1

AN INVESTIGATION OF STUDENTS’ RELIANCE ON

AI TOOL IN ACADEMIC WRITING AT UIN SJECH
M. DJAMIL DJAMBEK 6TH SEMESTER ENGLISH
DEPARTMENT STUDENTS: CASE STUDY
RESEARCH

CHAPTER I

BACKGROUND

A. Background of Problem

The goal of AI to develop machines that can behave in circumstances

surprisingly intelligently. For the present purpose we will define intelligence as

(John McCarthy, 1956)
the ability to achieve goals in the world . “A neural

network may be defined as a collection of nodes connected together in some

pattern, each node representing a neuron or some other simple processing unit.

We attempt to find ways of connecting these units so that the entire system can
(Marvin Lee Minsky, 1954)
perform intelligent behavior” . It means, AI is

artificial intellegence branch of computer sciences that focuses on the Machine

and system development to do any commands that give by Users or

Administrator. AI or artifical Intellegence are able to presenting a data or fact

base on the digital informations and could arrange also unify it obviously to

help human work to make the job more easy and effective.
 2

On the other hand, Overreliance occurs when people take AI output as

correct, in places where AI might not do well We find three main reasons: how

good people think the AI is, if they treat the AI as if it is not to blame, and

interface factors that make people less involved. Overreliance makes worse
(Glikman, 2023)
mistakes, and reduce awareness of next events . Effective

human-AI teaming needs clear and honest trust, which means the trust you give

matches how much trust the AI should really get for the task at hand and in the

time given. Honest trust is shaped by how clear the AI is on what it can do,

how well you know how to use the tool, and how well you know how the tool

can help you. If trust is not right then a group will not do their action at their
(Lee, 2022)
best

A preliminary research, Relience are often occurred by the students

based on preliminary research study through document analysis framework

MT, AM, IS, and , DA informants, they are proven who relience towards AI.

They have seventy up to hundred percentage, categorizes as High similarity

generated by AI, Through detecting using Quilbot and ZeroGpt. Students often

do copy paste after using Generative AI chatbots for finishing their works

without paraphrasing it using their own words and lack of interpretation. The

findings uncovered several significant problems regarding students' use of AI

tools in academic writing, which can be grouped into three main categories.

First, a lack of deep engagement with the writing process , as students tend to

depend heavily on AI-generated content without critically analyzing or

 3

internalizing the material. This leads to low knowledge of the subjects and low

personal work for their tasks. Second, there is poor growth in skills to write,

like rewording, making a point, and putting their thoughts in order. As so many

students copy or just change a little bit the work by AI, they do not get the right

amount of work needed to make their writing better. Third, misuse of AI

leading to integrity violations when students send in work that is mainly or

totally made by computers with no proper note on it. This not only fools about

how much they know and can do but it also makes the schools have to change

rules about how they use AI on tests that students take. Then, if they proven

overrelience through detecting such as Zero GPT, Quillbot, Grammarly,

Originallity AI, and Smodin, they could not allowed for submit the Assignment

look like that.

According to (Fieseler, 2023) argue that AI literacy must include not

only understanding how AI works but also recognizing its limitations through

detection tools. Training users to interpret AI detector signals improves their

ability to critically assess AI outputs, thereby reducing blind reliance and

enhancing informed usage. research show a way where AI sets can have ways

to tell users when AI says something that might be wrong, fake, or just not

good enough. They tell users to think twice about what they just saw. But we

will have to tell how we can make these work well in real-life.

On the other hand, High AI similarity generated by AI percentage at

the students are concerning, especially for university stage for academic
(Wiese et al., 2025)
integrity. find that efforts to teach AI ethics do helpfully
 4

draw on a holistic view (as opposed to a narrow view), and utilize progressive

pedagogies like case studies and group projects that aim to meaningfully

challenge students’ ethical reasoning skills in applied practices. Either, it would

be bad influnce on students academic integrity, Honest, Ethic, Originality, and

Responsibility. The consequences make students to be worse comprehension in

reading, writing, and critical thinking. This icasion could be impacted to

students, institution, and degrees. This research aim to investigate the student

movement towards using AI on their academic Writting.

This serves as a foundational basis for further exploration of students'

conditions in using AI tools within academic environments, particularly in

completing academic writing tasks at the Faculty of Teacher Training and

Education, Department of English Language and Literature, UIN Sjech M.

Djamil Djambek , Bukittinggi. The use of AI tools such as Grammarly,

QuillBot, and ChatGPT has increasingly spread among students, especially

those in their sixth semester who are preparing for final assignments and field

research.

This study aims to explore in depth the process of AI tool usage by

sixth semester students in completing academic writing tasks, as well as to

analyze the level of their dependence on these technologies. In addition, this

research will identify the factors influencing students’ reliance on AI tools,

viewed from their linguistic competence, academic understanding, and learning

motivation in completing writing tasks.

Ultimately, this research will have some ways to help plan rule about
 5

how to use AI in advance learning, especially when it comes to how people

learn to write in a second language. The results of the research will also help

teachers and schools who aim to write better essays while trying new types of

teaching methods that use technology. In addition, it can help teachers and

those who run schools plan clear rules about how to follow the rules for using

AI tools that will fit with what it means to digital and learning skills.
 6

B. Identification of problem

Based on the background of the study, several problems can be identified

about the use of AI Generated tools at 6th-semester students in English Education

Department at the Islamic University of Bukittinggi while doing academic writing;

C. Formulation of problem

Based on the identification of problem using AI on Academia through

seeing students effort this study have some research Question:

1. How prevalent is the reliance on AI tools among sixth-semester

English Education Department students at UIN Sjech M. Djamil

Djambek in completing academic writing tasks, based on preliminary

document analysis framework of selected informants?

2. What factors contribute to students' reliance on AI tools in academic

writing, particularly among those who are identified as highly

dependent based on AI detection results?

 7

D. Purpose of research

This Research aims to investigate why students avoiding their Relience

through using AI Digital Software. The main purpose of this study is to

investigate the students effort in globalization era of tremandously AI

influenced on the society, especially for academic context purposely create

the original, authentic, also can be acceptable manifestation of work ;

1. To examine the extent of students’ reliance on AI tools in academic

writing at the English Department of UIN Sjech M. Djamil Djambek,

particularly among sixth-semester students, based on preliminary

observations of selected informants.

2. To analyze the level of AI-generated content in students' academic

writing using detection tools such as QuillBot and ZeroGPT, and to

investigate the contributing factors behind high reliance on AI tools,

especially among those categorized as having seventy up to hundred

percentage AI similarity.

E. Significance of Problem

1. For Students : This study will help student to create some scientific

literatures maybe paper, scientific journal, or article to be original

as they were created by themselves through using AI generated

chatbot purposely avoiding to exposing plagiarism. AI could also

brainstroming several Ideas help student to find a legit information

and then they arrange it with their own way by using AI software.
 8

2. For Lecturers : This study also tremendously give lecturer chance

to proofing the student’s scientific literatures through to investigate

and analyze it towards AI Software. This study also makes

lecturers work easier because of ai detector

3. For Public Society : This study will minimalize unqualified people

rised by the eduactional institution who claim that they were an

educated person. but in fact they just did so many plagiarism which

not come from their own proficiency and ability also acrossing the

Academic Integrity.

F. Definition of key terms

AI generated chatbot - AI-generated chatbots are computer programs

powered by artificial intelligence (AI) that simulate human-like

conversations through text or voice interactions. These chatbots utilize

natural language processing (NLP), machine learning, and deep learning

algorithms to understand user input, generate contextually relevant

responses, and improve over time through experience. Common examples

include ChatGPT, Google Bard, and Bing AI. They are widely used in

education, customer service, content creation, and personal assistance. In

academic contexts, AI-generated chatbots are increasingly used by

students for writing support, research, and problem-solving, raising

important questions about authorship, originality, and learning

 9

authenticity.

Reliance - in its general and literal meaning, refers to the act of

depending on something or someone for support, guidance, or fulfillment

of a need. It implies a sense of trust or confidence in the reliability,

accuracy, or effectiveness of the entity being depended upon. In a broad

sense, reliance can apply to various contexts such as emotional, physical,

or technological dependence and reflects the extent to which an individual

or system leans on another factor to function or make decisions. The term

originates from the Latin religare , meaning "to bind again," suggesting a

binding connection or dependency between two entities.

 10

CHAPTER II

REVIEW OF RELATED LITERATURES

A. Review of Related Theories

1. Basic Concepts of Academic Writing

a. Definition and Characteristics of Academic Writing

Academic writing refers to a formal style of communication

used primarily in scholarly contexts, such as universities, research

institutions, and professional publications (Hyland, 2009). It is

characterized by clarity, precision, and logical organization, aiming to

convey complex ideas and arguments effectively. Unlike casual

writing, academic writing adheres to specific conventions, including

proper citation styles, objective language, and evidence-based

reasoning. These features help maintain credibility and ensure that

information is presented in a manner suitable for critical evaluation by

peers and scholars.

One key characteristic of academic writing is its reliance on

evidence and data to support claims (Swales & Feak, 2012). Writers

are expected to back up their arguments with credible sources, such as

peer-reviewed journals, books, and empirical studies. This emphasis

on evidence ensures that academic discourse remains grounded in

verifiable facts rather than personal opinions. Furthermore, academic

 11

writing often employs third-person perspective to maintain objectivity

and avoid bias, although somes disciplines may allow first-person

narration depending on the context and purpose of the writing.

Another defining feature of academic writing is its structured

format, which typically includes an introduction, body paragraphs,

and a conclusion (Northcut, 2017). Each section plays a specific role:

the introduction sets the stage by outlining the thesis or purpose, the

body provides supporting evidence and analysis, and the conclusion

summarizes the findings and suggests implications. This structure

helps readers follow the writer's logic and facilitates comprehension.

In essence, academic writing is not only about what is said but also

how it is communicated, making it a vital skill across all fields of

study.

b. Purpose and Functions of Academic Writing

The primary purpose of academic writing is to communicate

knowledge, engage in intellectual discourse, and contribute to the

development of new ideas within a field (Lillis, 2001). Scholars use

writing to present their research findings, critique existing theories,

and propose innovative solutions to real-world problems. Through this

process, Academic writing serves as a vehicle for intellectual

exchange and progress. It enables researchers to build upon each

other’s work, ensuring that knowledge evolves over time through

 12

rigorous examination and debate.

Beyond communication, academic writing functions as a tool

for learning and critical thinking (Bean, 2011). As students engage in

writing tasks, they are required to analyze information, synthesize

diverse perspectives, and articulate coherent arguments. This

cognitive engagement deepens understanding and enhances analytical

skills, which are essential for academic and professional success.

Moreover, academic writing encourages reflection and self-

assessment, prompting writers to evaluate the strength of their

arguments and consider alternative viewpoints.

Additionally, academic writing plays a crucial role in

establishing authority and credibility within academic communities

(Paré, 2018). Properly cited and well-structured writing demonstrates

a writer's familiarity with existing literature and adherence to ethical

standards. It allows scholars to situate their work within broader

academic conversations and gain recognition for their contributions.

Whether in the form of essays, research papers, or journal articles,

academic writing serves multiple purposes that extend beyond mere

information sharing—it shapes the way knowledge is created,

validated, and disseminated.

c. Common Challenges in Academic Writing

One of the most common challenges faced by students and

 13

emerging scholars is mastering the formal tone and structure required

in academic writing (Belcher, 2009). Many learners struggle with

transitioning from informal or personal writing styles to the more

objective and precise language expected in academic contexts. This

challenge is particularly pronounced among non- native English

speakers, who must simultaneously navigate linguistic complexities

and cultural expectations surrounding academic expression. Without

sufficient guidance or practice, these difficulties can hinder effective

communication and lead to frustration.

Another significant obstacle is the ability to construct and

support a coherent argument (Gosden, 2003). Academic writing

demands clear thesis statements, logical progression of ideas, and

evidence-based reasoning—skills that many novice writers find

difficult to develop. Often, students either fail to clearly articulate

their main points or rely too heavily on summarizing sources without

engaging critically with them. This results in weak arguments that

lack depth and persuasive power. Learning how to integrate external

sources while maintaining one’s own voice is a nuanced skill that

requires sustained effort and feedback.

Time management and writer’s block also pose considerable

challenges in academic writing (Rose & McClafferty, 2001). The

pressure to meet deadlines, combined with perfectionism or fear of

criticism, can paralyze the writing process. Procrastination becomes a

 14

coping mechanism, leading to rushed and substandard work.

Additionally, the revision process—often neglected—is essential for

producing high-quality academic writing. Many students

underestimate the importance of drafting, revising, and editing,

treating writing as a one-time task rather than an iterative process.

Overcoming these obstacles requires developing effective strategies,

such as setting realistic goals, seeking peer feedback, and practicing

regularly.

2. Introduction to Artificial Intelligence (AI) in Education

a. Definition and Evolution of AI Technology

Artificial Intelligence (AI) refers to the simulation of human

intelligence in machines designed to perform tasks that typically

require human cognition, such as problem-solving, decision-making,

and language understanding (Russell & Norvig, 2021). AI

encompasses various technologies, including machine learning,

natural language processing, and neural networks, which enable

computers to learn from data and improve performance over time

without explicit programming. Since its inception in the mid-20th

century, AI has evolved significantly, progressing from rule-based

systems to sophisticated algorithms capable of handling complex,

unstructured data. The evolution of AI can be traced through several

key phases. Early AI research in the 1950s focused on symbolic

 15

approaches, where systems followed predefined rules to mimic

logical reasoning (Haugeland, 1985). The late 20th century saw the

rise of machine learning techniques, allowing computers to learn

patterns from large datasets. In the 21st century, advancements in

computational power and the availability of big data have propelled

AI into mainstream applications, including image recognition,

speech synthesis, and autonomous vehicles. Today, AI is

increasingly integrated into education, healthcare, finance, and other

sectors, transforming traditional practices and enabling new forms of

interaction and automation.

Recent developments in generative AI, such as large

language models like GPT-4, have further expanded the capabilities

of AI systems in understanding and generating human-like text

(Brown et al., 2020). These models are trained on vast amounts of

text data and can generate coherent, contextually relevant responses

to user prompts. This breakthrough has opened new possibilities for

AI in education, particularly in areas like personalized learning,

automated assessment, and academic writing support. As AI

continues to evolve, its impact on educational practices and policies

will likely grow, necessitating ongoing research and ethical

considerations.

b. Role of AI in Educational Contexts

 16

AI has become a transformative force in modern education,

offering tools that enhance teaching, learning, and administrative

processes (Holmes et al., 2022). One of its primary roles is in

personalized learning, where AI-powered platforms adapt

instructional content to individual student needs based on

performance data. Systems like intelligent tutoring systems and

adaptive learning software provide customized feedback and

resources, helping students grasp concepts at their own pace. This

level of personalization was previously difficult to achieve in

traditional classroom settings, especially in large-scale educational

environments.

In addition to personalized instruction, AI supports educators

by automating routine tasks such as grading, attendance tracking,

and curriculum planning (Luckin et al., 2016). Automated grading

systems can assess multiple-choice and short-answer questions with

high accuracy, freeing teachers to focus on higher-order activities

like mentoring and facilitating discussions. Moreover, AI analytics

can identify at-risk students by analyzing behavioral and academic

data, allowing early intervention strategies to be implemented. These

Efficiencies contribute to improved educational outcomes and more

effective resource allocation in schools and universities. AI also

fosters inclusivity and accessibility in education by providing tools

for students with disabilities (Bouck et al., 2021). For example,

 17

speech-to-text and text-to-speech technologies assist visually

impaired or dyslexic learners, while AI-driven captioning improves

access for hearing-impaired individuals. Furthermore, AI chatbots

and virtual assistants offer 24/7 academic support, answering queries

and guiding students through course materials. As AI continues to

evolve, its integration into educational contexts promises to make

learning more equitable, efficient, and engaging for diverse

populations.

c. AI Tools Used in Academic Writing (e.g., Grammarly, Turnitin,

ChatGPT, etc.)

Several AI-powered tools have been developed to assist with

academic writing, improving grammar, enhancing clarity, and

detecting plagiarism (Stoian & Albulescu, 2023). Among the most

widely used is Grammarly , an AI-driven writing assistant that

checks for grammatical errors, punctuation issues, and stylistic

inconsistencies. It provides real-time suggestions to improve

sentence structure and word choice, helping users produce polished

and professional documents. While Grammarly is particularly

beneficial for non-native English speakers, it also aids native

speakers in refining their writing for academic rigor and clarity.

Another prominent tool is Turnitin , which leverages AI and

machine learning to detect instances of plagiarism by comparing

 18

submitted texts against a vast database of academic sources (Elliot,

2020). Turnitin generates similarity reports that highlight matching

content, allowing instructors to assess originality and guide students

toward proper citation practices. Although Turnitin does not

determine whether plagiarism has occurred, it serves as an essential

deterrent and educational tool in promoting academic integrity. Its

widespread adoption in educational institutions underscores the

growing role of AI in maintaining ethical standards in academic

writing.

More recently, ChatGPT and similar large language models

have revolutionized academic writing by offering advanced text

generation and idea development capabilities (OpenAI, 2023). Users

can input prompts related to topic suggestions, outlines, or even full

drafts, and receive coherent, contextually appropriate responses.

While these tools can significantly speed up the writing process and

aid brainstorming, they also raise concerns about authorship,

originality, and overreliance on AI-generated content. As such,

educators and institutions are actively exploring ways to integrate AI

tools responsibly into academic workflows while preserving the

integrity of scholarly writing.

3. The Use of AI Tools in Academic Writing

a. Types of AI Writing Assistance Tools

 19

AI writing assistance tools can be broadly categorized into

three types: grammar checkers , plagiarism detectors , and content

generators (Kessler, 2023). Grammar checkers like Grammarly and

Ginger are designed to correct spelling, grammar, and punctuation

errors, enhancing the readability and professionalism of written

work. These tools employ natural language processing (NLP)

algorithms to analyze sentence structures and suggest improvements

tailored to the intended audience and tone. They are especially useful

for students and researchers aiming to refine their manuscripts before

submission.

Plagiarism detection tools, such as Turnitin and Copyscape,

play a critical role in maintaining academic integrity by identifying

unoriginal content in documents (Park, 2003). These tools compare

submitted texts against extensive databases of published works,

internet sources, and institutional repositories to generate similarity

scores. By highlighting potential overlaps, they encourage proper

citation and attribution practices. While these tools do not

automatically flag plagiarism, they serve as preventive measures and

educational instruments for fostering responsible writing habits.

Content generation tools powered by AI, such as ChatGPT,

Google Bard, and Jasper, go beyond error correction and plagiarism

detection by assisting in drafting and ideation (Shum et al., 2018).

These tools can generate summaries, outlines, and even complete

 20

paragraphs based on user prompts, significantly reducing the time

required for initial writing stages. However, their use raises

important ethical questions regarding originality, authorship, and the

potential erosion of critical thinking skills. As AI continues to

advance, the diversity and function of writing assistance tools will

expand, influencing how academic writing is approached and

evaluated.

b. Benefits of Using AI in Writing Processes

One of the most notable benefits of using AI in academic

writing is the enhancement of efficiency and productivity (Mayfield,

2023). AI tools streamline the writing process by offering instant

feedback on grammar, coherence, and structure, allowing writers to

focus on content rather than technical correctness. For instance,

Grammarly not only identifies errors but also explains why certain

changes are suggested, thereby promoting learning and improvement

over time. This immediate feedback loop can reduce the number of

revisions needed, saving valuable time for both students and

professionals.

Another advantage is the improvement in writing quality and

clarity (Cotos, 2021). AI writing assistants can help users rephrase

awkward sentences, choose more precise vocabulary, and maintain a

consistent tone throughout a document. This is particularly beneficial

 21

for non-native English speakers who may struggle with idiomatic

expressions or complex syntactic structures. Additionally, AI tools

can assist in organizing thoughts logically, ensuring that arguments

flow coherently and supporting evidence is appropriately placed. As

a result, the overall readability and persuasiveness of academic texts

can be significantly enhanced.

Furthermore, AI contributes to enhanced accessibility and

inclusivity in academic writing (Woolf, 2010). Tools such as text-to-

speech and speech-to-text software empower individuals with

disabilities to participate more fully in academic discourse. Voice

recognition programs like Dragon NaturallySpeaking enable users to

compose documents through verbal input, while AI-driven

translation tools facilitate multilingual communication. These

innovations democratize access to academic resources and support

diverse learning needs, reinforcing the principle that high-quality

education should be accessible to all.

c. Limitations and Risks of Overreliance on AI

Despite their advantages, AI writing tools come with

limitations and risks, particularly when users become overly

dependent on them (Rada, 2022). One major concern is the potential

decline in critical thinking and writing skills . When students rely

excessively on AI-generated suggestions or entire passages, they

 22

may miss opportunities to develop their own analytical and

compositional abilities. Writing is not merely about producing error-

free text; it involves crafting arguments, synthesizing ideas, and

articulating thoughts in a meaningful way. Overreliance on AI may

hinder the development of these essential cognitive skills.

Another limitation is the lack of contextual understanding

and nuance exhibited by AI systems (Bender et al., 2021). While AI

can generate grammatically correct sentences, it may not always

grasp the subtleties of academic discourse, such as disciplinary

conventions, rhetorical strategies, or ethical considerations. For

example, a language model might generate a technically sound

paragraph that lacks relevance or fails to align with the intended

argument. This risk is heightened when users accept AI-generated

content uncritically without reviewing or modifying it according to

their specific needs.

Finally, there are ethical and academic integrity concerns

associated with AI use in writing (Heiligenstein & Zwick, 2023).

The increasing sophistication of AI tools makes it easier for students

to submit AI-generated work as their own, raising questions about

authenticity and authorship. Institutions are grappling with how to

define acceptable use policies and enforce guidelines that uphold

academic honesty. While AI can be a valuable aid, its misuse

undermines the principles of scholarship and diminishes the value of

 23

genuine intellectual effort. Therefore, it is essential to strike a

balance between leveraging AI's benefits and maintaining the

integrity of academic writing.

In addition, Capinding (2024) stated that the influence of AI

on writing is clearly seen in areas such as grammar correction,

content creation, and style improvement. As noted by Malik et al

(2023)., students generally have a favorable view of AI-powered

writing tools, recognizing their usefulness in checking grammar,

detecting plagiarism, translating languages, and organizing essay

structures. The study also found that incorporating AI into writing

processes can boost students’ writing skills, confidence, and

awareness of academic honesty.

While these tools can significantly improve writing

efficiency, some concerns remain about their potential effects on

creativity, originality, and the development of a unique writing

voice. Huang and Tan (2023) argued that excessive dependence on

AI may weaken critical and creative thinking abilities, as well as the

capacity to independently assess writing quality. Similarly, students

may become hesitant to write on their own, turning instead to AI-

driven tools like autocomplete and predictive text. These features are

easily embedded into digital writing platforms, offering suggestions

for online searches and message composition. Although students

appreciate the support these tools provide, surveys indicate worries

 24

about possible negative consequences—such as declining spelling

abilities and an increasing tendency to take the easier route through

spellcheck and AI writing software (Baron, 2023).

Furthermore, Capinding (2024) provided a validated

questionnaire to asses students’ dependency toward AI on writing.

They are such as follow:

1) I utilize AI-based spelling and grammar checkers when composing

written content.
2) I rely on AI tools for generating ideas or suggestions when drafting
written documents.
3) I use AI-driven writing assistants to enhance the overall quality of my
written work.
4) AI-generated templates or frameworks assist me in structuring my
written content.
5) I depend on AI-powered paraphrasing tools to rephrase sentences in
my written work.
6) I use AI-driven auto-complete features when composing emails or
other written correspondence.
7) AI-driven content creation tools help me generate creative or
promotional written materials.
8) I rely on AI-enhanced proofreading tools to identify errors in my
written work.
9) AI-based sentiment analysis tools assist me in gauging the tone of my
written communication.
10) I use AI-powered social media management tools to generate or
schedule written posts.
11) AI-generated suggestions for email subject lines improve the
effectiveness of my written communication.
12) I depend on AI-based content summarization tools for condensing
lengthy written materials.
13) I use AI-driven brainstorming tools to generate ideas for written
projects.
 25

14) AI-powered writing prompts or exercises help me overcome writer’s

block.
15) I rely on AI-generated suggestions for enhancing the clarity and
coherence of my written content.

P= × 100%

P = Percentage f = Frequency
n = Number of question
Lastly, this survey is an integrated and tested instrument for

establishing the level of students' dependency on artificial intelligence (AI) in

various aspects of their writing process. By covering a wide range of AI-

assisted writing activities—from simple grammar repair to advanced content

generation and creative ideation—this instrument enables researchers and

teachers to pinpoint in detail the manner in which AI technologies are being

used to learning and solo writing activities.

Moreover, the survey highlights key areas in which students are

becoming reliant on AI tools of great consequence to pedagogy, academic

honesty, and the development of autonomous writing skills. It sets the stage

as well for more research into the evolving role of AI teaching, particularly as

it relates to literacy, critical thinking, and authorship. Overall, Capinding's

(2024) empirically tested survey is a significant instrument for understanding

the influence of AI on contemporary writing practices and for guiding

reasonable discussion of responsible use of AI in education.

B. Review of Related Studies

 26

The First resesarch by Elisha L. Nañola with the tittle “Recognizing

the artificial: A comparative voice analysis of AI- Generated and L2

undergraduate student-authored academic essays” Recent developments in AI

content generation software have rendered human-authored texts

indiscernible from AI-generated ones. This progression introduced challenges

in the academic field, raising concerns over the devaluation of academic

integrity. As an inevitable aspect of written texts and an indicator of writing

identity, authorial voice is a potential distinguishing factor between both text

types. Given this, the researchers examined the differences in the authorial

voice of student-written and AI-generated essays using 12 student-written

academic essays and 12 AI-generated academic essays. The samples were

coded and analyzed using Lehman and Sułkowski's (2020) Voice Analytic

Rubric. The findings revealed that Collective (C) voice was the dominant

voice in student-written essays whereas AI-generators primarily employed

Individual (I) voice. Further comparison and closer analysis showed that: a) I-

voice is not the consistent dominant textual identity for AI-generated texts; b)

AI-generated text is closer to an expert's writing whereas student-written are

closer to a novice's; and c) the writing style of AI-generated texts lean toward

predictability. These findings contribute to understanding students' authorial

voice construction vis-a-vis the relatively underexplored authorial voice of

large language systems.

The Second resesarch by Shaun A. Hanycz Pavel Antiperovitch with

the tittle “A practical review of generative AI in cardiac electrophysiology”

 27

medical educationThe potential of AI technology to transform human life,

well-being, and daily work is faced with numerous risks and challenges yet to

be fully accounted for. However, the complexity of AI ethics makes it hard to

pin down what to teach, how to teach it, and how to assess its effectiveness.

Drawing on an educational perspective, this paper presents a systematic

literature review and qualitative analysis of the early years of AI ethics

education as a formalized field to analyze whether its future trajectory is

aligned with educational best practices. Our review highlights core challenges

in AI ethics education and the content, assessment, and pedagogy used in real

interventions over recent years. We find that efforts to teach AI ethics do

helpfully draw on a holistic view (as opposed to a narrow view), and utilize

progressive pedagogies like case studies and group projects that aim to

meaningfully challenge students’ ethical reasoning skills in applied practices.

However, many real- world AI ethics teaching interventions do not leverage

well-supported assessment techniques known to support student learning;

rather, assessment is conducted primarily for research evaluative purposes.

This gap in rigorous assessment raises implications for researchers and

practitioners, as responsible development and use of AI will be stymied if

educators cannot successfully determine whether students have truly learned

relevant AI ethics content or skills.

The Third resesarch by Okky Putra Barus with the tittle “Shaping

generative AI governance in higher education: Insights from student

perception” This study explores student perspectives on generative AI

 28

governance in higher education to ensure responsible and ethical AI

integration. Employing a mixed-methods approach with an explanatory

sequential design, the study gathered data from 111 undergraduate students at

Universitas Pelita Harapan through an online survey and semi-structured

interviews with 53 students. The study covered key aspects of AI governance,

such as ethics, curriculum integration, misuse detection, and academic

sanctions, which are crucial for ensuring responsible AI implementation. The

findings reveal that students are highly aware of the potential benefits of

Generative AI and support its integration into the curriculum. However, they

also emphasize the need for clear guidelines, ethics training, and plagiarism

detection mechanisms to prevent misuse and uphold academic integrity.

These findings provide valuable insights for higher education institutions to

develop a comprehensive AI governance framework that balances the

potential of Generative AI with the ethical considerations of its use.

The Fourth resesarch by Davy Tsz Kit Ng with the tittle

“Opportunities, challenges and school strategies for integrating generative AI

in education” The increasing accessibility of Generative Artificial

Intelligence (GenAI) tools has led to their exploration and adoption in

education. This qualitative study investigates the opportunities and challenges

associated with integrating GenAI in education, and the strategies that

encourage teachers and students to embrace GenAI in school settings. We

recruited 76 educators in Canada to participate in a professional training

seminar about GenAI and expressed their views through online surveys.
 29

Through written reflections, an optimistic outlook on GenAI's role in

education was identified among the teachers, and some discipline-specific

ideas were proposed. Thematic analysis reveals three key practices of AI

implementation: teaching/learning, administration and assessments. However,

three major challenges are also identified: school's readiness, teachers' AI

competencies, and students' AI literacy and ethics. Teachers suggest several

strategies to motivate GenAI integration, including professional development,

clear guidelines, and access to AI software and technical support. Finally,

Singh's Teach AI Global Initiative Guidance and Socio-ecological Model are

adapted and proposed to support schools in becoming AI-ready by addressing

teachers' and students' needs, facilitating organizational learning, and

promoting improvement and transformation to foster their literacy

development. Recommendations were provided for developing effective

strategies to embrace GenAI in education.

The Fifth resesarch by Marina Belkina with the tittle “Implementing

generative AI (GenAI) in higher education: A systematic review of case

studies” The introduction of Generative Artificial Intelligence (GenAI) tools,

like ChatGPT, into higher education heralds a transformative era, reshaping

instructional methods, enhancing student support systems, and redefining the

educational landscape. Recent literature reviews on GenAI highlight a lack of

focus on how these tools are being practically implemented in educational

settings. Addressing this gap, the present study systematically examines

empirical case studies that demonstrate the integration of GenAI into teaching
 30

and learning in higher education, offering actionable insights and guidance

for academic practice. We conducted a search of relevant databases and

identified 21 empirical studies that met our inclusion criteria. The selected

studies cover a diverse range of disciplines, locations, types of participants

(from first-year students to postgraduates and academics), and a variety of

methodologies. We classified the selected publications based on the

pedagogic theory of Laurillard's Conversational Framework (LCF) and the

Substitution, Augmentation, Modification, and Redefinition (SAMR)

framework. We also synthesized definitions from selected empirical studies

and recent research exploring Technological Pedagogical Content Knowledge

(TPACK) in the age of GenAI, providing a comprehensive understanding of

Gen AI-TPACK factors. Limitations and future research opportunities are

also discussed. The paper cocludes by providing a GenAI-TPACK diagram to

guide educators in effectively incorporating GenAI tools into their teaching

practices, ensuring responsible and impactful use in higher education.

Sixth resesarch by Jérémie F. Cohen with the tittle “Generative

artificial intelligence and academic writing: friend or foe?” GenAI is

revolutionizing scientific writing and publishing. While it promises enhanced

scientific productivity, it requires caution and scrutiny. Critical thinking and

accountability remain the exclusive domain of human authors. We must

safeguard the essence of scientific writing by embracing these tools

judiciously, with a commitment to the principles of scientific integrity.

Authors must declare whether they have used genAI in the production of a
 31

manuscript and provide details as to how or in what aspect of the submission

genAI was used.

Seventh resesarch by Amy Wanyu Ou with the tittle “Conceptualising

and cultivating Critical GAI Literacy in doctoral academic writing”

Generative artificial intelligence (GAI) has revolutionised the landscape of

academic writing, presenting both advantages and risks to learning for L2

writers. It is thus imperative that L2 writers, especially at advanced academic

levels, develop the critical skills necessary for employing GAI tools ethically

and effectively in their writing processes. Our study addressed this need by 1)

conceptualising Critical GAI Literacy based on current research and our

collected data, and 2) developing a self-regulated learning-based micro-

curriculum for L2 doctoral students to cultivate knowledge and skills using

GAI for academic writing. We collected interactive and reflective data in an

introductory-level academic writing course at a Swedish university enrolled

with 60 PhD students from diverse backgrounds and examined their evolving

perspectives and strategies for engaging in GAI-mediated writing. Findings

show a spectrum of initial attitudes among students and limited knowledge of

GAI use. Final reflections illustrate de-enchantment with GAI, recalibrated

and enhanced understanding of ethical issues, developed prompting methods,

and increased awareness of text ownership through the self-directed learning

process. Furthermore, students demonstrated a discerning approach in

evaluating GAI-generated suggestions and sociolinguistic impacts, indicating

a growing criticality in L2 writing practices.

 32

Ultimately, The similarity between of previous research and mine are

discussed about the use of AI in academical environment, However this

research is aim to investaigate person who is students individually, Further

discussion about motivation, habit, and engagement of student conduct

academic writing.

I. Conceptual Framework

AI in Education
↓
Use of AI Writing Tools (ChatGPT, GWEN, Deepseek, etc.)
↓
Academic Writing Output
↓
Problems Identified
→ Linguistic Issues
- Grammatically correct but semantically weak content
- Repetitive sentence structures
- Awkward phrasing and unnatural tone

→ Non-Linguistic Issues
- Overreliance on AI-generated content
- Lack of personal voice and originality
- Minimal paraphrasing or interpretation
- Violation of academic integrity norms
 33

→ Contextual & Cognitive Issues

- Incomplete understanding of topics
- Weak argument development
- Poor integration of sources

→ Institutional & Ethical Issues

- Lack of clear university policy on AI use
- Low awareness of ethical writing practices
The increasing use of AI tools like ChatGPT and Grammarly in

academic writing has raised concerns about overreliance among students.

Preliminary findings show that some students submit work with high

similarity scores (70–100%), indicating heavy dependence on AI-generated

content. This reliance is marked by minimal paraphrasing, lack of personal

voice, and repetitive sentence structures typical of AI outputs. Contributing

factors include time pressure, low confidence in writing skills, peer

normalization, and the perceived efficiency of AI tools. Overreliance hinders

critical thinking, weakens writing development, and threatens academic

integrity. Institutional policies, such as limiting AI use to below 10%, are

often unclear or unenforced. This study investigates how and why sixth-

semester English students at UIN Sjech M. Djamil Djambek use AI tools and

explores the impact of this trend on their writing quality and learning

outcomes.
34
 35

CHAPTER III

RESEARCH METHOD

A. Research Design

This study uses a qualitative case study approach to investigate

students’ reliance on AI tools in academic writing among sixth-semester

English Department students at UIN Sjech M. Djamil Djambek. The

qualitative case study design enables an in-depth examination of the

specific issues related to the extent of AI use, the factors contributing to

overreliance, and the impact of AI dependence on students’ writing skills

and academic integrity. Because research would focusing on selected

informants, the study explores how and why students use AI tools,

revealing patterns of behavior, perceptions, and underlying motivations

that shape their engagement with artificial intelligence in academic writing

tasks.

B. Research Setting

The reseacth is conducted at the English Department of the Faculty

of Tarbiyah , UIN Sjech M. Djamil Djambek, Bukittinggi, Indonesia. The

setting was chosen due to its integration of digital learning tools into the

curriculum, which includes the use of AI-based writing assistants such as

BlackBox, Deepseek, and ChatGPT. The study focuses on sixth-semester

students who are actively engaged in academic writing tasks as part of

 36

their assignment. The research setting provides a realistic context for

examining how students interact with AI tools during the writing process,

especially in relationship to essay writing, literature reviews, and research

paper.

C. Informants

The informants in this study are sixth-semester English Department

students at UIN Sjech M. Djamil Djambek who have been identified as

being highly reliant on AI tools in their academic writing. Based on initial

findings, four informants MT, AM, IS, and DA were documented to

produce written assignments with 70% to 100% similarity scores ,

categorized as high AI-generated content using detection tools such as

QuillBot and ZeroGPT . These students were selected because they

represent extreme or typical cases of AI reliance, making them ideal

subjects for understanding the underlying factors and behaviors associated

with overdependence on AI.

D. Sampling Technique

This study uses purposive sampling, a non-probability sampling

technique commonly used in qualitative research. Informants are selected

based on specific criteria relevant to the study, including their dependency

during work academic writing witb AI generated text such as Deepseek,

ChatGPT, Gwen, etc. This technique ensures that the informants have the

necessary background to provide rich and meaningful data related to the

 37

research focus, which is

 38

E. Instruments and Technique data collection

To gather comprehensive data, the study utilizes three primary

instruments and technique of ddata collection:

Questionare : Several question which give to students 6 th semester to

assess student dependency towards AI generated text. Questions Further

investigate students Reliance or Dependence including rephrasing,

enhancing also overcoming the academic assignment.

Semi-Structured Interviews : Individual interviews are conducted

with each informant using a flexible interview guide. Questions explore

students’ experiences, attitudes, motivations, and perceived benefits or

drawbacks of using AI in academic writing.

Document Analysis : Students’ academic writing assignments are

collected and analyzed to identify is it AI Generated Content, using

detection tools like ZeroGPT and QuillBot . This analysis focuses on

accuracy, originality, linguistic quality, and stylistic consistency.

G. Triangulation of Data

To ensure the trustworthiness of the data, the researcher adheres to

Lincoln and Guba’s criteria;

Credibility : Achieved through prolonged engagement, triangulation,

peer debriefing, and member checking. Researcher is engage with

 39

interviewy have question and answer deeply about their dependence of AI

while conducting academic writing.

Transferability : Supported by detailed descriptions of the research

context, participants, and procedures, allowing readers to assess the

applicability of findings to similar settings. Reader can comprehend about

the participants, procedure during research and the context then could

assess the findings.

Dependability : Ensured by maintaining an audit trail, including field

notes, audio recordings, and documentation of all stages of the research

process. This research is have concrete evidence during collecting the data

particularly on semi structured interview such as video recording and

picture.

Confirmability : Achieved by acknowledging and documenting the

researcher’s role and potential biases, ensuring that interpretations remain

grounded in the data rather than personal assumptions.

H. Technique of Data Analysis

The data are analyzed using thematic analysis , a widely used qualitative

method that involves identifying patterns and themes within the dataset.

The analytical process follows these steps:

Data Familiarization : Transcribing interviews, reviewing observation

notes, and analyzing document samples to gain a deep understanding of

the data.
 40

Initial Coding : Identifying meaningful segments of text and assigning

descriptive codes related to AI reliance, writing habits, and challenges.

Theme Development : Grouping related codes into broader themes such as

motivations for AI use , linguistic dependency , academic pressure , and

perceived benefits/risks .

Reviewing Themes : Ensuring coherence and internal consistency of each

theme and revising where necessary.

Defining and Naming Themes : Finalizing themes with clear definitions

and illustrative examples.

Reporting Findings : Presenting the results in a structured narrative format

supported by direct quotes from participants and excerpts from writing

samples.
 41

Bibliography

Fieseler, C. , F. N. , & M. M. (2023). AI Literacy: Conceptualization, Measurement, and

Validation. Computers in Human Behavior , 142, 107682.
Glikman, M. R. (2023). Overreliance on AI: When Trust Is a Bad Thing.
John McCarthy 1956. (1956). A PROPOSAL FOR THE DARTMOUTH SUMMER RESEARCH
PROJECT ON ARTIFICIAL INTELLIGENCE.
Lee, K. M. , & S. K. A. (2022). Trust in Automation: Designing Effective Human-AI Teams.
Marvin Lee Minsky. (1954). Neural Nets and the Brain Problem.
Wiese, L. J., Patil, I., Schiff, D. S., & Magana, A. J. (2025). AI ethics education: A systematic
literature review. In Computers and Education: Artificial Intelligence (Vol. 8). Elsevier
B.V. https://doi.org/10.1016/j.caeai.2025.100405