuThe immediate feedback from AI tools is extremely valuable.
enables students to recognize their errors and make adjustments while still engaged in the learning processAy.
Additionally, 67% of the lecturers pointed to the development of writing skills as another major benefit.
AI tools help students improve their grammar, style, and sentence structure, providing the scaffolding necessary for mastering complex language rules.
As P7 emphasized.
AuAI tools are particularly beneficial for non-native English They help students grasp difficult grammar rules and enhance their writing skillsAy.
Another 55% of the participants emphasized time efficiency as a key advantage of AI.
Tools that automate processes such as grammar checks and plagiarism detection significantly reduce the workload for both students and lecturers, allowing more time to focus on content Lecturer P5 noted.
AuAI tools save a lot of time by automatically handling error detection, allowing both students and teachers to focus on developing ideas and contentAy.
Figure 1.
Perceived benefits of AI tools Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
Figure 1 displays the AI toolsAo perceived benefits in English writing assignments as reported by lecturers.
The standout benefit, highlighted by 75% of participants, is real-time feedback, which enables students to make immediate corrections during their Additionally, 67% noted that these tools help develop writing skills, particularly in grammar, style, and structure, which is especially valuable for non-native English speakers.
Finally, 55% of respondents appreciated the time efficiency that AI tools provide by automating error detection, allowing both students and lecturers to concentrate more on creating content.
Challenges of AI Tools While the AI toolsAo benefits were acknowledged, several challenges were also identified.
The most prominent concern, reported by 70% of the lecturers, was the excessive dependence on AI tools, which may hinder students' independent writing development.
Lecturer P1 expressed caution, stating.
AuWe need to guarantee that students do not become very dependent on AI tools, as it could hamper the advancement of their writing abilitiesAy.
Another significant concern was algorithmic bias.
Half of the participants raised concerns about potential biases embedded in AI systems, especially when the tools are trained on datasets that do not reflect diverse linguistic or cultural backgrounds.
Lecturer P4 explained.
AuAI systems may carry the potential to reinforce biases present in their training data, resulting in evaluations that lack fairnessAy.
Lastly, insufficient clarity in the processes behind AI decisionmaking was cited by 60% of participants as a critical issue.
This opacity makes it difficult for lecturers and students to fully trust AI-generated P7 commented.
AuItAos hard to trust AI when its decisionmaking process is not transparentAy.
In light of the challenges, lecturers maintain a positive outlook concerning the possible influence of AI on education.
Lecturer P1 predicted that AI will continue to be a significant tool in improving both teaching and learning experiences in writing courses.
However, continuous improvements and careful monitoring are necessary to ensure AIAos effectiveness and fairness.
Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
Figure 2 highlights the challenges lecturers face when utilizing AI tools for English writing tasks.
A significant concern, highlighted by 70% of participants, is the potential for students to develop an extreme reliance on AI, which may impede their ability to write A further big problem that was mentioned by fifty percent of the respondents is algorithmic bias, which is when artificial intelligence systems may reinforce prejudices that are already present in their training data, which can lead to unjust evaluations.
addition, sixty percent of the participants expressed worries over the transparency of the decision-making processes of artificial intelligence, which makes it more difficult for both teachers and students to have complete faith in the feedback and evaluations that are provided by AI.
Another significant concern was algorithmic bias.
Half of the participants raised concerns about potential biases embedded in AI systems, especially when the tools are trained on datasets that do not reflect diverse linguistic or cultural backgrounds.
Lecturer P4 explained.
AuAI systems may carry the potential to reinforce biases present in their training data, resulting in evaluations that lack fairnessAy.
Lastly, insufficient clarity in the processes behind AI decisionmaking was cited by 60% of participants as a critical issue.
This opacity makes it difficult for educators and students to fully trust AI-generated P7 commented.
AuItAos hard to trust AI when its decisionmaking process is not transparentAy.
In light of the challenges, educators maintain a positive outlook concerning the possible influence of AI on the education.
Lecturer P1 predicted that AI will continue to be a significant tool in improving both teaching and learning experiences in writing courses.
However, continuous improvements and careful monitoring are necessary to ensure AIAos effectiveness and fairness.
Figure 2 highlights the challenges educators face when utilizing AI tools for English writing tasks.
A significant concern, highlighted by 70% of participants, is the potential for students to develop an extreme reliance on AI, which may impede their skills to write autonomously.
A further big problem that was mentioned by fifty percent of the respondents is algorithmic bias, which is when artificial intelligence Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
systems may reinforce prejudices that are already present in their training data, which can lead to unjust evaluations.
In addition, sixty percent of the participants expressed worries over the transparency of the decision-making processes of artificial intelligence, which makes it more difficult for both teachers and students to have complete faith in the feedback and evaluations that are provided by AI.
Figure 2.
Challenges faced using AI tools Proposed Solutions and Institutional Recommendations Figure 3 illustrates the proposed solutions for integrating AI tools in English writing assignments.
The most recommended solution, highlighted by 85% of participants, is the need to equalize AI tools with traditional teaching methods, ensuring that AI supports but does not replace human instruction.
78% of participants emphasized the necessity of providing ethical AI use training for both students and lecturers, while 70% called for clear institutional guidelines to regulate AI use in writing assignments.
When it comes to incorporating AI tools into English writing assignments, a methodical and organized strategy is required in order to make the most of their potential while simultaneously addressing the obstacles that have been recognized by lecturers.
Following the feedback of thirty lecturers, a number of proposals that are both practical and ethical have been developed for the purpose of ensuring that artificial intelligence is employed in the classroom in an effective Many participants emphasized the importance of setting clear Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
limitations on AI useAifor instance, limiting AI-generated content to 30% or less of an assignment .
P8.
P17.
Others suggested using AI only as a supportive tool for idea generation or grammar checking, not as the main source of writing .
P6.
P12.
Figure 3.
Proposed solutions for integrating AI tools in English writing assignment Ethical concerns such as plagiarism and intellectual ownership were also widely discussed.
One lecturer noted: AuI usually use an AI detector or plagiarism checker and ask students to rework if they exceed the limitAy (P.
Several participants .
P9.
P14.
agreed that institutions must provide clear guidelines and training to both lecturers and students on how to use AI tools responsibly.
These insights highlight that while AI has a growing role in writing instruction, its implementation must be carefully framed by ethical boundaries, institutional policy, and instructional clarity to promote responsible use.
The recommendation to balance AI with traditional teaching methods is supported by 85% of the participants who emphasized that while AI can assist students in writing tasks, it should not replace the critical role of lecturers.
Lecturer P2 highlighted this concern:
"Artificial intelligence tools should complement, not replace, human Students still need to develop their analytical thinking and " This sentiment reflects a widespread belief among lecturers that AI's role should be supplemental.
AI can provide Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
immediate feedback on technical aspects, but deeper learning processes like critical thinking and creative problem-solving still require human interaction.
The findings also reveal that 50% of participants expressed concerns about students potentially becoming over-reliant on AI tools.
AI should be integrated thoughtfully to encourage the development of essential academic skills like independent writing.
For instance.
Participant 42 stated, "AI can assist with the mechanics, but understanding the subtleties of a student's argument or the creativity in their expression is something only a human can do.
" This highlights the importance of fostering student autonomy while providing appropriate technological To ensure effective AI integration, lecturers advocate for a "blended learning" approach, in which AI tools are implemented to augment traditional teaching methods while ensuring that the educator's role remains prominent.
The aim is to encourage students to engage critically with AI feedback, using it as a tool to improve technical elements of writing while reserving complex, higher-order thinking for human guidance.
Seventy-eight percent of those who took part in the study stressed how important it is to provide comprehensive training to both teachers and pupils on all aspects of the ethical application of artificial intelligence tools.
Lecturer P3 made the following observation:
"Training is crucial for ensuring that lecturers and students comprehend the effective and ethical use of AI tools.
" This training would concentrate on avoiding academic dishonesty, eliminating an excessive dependence on artificial intelligence, and gaining a knowledge of the biases that are inherent in AI technologies.
Several lecturers .
P7.
P10.
recommended the inclusion of workshops or formal modules that teach students when and how to appropriately use AI tools, including how to acknowledge AI assistance in their assignments.
A consistent theme among 70% of lecturers was the need for institutional guidelines to standardize AI use across educational These guidelines would provide clear rules on when and how AI tools can be used, ensuring that students are held to uniform standards of academic integrity.
Lecturer P5 remarked, "We need clear institutional policies to define acceptable uses of AI, especially in areas such as Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
academic integrity and the appropriate balance between AI and student " The absence of such guidelines leads to inconsistent practices and places undue responsibility on individual lecturers to make ethical judgments.
Moreover, lecturers expressed that guidelines should address specific concerns like AI bias, over-reliance, and transparency in AIgenerated feedback.
To support academic integrity, institutions would need to incorporate explicit rules on AI usage in assessment criteria.
This necessitates that students disclose their utilization of AI tools and obtain training on how to successfully incorporate AI assistance with their personal contributions.
Institutional support plays a crucial role in effectively integrating AI within the classroom environment.
Lecturers who operate in environments where there is clear institutional backing feel more confident in using AI tools.
In contrast, those without sufficient support or guidelines tend to avoid using AI, as they are unsure of how to ethically manage its integration.
Exploring the Ethical Considerations Influencing Educators' Perceptions Among the participants, numerous important ethical issues arose concerning the incorporation of AI tools into English writing The primary ethical considerations include academic integrity, bias and fairness, as well as transparency and accountability.
These factors heavily influence lecturers' perceptions of AI, shaping their views on its acceptability and limitations in academic settings.
Academic Integrity Maintaining academic integrity is a major ethical concern shared by 70% of participants.
Lecturers expressed worry that while AI tools can help detect plagiarism, they may also enable students to produce work that lacks originality.
For instance.
Lecturer P8 stated, "AI tools are useful for detecting plagiarism, but students often rely on them to generate content, raising questions about the authenticity of their work.
Moreover, many lecturers emphasized the need for institutional guidelines to clearly delineate when and how AI tools can be ethically Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
used in writing assignments.
Lecturer P5 remarked, "We need stringent rules regarding AI usage to ensure students' work remains authentic.
" This suggests a broader institutional responsibility to set boundaries, ensuring that AI functions as an aid to enhance analytical processes instead of a means to bypass essential reasoning.
Bias and Fairness Bias in AI systems is another significant concern, cited by 50% of participants.
Lecturers worried that AI tools, which are often trained on large datasets, could unintentionally reflect or reinforce existing biases.
Lecturer P3 explained, "It is necessary to verify that AI tools do not reinforce existing biases.
, especially when evaluating student work from diverse backgrounds.
" This highlights the importance of assessing the fairness of AI algorithms, particularly in multicultural settings like Indonesia, where linguistic and cultural diversity must be considered.
Additionally, lecturers called for greater transparency in how AI systems are trained and how they make decisions.
A participant from University D pointed out, "If we don't know the datasets used to train these AI tools, we can't ensure they are fair.
" This observation highlights the necessity for educational institutions to collaborate with AI creators to guarantee that these tools maintain neutrality and do not perpetuate biased viewpoints.
Transparency and Accountability Sixty percent of participants expressed concern regarding the absence of transparency in AI decision-making developments, highlighting the necessity for well-defined guidelines and accountability in the application of AI systems within educational Lecturer P7 noted, "The proprietary nature of AI tools makes it difficult to understand their decision-making process, and this opacity creates a trust issue.
" Without insight into how AI-generated assessments are made, lecturers find it challenging to rely on these tools for objective evaluations.
To address this, lecturers recommended that AI developers provide detailed information about how their algorithms operate and the data used to train them.
Lecturer P3 remarked, "We need clear Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
guidelines for AI use in writing assignments, particularly when it comes to explaining how AI tools work and ensuring their accountability.
" The call for increased transparency reflects a broader push to guarantee that AI systems function effectively while maintaining fairness and trustworthiness, giving both lecturers and students confidence in their application.
DISCUSSION
This study provides timely insight into Indonesian university lecturersAo perceptions of AI-assisted writing tools in English language The results reveal a nuanced balance between enthusiasm for AIAos pedagogical affordances and concern for its ethical, institutional, and cognitive implications.
These findings not only address practical classroom realities but also contribute to theoretical conversations surrounding digital literacy, constructivist learning, and teacher readiness in AI-mediated education.
The advantages associated with AI tools, particularly in improving grammar, structure, and coherence through real-time feedback, were widely acknowledged by the participants.
Such feedback was seen as especially advantageous for non-native English speakers who rely on immediate corrective input to master complex language patterns.
These findings align with earlier studies that highlight AIAos potential in scaffolding academic writing for EFL learners (Han & Li, 2024.
Hooda et al.
, 2022.
Fathi & Rahimi, 2.
Participants emphasized that tools like Grammarly and ChatGPT support technical fluency while allowing students to concentrate on content development, corroborating constructivist views that position learners as active agents in the knowledge-building process (ZawackiRichter et al.
, 2019.
Selwyn, 2.
However, lecturers cautioned against over-reliance on these technologies, expressing concern that students may bypass critical thinking and creativity in favor of surface-level corrections.
This echoes warnings in the literature that excessive reliance on AI tools can undermine metacognitive engagement and reduce learnersAo capacity for deep reflection (Ersyz & Engin, 2024.
Holmes et al.
, 2.
These concerns reinforce the principles of constructivist pedagogy, which Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
highlight the need for students to actively process and reconstruct meaning through self-regulated learning rather than passively accept machine-generated suggestions.
One of the most significant concerns reported by participants was the threat to academic integrity.
Although AI tools can detect plagiarism, they can also be misused to generate unoriginal content, making it difficult to assess studentsAo authentic voices and learning processes (Uzun, 2024.
Gustilo et al.
, 2.
As AI-generated writing becomes more fluent and human-like, the boundaries of authorship and originality become increasingly ambiguous (Hutson, 2.
Nguyen .
notes similar findings in a student-focused study, where learners expressed uncertainty about what constitutes acceptable AI use in academic writing and demanded more explicit institutional guidance.
Lecturers also expressed concern about algorithmic bias, particularly how AI tools might reinforce linguistic and cultural norms that disadvantage students from diverse backgrounds.
This concern is particularly pressing in multilingual contexts like Indonesia, where inclusivity in assessment is essential.
AI writing assistants, often trained on monolingual English corpora, may evaluate EFL writing ineffectively or unfairly (Toscu, 2.
, posing a risk of marginalizing students from non-dominant linguistic groups.
Khan et al.
emphasize that equitable AI integration requires institutional frameworks that address fairness, bias, and data Their work stresses the importance of deploying AI in ways that bridge, rather than widen, the digital divide, especially in resource-limited educational environments.
Building on these concerns.
McDonald et al.
highlight that even institutions embracing Generative AI (GenAI) often fail to address the needs of diverse educational contexts, such as STEM tasks or culturally sensitive writing.
This oversight results in a limited pedagogical approach that replicates biases rather than mitigating Similarly.
Shahzad.
Xu, and Zahid .
explore the influence of generative AI technologies on learning outcomes in higher education institutions, noting that AI tools can significantly impact fairness, self-efficacy, and creativity.
These factors are particularly Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
crucial in ensuring equitable outcomes for diverse learners.
Their findings align with the present study, in which lecturers voiced the need for not only access to AI tools but also clear training and policy frameworks that support equitable, pedagogically sound use.
Han et al.
stress that such guidelines should also consider the relational and pedagogical impacts of AI on students, ensuring its use fosters learner autonomy and transparency rather than diminishing them.
This concern reflects lecturersAo consistent calls for explicit standards that define acceptable AI use in academic contextsAiespecially to avoid academic dishonesty, misuse, and learner dependency (Nguyen, 2025.
Han et al.
, 2.
Incorporating ethical training into the curriculum also aligns with constructivist learning theories, which emphasize that students construct knowledge through active engagement and reflection, rather than passive dependence on machine-generated output.
In support of this, lecturers in the study advocated for ethicsfocused workshops and formal instruction that teach students how to balance AI-generated content with their own contributions.
Suggested topics for training modules included when and how to use AI tools responsibly and how to declare AI assistance appropriately in Additionally, providing training resources for lecturersAisuch as workshops on using AI-assisted writing rubrics and identifying over-reliance on AI toolsAiwas considered essential for ensuring consistent and responsible classroom implementation.
McIntyre .
also emphasizes the importance of equitable classrooms and programs in the age of AI, stressing the need for writing lecturers to maintain a focus on individualized feedback and critical engagement, particularly in first-year writing courses.
Her argument for "informed refusal" allows lecturers to acknowledge AI's presence without relying on it excessively, thereby preserving the authenticity of student learning and engagement.
Therefore, to ensure equitable outcomes.
AI tools must be designed with attention to linguistic diversity and cultural sensitivity, especially in contexts like Indonesia, where inclusivity in educational assessment is crucial.
must be implemented with care to ensure it accommodates all students, irrespective of their backgrounds, promoting a genuinely inclusive and Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
efficient learning environment.
Moreover, lecturers expressed concerns regarding the absence of transparency in AI decision-making, which complicates its application in the classroom.
Many lecturers found it difficult to interpret or trust AI-generated feedback, given the lack of transparency in how these systems are trained or how they function (Saeidnia et al.
, 2024.
Holmes et al.
, 2.
This "black-box" nature undermines both educator confidence and student understanding.
McDonald et al.
note, although over half of U.
analyzed offered syllabi and classroom activities involving GenAI, there was little discussion of how to support lecturers in navigating ethical grey zones or in managing the cognitive burden that may accompany GenAI integration.
This study also revealed a gap between the pace of AI development and institutional preparedness.
Many lecturers noted the absence of clear policies or training on ethical AI use, placing the burden of decision-making on individual lecturers.
This finding aligns with previous research that underscores the need for robust institutional frameworks to ensure consistent and equitable AI integration (Williamson et al.
, 2020.
McDonald et al.
, 2025.
Singun.
The demand for guidance is echoed on the student side as well.
Nguyen .
, in a survey of students in a U.
university, found that although many students use generative AI to brainstorm and support their writing, they are critically engaged with AI output and emphasize the importance of explicit ethical guidance from both syllabi and institutional policies.
This reflects a shared concern among both lecturers and students regarding the lack of clarity in appropriate AI CelikAos .
Intelligent-TPACK model is particularly useful for interpreting these findings.
His framework expands the traditional TPACK (Technological.
Pedagogical, and Content Knowledg.
model by incorporating ethical dimensions, asserting that effective utilization of AI necessitates a combination of technical expertise, pedagogical understanding, and ethical considerations.
This model supports the present studyAos conclusion that teacher development should include AI-specific training that balances technological familiarity with Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
pedagogical purpose and ethical awareness.
Lecturers in this study advocated for a blended learning approach, incorporating AI tools to enhance and support human instruction instead of substituting it.
AI facilitates the automation of routine feedback, allowing lecturers to concentrate on advanced cognitive skills, including argumentation, originality, and the organization of discourse (Kim et al.
, 2022.
Fathi & Rahimi, 2.
However, to achieve this balance, students and teachers alike require training in AI literacyAihow to use tools appropriately, interpret results, and reflect on ethical implications (Celik, 2.
In practical terms, institutions should implement standardized guidelines, provide hands-on AI integration workshops, and offer access to ethically vetted technologies.
Furthermore, equitable access to AI tools is essential to avoid reinforcing digital inequality, especially in resource-constrained environments or among students with limited technological proficiency (Lee, 2024.
Hossain, 2.
These actions would reduce discrepancies between individual lecturers and foster a more equitable educational environment.
According to the findings of this study, the deployment of AIassisted writing tools must be driven by ethical reflection, institutional support, and pedagogical intentionality.
This is despite the fact that these tools offer major educational benefits.
This is only possible if the use of artificial intelligence is clear, inclusive, and founded on strong educational principles.
AI has the potential to empower learners and enrich instruction.
When it comes to deciding whether or not the incorporation of artificial intelligence into English writing instruction will be successful, one of the most important factors to consider is whether or not lecturers, institutions, and policies are willing to humanize and contextualize the application of AI in English writing CONCLUSION This study has provided insights into Indonesian university lecturersAo perceptions of AI-assisted writing tools, focusing on ethical considerations surrounding their use in English language instruction.
The findings suggest a nuanced perspective: while lecturers Yunita.
Fitrianna.
, & Maulana.
Artificial Intelligence in English Writing: Ethical Integration and LecturersAo Perspectives.
JEELS, 12.
, 641-666.
acknowledge the benefits of AIAisuch as real-time feedback and improved writing qualityAithey also express apprehensions about academic integrity, student dependency, algorithmic bias, and the lack of institutional guidelines.
A key implication of this research is the urgent need for clear, standardized policies that support ethical and effective AI integration in higher education.
Without such frameworks, the use of AI may exacerbate inequality and hinder the development of independent writing skills.
Institutions should take the initiative to formulate guidelines that tackle concerns related to academic integrity, bias, and This approach will help ensure that AI tools serve as supportive instruments instead of substitutes for critical thinking and original authorship.
This study, while contributing valuable insights, is constrained by its sample size and dependence on self-reported data from a singular national context, potentially limiting its applicability to wider Future research ought to investigate student perspectives, assess the long-term effects of AI-assisted learning on academic performance, and analyze institutional policies across various educational and cultural contexts.
These studies would offer a deeper insight into the ethical and efficient incorporation of AI in writing instruction on a global scale.
ACKNOWLEDGMENTS
This research was funded by research grants of Litapdimas 2024 under the Ministry of Religious Affairs of Indonesia.
DECLARATION OF AI AND AI-ASSISTED TECHNOLOGIES
During the preparation of this work the authors used ChatGPT for optimizing language and readability.
The authors have reviewed and edited the content as needed and take full responsibility for the content of the publication after the use of this tool/service.
REFERENCES