ChatGPT in education

Abstract

The aim of this study is to reveal students' thoughts on the use of ChatGPT in education. Therefore, the study was designed and conducted according to a qualitative research method. Data was collected through semi-structured interviews. The study group of the research consisted of voluntary students from one of the 4th grade classes of a public school in Türkiye. An interview form was developed and used as the data collection tool. The obtained data was analyzed using the content analysis method. Based on the results, it was observed that the students found the use of ChatGPT in teaching to be engaging and enjoyable. Moreover, incorporating artificial intelligence in teaching was found to promote academic achievement and offer a greater amount of information compared to conventional printed resources such as textbooks. Additionally, ChatGPT was perceived to provide accurate, prompt, and clear responses to queries posed by students. The students expressed their recommendation to use ChatGPT in studying other subjects like social studies, mathematics, and Turkish, as it is a valuable tool for enhancing academic success and expanding access to information.

Full text

Türk Akademik Yayınlar Dergisi (TAY Journal), 2023, 7(2), 378-406
378
http://www.tayjournal.com
https://dergipark.org.tr/tr/pub/tayjournal
ChatGPT in Education
Sayım Aktay, Assoc. Prof. Dr., Corresponding Author
Muğla Sıtkı Koçman University, Türkiye
sayimaktay@gmail.com
Orcid ID: 0000-0002-5301-0099
Seçkin Gök, PhD Student
Muğla Sıtkı Koçman University, Türkiye
seckingok@posta.mu.edu.tr
Orcid ID: 0000-0001-6095-9828
Dilşat Uzunoğlu, PhD Student
Muğla Sıtkı Koçman University, Türkiye
dilsatuzunogluu@gmail.com
Orcid ID: 0000-0003-2591-3159
Article Type: Research Article
Received Date: 03.03.2023
Accepted Date: 08.05.2023
Published Date: 08.05.2023
Plagiarism: This article has been reviewed by at least two
referees and scanned via a plagiarism software
Doi: 10.29329/tayjournal.2023.543.03
Citation: Aktay, S., Gök, S., & Uzunoğlu, D. (2023). ChatGPT in
education. Türk Akademik Yayınlar Dergisi (TAY Journal), 7(2),
378-406

ChatGPT in ….. Aktay, S., Gök, S., & Uzunoğlu, D.
379
Abstract
The aim of this study is to reveal students' thoughts on the use of ChatGPT in education. Therefore, the study
was designed and conducted according to a qualitative research method. Data was collected through semi-
structured interviews. The study group of the research consisted of voluntary students from one of the 4th grade
classes of a public school in Türkiye. An interview form was developed and used as the data collection tool. The
obtained data was analyzed using the content analysis method. Based on the results, it was observed that the
students found the use of ChatGPT in teaching to be engaging and enjoyable. Moreover, incorporating artificial
intelligence in teaching was found to promote academic achievement and offer a greater amount of information
compared to conventional printed resources such as textbooks. Additionally, ChatGPT was perceived to provide
accurate, prompt, and clear responses to queries posed by students. The students expressed their
recommendation to use ChatGPT in studying other subjects like social studies, mathematics, and Turkish, as it
is a valuable tool for enhancing academic success and expanding access to information.
Keywords: Education, artificial intelligence (ai), ChatGPT
Introduction
Technological developments have brought many new dimensions to the field of education. At
this point, educational environments have also undergone changes from a technological perspective.
Children born in this era are referred to as the digital generation and their immediate surroundings are
surrounded by technological tools. In this context, it is observed that learning environments are affected
by studies in the field of artificial intelligence (Kayahan, 2018). Artificial intelligence is a field of study
that aims to produce computers and machines (Coppin, 2004) with human-like intelligence and the
resulting innovations and developments (Chen et al., 2020). Intelligence in humans can be defined as
the ability to think, reason, judge, and draw conclusions. Artificial intelligence emerged as a result of
studies aimed at enabling machines to perform these abilities. In other words, artificial intelligence aims
to mimic the mental skills of the human brain, such as problem-solving, decision-making, prediction,
and inference (Arslan, 2020; Drigas et al., 2009; Kış, 2019). Although it is known that artificial
intelligence is a product of human intelligence and human intelligence is still superior to artificial
intelligence in many aspects (Teng, 2019), there are differences in favor of artificial intelligence in terms
of speed, accuracy, decision-making, adaptation, energy consumption, function, awareness level, and
interaction points (Aksut, 2021). AI provides opportunities to improve the teaching environment with
technologies such as the Internet of Things (IoT), virtual reality (VR), and augmented reality (AR)
(Vincent-Lancrin and van der Vlies, 2020).
The term artificial intelligence was first used by John McCarthy at a workshop held at Dartmouth
College in 1956 (Arslan, 2020; Coşkun and Gülleroğlu, 2021). In the same year, Allen Newell, Herbert A.
Simon, and Cliff Shaw presented and discussed the Logic Theorist, which is considered the first artificial
intelligence program (Fadel et al., 2019). Artificial intelligence algorithms are artificial neural networks
produced by taking inspiration from the brain's neural networks. At this point, artificial intelligence
requires dense and high-quality data. The abilities of artificial intelligence programs increase as a result
of dense and high-quality data (Kış, 2019; Kolchenko, 2018).
The use of artificial intelligence in education has a relatively short history. The first examples of
AI in education emerged in the 1960s and 1970s with the goal of personalizing education. Then, in the
1980s and 1990s, the proliferation of personal computers and the internet facilitated the development

Türk Akademik Yayınlar Dergisi (TAY Journal), 2023, 7(2), 378-406
380
of more complex educational software and online learning platforms. Since the 2000s, developments in
machine learning and natural language processing have led to the emergence of more advanced
education tools supported by artificial intelligence (Zhai, 2022). In parallel with the increase in the use
of the Internet in daily life, the use of artificial intelligence in education has also gained increasing
momentum. In this context, intelligent tools have been developed in many fields, such as natural
language processing, inference, intelligent agents, the semantic web, ontology, and speech recognition
(Drigas et al., 2009). The techniques used in education with artificial intelligence can be listed as
chatbots, expert systems, and intelligent tutoring systems (Meço and Coştu, 2022). Artificial intelligence
in education is part of the concept of advanced learning technologies, which refers to intelligent tutoring
systems that leverage artificial intelligence to determine what, how, and to whom to teach. In other
words, intelligent tutoring systems are software programs that teach complex subjects by modeling a
human teacher and offering one-on-one teaching opportunities (Akdeniz and Özdinç, 2021; Bahçeci and
Gürol, 2010; Kayahan, 2018; Keleş ve Aytürk Keleş, 2002). Intelligent tutoring systems are the most
common application of AI in education. In these systems, as the student progresses, the system
automatically adjusts the level of difficulty (Fadel et al., 2019). Examples of AI-based intelligent
education systems include platforms such as Carnegie Learning, Jill Watson, EBA ADES, and iTalk2Learn
(Çetin and Aktaş, 2021).
The use of artificial intelligence in education is accepted that gaining benefits for students such
as monitoring educational progress, providing continuous access to education, and developing the
ability to use digital assistants. On the other hand, using AI is beneficial for teachers, managing student
populations, analyzing and evaluating learning outcomes, and providing feedback (Osetskyi et al., 2020).
Kış (2019) claims that AI-based instruction increases student engagement, motivation, and
independence. In addition, AI has improved the effectiveness, efficiency, and quality of work done by AI
teachers. In this context, productivity has been considered in terms of presenting relevant content in
accordance with the curriculum, retaining information in students' minds, and taking into account
students' abilities and interests (Chen et al., 2020; Joseph, 2019). AI contributes to the personalization
of learning materials. Personalized learning is an approach that aims to customize learning according to
students' individual needs and strengths (Vincent-Lancrin and van der Vlies, 2020). Furthermore,
teachers spend 20 to 40% of their hours on activities that can be automated using current technology.
However, teachers can allocate this time to activities that support students. Additionally, the areas with
the greatest potential for automation are preparation, management, evaluation, and feedback (Bryant
et al., 2020). The use of artificial intelligence (AI) in education has progressed from the use of computers
and related technologies to web-based and online intelligent learning systems, and finally to humanoid
robots and web-based chatbots (Chen et al., 2020). Chatbots are generally AI software that provides
information or performs a task through text-based dialogue with the user (Meço and Coştu, 2022; Özkol
et al., 2019). Some chatbots are rule-based, while others are AI-based. It is claimed that AI-based
chatbots are more useful in education (Palasundram et al., 2019). Chatbots are considered to be one of
the most suitable ways for students to study (Kasthuri and Boloji, 2021). Furthermore, chatbots provide
the opportunity for students to ask questions at any time of the day, particularly in situations where the
teacher is unavailable due to the limited duration of the lesson or psychological reasons (Deveci Topal
et al., 2021). Chatbots equipped with reasoning and knowledge perform exemplification much faster
than humans and use natural language. In this respect, the use of chatbots can be seen as a tool for the
constructivist learning approach (Palasundram et al., 2019).

ChatGPT in ….. Aktay, S., Gök, S., & Uzunoğlu, D.
381
Chatbots that contribute to the individual development of students have begun to spread in the
field of education (Arruda et al., 2019; Deveci Topal et al., 2021; Nghi et al., 2019). Chatbots such as Alex,
ELIZA, megaHAL, PARRY, ANTswers, A.L.I.C.E., and GPT-3 have been developed (Kane, 2016; Osetskyi,
2020; Özkol et al., 2019). One of the developed chatbots is ChatGPT. One of the most important features
that distinguishes ChatGPT from other AI-based chatbots is that it is more successful in activities that
require creativity. ChatGPT is a general-purpose chatbot developed by OpenAI. The effects of this
chatbot on education are not known. It is thought that the impact of ChatGPT on learning goals, learning
activities, and measurement and evaluation practices will be very significant as it may lead to changes
in these areas (Zhai, 2022).
AI is being applied in different ways in educational institutions, including the automation of
administrative processes, curriculum and content development, and instruction (Chen et al., 2020).
However, it is seen that AI-based applications in education are not sufficient. AI applications are
generally used for data storage and constitute a small part of classroom activities (İşler and Kılıç, 2021).
However, AI applications can enhance the quality of education in personalized education programs,
performance monitoring, and preparation of course content (Meço and Coştu, 2022). When the
literature is examined, it can be seen that there are many studies on the use of artificial intelligence in
education (Aygün, 2019; Kabiljagić et al., 2022; Kim and Han, 2021; Kim and Park, 2017; Ottenbreit-
Leftwich et al., 2021; Ryu and Han, 2017; Shin, 2020; Shin et al., 2018; Shin and Shin, 2020; Son, 2020).
However, it can be said that there is a lack of studies on the use of AI-based chatbots in education. Studies
have been conducted with undergraduate students (Essel et al., 2022; Mokmin and Ibrahim, 2021),
middle school students (Deveci Topal, 2021), elementary school students (Kabiljagic et al., 2022), and
foreign language education (Yıldız, 2022). However, no study has been found in which ChatGPT chatbot
is used in teaching.
The aim of this study is to reveal students' thoughts on the use of ChatGPT in education. It is
believed that this study will contribute to teachers in terms of implementing artificial intelligence in the
classroom, raise awareness about the use of AI in education, provide researchers with a new perspective
on the use of AI-based chatbots, and present a positive example to education officials about the use of
chatbots in the classroom.
Method
Study Design and Participants
This research was designed and conducted with the Natural analysis approach, which is a
qualitative research method. It has gained an important place in scientific research by combining both
natural examination, research and examination (Calhoun et al., 2007). In the natural analysis approach,
the researcher does not interact with the individuals involved in the application and examines the
application in its purely natural environment. While avoiding interaction in this approach, the
researcher can also use data collection tools such as interview and product analysis (Lindlof & Taylor,
2002). The study group of the research consists of 15 4th grade students studying in a public school in
Türkiye in the 2022-2023 academic year. The study group of the research was determined by
convenient sampling method. Convenient sampling is a fast and convenient sampling strategy (Patton,
2014/2018). Therefore, a 4th grade class was determined as the study group in the primary school
where the researcher worked. In addition, there is only one 4th grade class in the school.

Türk Akademik Yayınlar Dergisi (TAY Journal), 2023, 7(2), 378-406
382
Table 1: Demographic information of the study group
Age
Gender
Knowledge of ChatGPT
Previous Teaching Experience with
ChatGPT
Participant 1
9
F
Few
No
Participant 2
9
F
Few
No
Participant 3
9
F
Few
No
Participant 4
9
M
Few
No
Participant 5
10
F
Few
No
Paticipant 6
9
F
Few
No
Participant 7
10
F
Few
No
Participant 8
10
F
Few
No
Participant 9
10
M
Few
No
Participant 10
9
M
Few
No
Participant 11
9
M
Few
No
Participant12
10
F
Few
No
Participant 13
9
F
Few
No
Participant 14
9
M
Few
No
Participant 15
10
M
Few
No
Measurement Tools
In qualitative research, data is collected through interviews, observations and documents
(Creswell, 2013/2021; Patton, 2014/2018). For this reason, the data in the research were obtained
through semi-structured interviews. The interview form was developed by the researchers and the form
was finalized by making corrections in line with the opinions of two experts from the field of basic
education. The interview form consists of two parts. The first part includes demographic information.
The second part includes interview questions. In the demographic information section, students were
also asked about their knowledge of ChatGPT and their previous teaching experience with ChatGPT.
This form consists of open-ended questions aiming to reveal students' thoughts, experiences and
needs regarding the process of teaching through ChatGPT intelligence. In form; (a) What do you think
about ChatGPT? (b) What do you think about learning with ChatGPT? (c) What did you like about
learning with ChatGPT? (d) What did you dislike about learning with ChatGPT? (d) How can lessons be
better with ChatGPT? questions are included.
Reliability of The Research
The research process has not been strictly guided by the personal judgments of the researchers.
In line with the purpose of the research, semi-structured interviews were conducted with the students.
Questions were prepared to make the students participating in the practical interviews more sensitive.
In Interviewing process, It was tried to create an interview environment in which they could freely
express their ideas, compatible with their questions. In addition, the real names of the students were
not expressed in the research and were encrypted (P1, P2...etc.). Accordingly, in the session, the
participants were given the freedom to leave the study at any point in the interviewing process. The
information obtained during the semi-structured interviews was not used by any researchers other than
the researchers in the sessions.
Participants were informed by the researcher leading the discussion that no one other than the
researchers and participants was allowed to attend the interview sessions. Again, the participants were
convinced that the obtained material would be presented in an encrypted form in the research
presentations to be made about the study in the future. Before the semi-structured interviews, an

ChatGPT in ….. Aktay, S., Gök, S., & Uzunoğlu, D.
383
informed consent form was obtained from each parent stating that they allowed their child to
participate in the study. In order to support the results reached by the researchers and to reflect the
perspectives of the students, sample quotations from the interview texts are included. On the other
hand, in order to ensure validity in the study, detailed descriptions through direct quotations were also
included (Lincoln & Guba, 1985).
Procedure
In the research, the subject of recycling, which is included in the primary school 4th
grade Science program, was handled for three weeks, two hours a week, through an artificial
intelligence software, ChatGPT.
The procedures for the implementation process are listed below:
1. Before the application, four recycling-oriented images were created in different
themes through another artificial intelligence software, DALL-E, and presented to the students.
These pictures were discussed with the students. In addition, a brief information about the
DALL-E software was given to the students. In addition, these pictures were hung on the
classroom boards by the students.
2. ChatGPT was introduced to the students with the help of the smart board in the
classroom. Then, brief information about ChatGPT is given.
3. The students were informed that the recycling issue will be handled via ChatGPT.
4. ChatGPT was asked what the concept of recycling is. ChatGPT's answer is voiced via
the ReadAload plugin. In addition, ReadAloud plugin was used in all answers given by ChatGPT.
5. ChatGPT was asked to write a story about recycling, suitable for the level of 4th grade
primary school students. It was ensured that the story written by ChatGPT was read aloud by all
students. The story was then discussed in class.
6. The question of which wastes can be recycled was asked to ChatGPT and an answer
was received. It was ensured that the answer was read by all students.
7. A blank sheet of paper was distributed to the students. Students were asked to write
two questions each to be asked to ChatGPT about recycling. Students who wanted to write more
questions were not prevented. The questions were collected and the same questions were
separated.
8. The questions written by the students were put in an empty box. Each student was
allowed to choose a question from the box. Students read aloud the questions on the paper they
drew. Each question was directed to ChatGPT.
9. Volunteer students wrote down the questions answered by ChatGPT in their
notebooks.
10. Some examples of questions asked by students to ChatGPT are given below.
Q1. What is the purpose of recycling?
Q2. Where does the garbage we throw in the trash go?
Q3. What are the benefits of recycling?

Türk Akademik Yayınlar Dergisi (TAY Journal), 2023, 7(2), 378-406
384
Q4. How are metals recycled?
Q5. How is waste glass recycled?
Q6. How are plastics recycled?
11. Students were asked to produce slogans about recycling. The slogans created by the
students were read aloud in the classroom.
12. ChatGPT was asked to write a poem about recycling appropriate for the level of 4th
grade students in primary school. This poem was read by each student. Then the students wrote
this poem in their notebooks.
13. ChatGPT was asked to prepare 10 questions in the context of the studies. It was
ensured that these questions were appropriate to the level of the students. Students gave
written or oral answers to these assessment questions.
Data Collection
After the application via ChatGPT, appointments were made to interview the volunteer
participants. The interviews were conducted face-to-face in a quiet room (Parent meeting
room) allocated by the school administrators on dates agreed with the participants. In addition,
a voice recorder was used during the interview. Approval was obtained from the parents of the
students regarding the audio recording of the interview. All of the data obtained at the end of
the interviews were transcribed and transferred to Microsoft Word.
Data Analysis
The audio recordings of the semi-structured interviews conducted in this study were
transcribed using the Microsoft Word program. After this stage, the transcribed data were read
and preliminary exploration was made by taking notes. The content analysis method was used
when analyzing the data obtained from the qualitative phase of this study (Patton, 2018;
Saldana, 2015/2019). An inductive approach was adopted in the data analysis process and the
analysis process was carried out accordingly (Lincoln & Guba, 1985). First of all, the data
obtained were coded separately by the researcher, and then themes related to these codes were
created. This whole process was carried out with qualitative research analysis software called
"Nvivo for Windows".
Ethical Permits of Research
In this study, all the rules specified to be followed within the scope of "Higher Education
Institutions Scientific Research and Publication Ethics Directive" were complied with. None of the
actions specified under the heading "Actions Contrary to Scientific Research and Publication Ethics",
which is the second part of the directive, have been taken.
Ethics Committee Permission Information:
Name of the committee that made the ethical evaluation = Muğla Sıtkı Koçman University Social
and Human Studies Ethics Committee
Date of ethical review decision = 21.02.2023
Ethics assessment document issue number= 28

ChatGPT in ….. Aktay, S., Gök, S., & Uzunoğlu, D.
385
Findings
The participants' opinions obtained through semi-structured interviews in the research were
subjected to content analysis, and as a result of this analysis, 5 themes and codes related to these themes
were identified. The themes reached are (1) General thoughts regarding teaching with ChatGPT, (2)
Positive aspects of teaching with ChatGPT, (3) Negative aspects of teaching with ChatGPT, (4)
Recommendations for teaching with ChatGPT, and (5) Applications of ChatGPT for students.
Figure 1. General thoughts regarding teaching with ChatGPT
When the theme of students' attitudes towards teaching with ChatGPT was examined, opinions
such as liking teaching with ChatGPT, finding teaching with ChatGPT more beautiful, finding teaching
with ChatGPT fun, making students happy, being different with teaching with ChatGPT, and finding
teaching with ChatGPT exciting were expressed. All students stated that they liked teaching with
ChatGPT. In this context, one student expressed their opinion as "I liked everything, actually" [P5]. Nearly
half of the students expressed that teaching with ChatGPT was more beautiful, and one student
expressed this as "It's becoming more beautiful" [P2]. Additionally, one student emphasized that teaching
with ChatGPT was fun by stating "It's really fun" [P8], while another student expressed that teaching
with ChatGPT made them happy by saying "I was happy" [P11]. Furthermore, one student mentioned
that teaching with ChatGPT was different by stating "It seemed different" [P1]. Another student
emphasized the excitement of teaching with ChatGPT by stating "I was excited at first, then it turned out
to be very good" [P23].
Based on the above views, it can be said that students appreciate and find it enjoyable to engage
in the course process using ChatGPT, one of the artificial intelligence-based software. In addition, the
artificial intelligence course process has made students happy.

Türk Akademik Yayınlar Dergisi (TAY Journal), 2023, 7(2), 378-406
386
Figure 2. Positive aspects of teaching with artificial intelligence
When examining the positive aspects of teaching with ChatGPT, it is evident that teaching with
ChatGPT contributes to academic success, provides more information, gives accurate and
understandable answers to questions, provides short and concise answers, allows for knowledge
acquisition, enables the asking of desired questions, writes text, responds promptly, makes the lesson
more enjoyable, provides knowledge, composes poems on specific subjects, allows open access to
everyone, responds in writing, writes stories on desired subjects, can answer any question, simplifies
our lives, facilitates learning of different information, includes visuals in the teaching process, enables
research with ChatGPT, and leads to effective teaching outcomes.
All of the students have stated that processing lessons with ChatGPT contributes to academic
success. In this context, two student views were expressed as "Because it can provide a good contribution
to us in our lessons" [P12] and "It does provide and has provided a little bit" [P6]. More than half of the

ChatGPT in ….. Aktay, S., Gök, S., & Uzunoğlu, D.
387
students emphasized that ChatGPT provides more information compared to textbooks and other
sources. Some of these students expressed their thoughts on this topic as "I also learn other things while
working with ChatGPT" [P3], "There is more information beyond the book" [P11], and "We obtain more in-
depth knowledge about ChatGPT than a normal science lesson" [P6]. While nearly half of the students
highlighted that ChatGPT provides accurate and understandable answers to the questions asked, one
student mentioned this topic as "ChatGPT gives correct and understandable answers" [P2]. While it is
observed that some students express their thoughts on ChatGPT giving short and clear answers, one
student mentioned this topic as "Max (the name given by the students to ChatGPT) explains briefly to us"
[P8]. One of the students emphasized the aspect of acquiring knowledge from ChatGPT as "We can ask
questions about things we have mixed up in our lessons and acquire knowledge" [P11], and another
student expressed the aspect of being able to ask desired questions to ChatGPT as "We can ask what we
want" [P6]. Additionally, one student elaborated on the ability of ChatGPT to write text as "We cannot
write text when we cannot see something or imagine it. But ChatGPT, in other words, can write a text for
us, like creating nature" [P12]. Furthermore, one student highlighted the aspect of ChatGPT providing
immediate answers to the questions asked as "Because you receive an answer. You get the answer the
moment you ask it" [P2]. Some students have found classes that are taught using ChatGPT to be more
fun. One student expressed their thoughts on this matter by stating, "I find the class both entertaining
and more informative, which is why I want to attend it" [P6]. Another student commented on the ability
of AI to provide information by stating, "It provides very good information" [P11]. Yet another student
mentioned the ability of AI to write original poetry by saying, "For example, it writes a poem when we tell
it to" [P2]. Additionally, one student emphasized the importance of AI being open source, stating, "For
instance, textbooks are written for one person, for children. AI is written for everyone" [P9]. One student
mentioned the ability of AI to respond in writing, saying, "...and it also writes to us" [P1], while another
student highlighted the ability of AI to write stories, saying, "It can write stories" [P2]. One of the positive
aspects of AI mentioned by the students is its ability to provide answers to any question. One student
remarked on this by saying, "It answered the questions I asked" [P11]. Another positive aspect mentioned
is the fact that AI makes life easier for humans. One student summarized this idea by stating, "AI makes
people's work easier" [P1]. Additionally, one student mentioned the ability of AI to facilitate learning,
saying, "We can study by asking something we don't know" [P8], and another student elaborated on this
idea by saying, "With AI, we can research things we don't know, even in our classes or in school, and find
the meaning or something else" [P12]. Based on these opinions, it is believed that ChatGPT is used as an
effective and efficient research method in both classes and other disciplines, although it may not be
prevalent. Finally, one student expressed their belief that classes are more efficient with the use of AI by
stating, "It helps us become more aware, which is why it is more efficient" [P6].
Based on the opinions above, it can be said that students liked and found the process of teaching
with ChatGPT, one of the artificial intelligence-based software, enjoyable. In addition, it is observed that
teaching with artificial intelligence contributes to academic success and provides more information than
printed sources such as textbooks. Furthermore, ChatGPT provides accurate, clear, and instant answers
to the questions asked. In short, from the perspective of students, artificial intelligence is a quality that
will make their lives easier.

Türk Akademik Yayınlar Dergisi (TAY Journal), 2023, 7(2), 378-406
388
Figure 3. Negative aspects of teaching with ChatGPT
When examining the negative aspects of teaching with ChatGPT (AI), several issues have
emerged, including limited access to AI, AI's limited ability to provide instruction, the potential for
inappropriate content generated by AI, the lack of any significant difference in AI-assisted instruction
compared to other forms of instruction, unfavorable views towards AI-generated poetry, the inability of
AI to replace books, and the potential for increased exposure to radiation.
Students commonly express difficulties in accessing AI. Two students stated, "It could be an
internet problem, that's why we couldn't get in" [24], and "It gives us an error message when we try to ask
questions" [20]. In contrast, other views were expressed less frequently (f=1). For example, one student
commented on the limited ability of AI to provide instruction, stating, "AI only explains. But normal
classes both explain and... " [22], while another student highlighted the potential for inappropriate
content generated by AI, stating, "While browsing through AI, we may come across scary things and the
like" [15]. Another student mentioned that there was no significant difference between AI-assisted
instruction and other forms of instruction, stating, "There was no difference between this and other
classes" [19]. Furthermore, one student did not appreciate the poetry generated by AI, stating, "The
poems it wrote weren't really like poetry" [18]. Only one student expressed the view that AI could not
replace books, stating, "Can it replace books? No, it can't" [13]. Additionally, one student's concern about
increased radiation exposure due to the use of AI is noteworthy. The student stated, "That's why we were
exposed to more radiation" [12]. In summary, students emphasize the access problem as the most
fundamental problem related to learning with ChatGPT.

ChatGPT in ….. Aktay, S., Gök, S., & Uzunoğlu, D.
389
Figure 4. Recommendations for teaching with ChatGPT
When examining the theme of suggestions for teaching with ChatGPT, several recommendations
have emerged. These include using ChatGPT for repeating science courses, processing social studies
courses with ChatGPT, teaching mathematics with ChatGPT, teaching Turkish with ChatGPT, using
ChatGPT for favored courses, processing all courses with ChatGPT, using ChatGPT for difficult courses,
excluding those who do not require it from application, using ChatGPT to remedy deficiencies in courses,
teaching English courses with ChatGPT, utilizing ChatGPT for mathematical operations, utilizing
ChatGPT for writing tasks in Turkish, providing personalized answers based on user age, using ChatGPT
for course repetition, enabling students to access ChatGPT individually during class for the purpose of
asking questions rather than chatting, providing more information in processed courses, and being able
to ask more unknown questions.
Most of the students have expressed their interest in revising their science lessons with ChatGPT.
In this context, one student expressed their opinion as "I want to revise my science lessons" [P3]. Almost
half of the students have expressed their interest in processing social studies, mathematics, and Turkish
lessons with ChatGPT. The students listed their suggestions as "Social studies, for example..." [P7],
"Mathematics, and something else" [P12], and "I want to process Turkish" [P1]. One student expressed
their suggestion about using ChatGPT in the students' favorite courses as "I like mathematics. I have some
difficulty with Turkish" [P5], while another student expressed the suggestion of using ChatGPT in all
courses as "Mathematics, social studies, everything" [P14]. Some students have expressed their opinions
about using ChatGPT in the courses that students find challenging, and one student emphasized this
issue as "English, as my English is not good" [P6]. One student's suggestion stands out, which argues that
"Even those who do not need it can access it. They hinder those who need it. I don't think such a thing should

Türk Akademik Yayınlar Dergisi (TAY Journal), 2023, 7(2), 378-406
390
be done" [P7], based on the fact that access to ChatGPT is not provided due to density. Additionally, it
has been seen that using ChatGPT to eliminate deficiencies in courses has been presented as a
suggestion. One student emphasized this issue as "Sometimes I can make mistakes in exams. It should be
used to correct these deficiencies" [P11]. The suggestion of processing English lessons with ChatGPT has
been expressed much less (f=1) than other courses. It was observed that one student expressed this
suggestion as "English lessons should be processed". Additionally, it has been seen that all other
suggestions were expressed by only one student. The suggestion of using ChatGPT in mathematics
operations was expressed as "We can do such an operation in mathematics" [P1], the suggestion of using
ChatGPT in Turkish writing tasks was expressed as "For example, we write something in Turkish" [P2],
the suggestion of using ChatGPT to revise the courses was expressed as "For example, if we don't
understand this subject, we can review it better on this smart board or TV, phone" [P12], the suggestion of
having individual access to ChatGPT during courses was expressed as "If everyone had a tablet or
something similar... Everyone could ask and answer questions themselves" [P3], the suggestion of using
ChatGPT for asking questions instead of chatting was expressed as "Well... If there is something to ask a
question about, it is used not in a conversation or in a ranking format" [P8], the suggestion of providing
more information during the course by researching more was expressed as "Gaining more information
by researching more" [P1], and the suggestion of asking more unknown questions was expressed as
"Well... Asking more questions about something we don't know frequently" [P9]. In summary, students
have expressed various suggestions about using ChatGPT in courses, ranging from revising courses to
writing tasks and asking questions. From another perspective, a suggestion regarding the provision of
responses by ChatGPT based on the user's age was elaborated by one of the students as follows: "So, if
there was such a thing in ChatGPT, wouldn't it be necessary to categorize it for children aged 7-11, 11-18
years old, etc.?" [P4].
Based on the above opinions, it can be seen that students suggest processing not only science
courses but also social studies, mathematics, and Turkish courses with ChatGPT. In addition, students
recommend the use of ChatGPT in reviewing lessons and completing academic deficiencies.
Figure 5. Applications of ChatGPT for students

ChatGPT in ….. Aktay, S., Gök, S., & Uzunoğlu, D.
391
When examining the theme of the applications of ChatGPT for students from the perspective of
students, it is observed that usage areas such as acquiring information on unknown topics, processing
lessons with ChatGPT, playing games with ChatGPT, obtaining information on health, and expressing
thoughts have emerged. From the perspective of students, ChatGPT is primarily used to acquire
knowledge on unknown topics and to process lessons. Students expressed these topics as "For us to
learn, to learn something that nobody knows" [P8] and "It can be used for teaching" [P6]. Additionally,
playing games with ChatGPT was also mentioned as one of the areas of usage. One of the students
mentioned this topic as "New games can be played" [P3]. Furthermore, while one student expressed the
area of obtaining health-related information as "In subjects like health" [P1], another student
emphasized the area of expressing thoughts as "For example, we can express our thoughts" [P10].
In summary, students listed acquiring knowledge, studying, playing games, and healthcare as
the areas of application for ChatGPT. It can be seen that students have limited knowledge about the areas
of application of ChatGPT. The most important reason for this may be insufficient knowledge and
experience in this context.
Discussion and Conclusion
The present study found that students enjoyed and found it fun to learn science through
ChatGPT. However, students found learning with an AI-based software like ChatGPT to be different
experience. These findings are similar to the results of many studies focusing on the use of chatbots in
education, in terms of students' emotional responses (Chen et al., 2020; Deveci Topal et al., 2021; Essel
et al., 2022; Fryer and Carpenter, 2006; Liu et al., 2022). When examining the results of these studies, it
can be seen that students liked, found interesting, and enjoyed the teaching processes that were
designed and implemented interactively with chatbots. Additionally, the chatbot named "Jill Watson,"
developed by the University of Georgia, was used in computer science classes, and students expressed
more interest in the course. In Yıldız's (2022) study, students preferred using chatbots as a resource
that provided feedback. Similarly, in Nghi et al.'s (2019) study on English language education, students
found chatbots to be fun and exciting. From another perspective, Ryu and Han (2017) examined the
images of artificial intelligence in students' minds and found that students described AI as innovative
and unprecedented. In some studies, the use of artificial intelligence in the education of children has
been evaluated as an interesting and motivating experience for researchers and teachers (Akdeniz,
2019; Kamite et al., 2019; Jia and Chen, 2009; Keleş and Aytürk Keleş, 2002; Kolchenko, 2018).
Furthermore, it can be seen that children are positively affected by the process, particularly when
chatbots are used in education. Students support the educational use of chatbots both inside and outside
the classroom, a view that is consistent with general approaches in the literature. In this context, the use
of artificial intelligence in education has been noted to have features such as personalized learning,
encouraging student participation, and stimulating student interest (Ryu and Han, 2017). However, in a
study aimed at revealing the opinions of elementary school students regarding artificial intelligence,
Shin and others (2018) found that students metaphorically described artificial intelligence as a servant,
friend, and enemy. For this reason, artificial intelligence programs should be designed considering that
some students may be adversely affected. This study also highlights that children have pointed out that
artificial intelligence can generate inappropriate content.
In this study, students expressed their belief that using ChatGPT in teaching would contribute to
their academic success. This finding is consistent with many studies conducted with chatbots. For

Türk Akademik Yayınlar Dergisi (TAY Journal), 2023, 7(2), 378-406
392
example, a chatbot developed by Lin and Chang (2020) to facilitate psychology students' thesis writing
had a significant impact on their success. Chen et al. (2020) found that a chatbot designed for foreign
language learning significantly improved student achievement. Essel et al. (2022) conducted a study
showing that interacting with a chatbot as a teaching assistant positively affected students' academic
performance. However, Yin et al. (2020) determined in their study with university students that chatbot
applications had positive effects on the experimental group's intrinsic motivation, but there was not
much difference between the performance of the experimental and control groups. Overall, it has been
stated that the use of artificial intelligence in education contributes to students' academic success
(Grudin and Jacques, 2019; İşler and Kılıç, 2021; Kim and Han, 2021; Meço and Coştu, 2022;
Pokrivcakova, 2019).
The use of chatbots in education has many positive aspects from the perspective of students. For
instance, ChatGPT provides accurate, understandable, concise, and clear answers. This is due to
ChatGPT's ability to process information in a correct, efficient, systematic, and informative manner
(Zhai, 2022). According to Kasturi and Balaji (2021), chatbots can accurately answer user questions.
Additionally, in the study conducted by Deveci Topal et al. (2021), using a chatbot application
specifically programmed for the "States of Matter" unit of the 5th-grade Science course, students were
able to ask the chatbot any question, and the chatbot was able to answer all of their questions and write
text. In this study, students reported that they could acquire knowledge related to the lessons and other
areas from ChatGPT. This finding is consistent with the results of Brandtzaeg and Folstad's (2017) study,
which aimed to identify why people use chatbots. According to Brandtzaeg and Folstad (2017), one
reason people (aged 16-55) prefer chatbots is to acquire knowledge. In short, chatbots are mostly used
to provide information to users and to answer their questions quickly (Uzun et al., 2021). Chen et al.
(2020) also stated that chatbots increase students' knowledge. In this context, obtaining reliable
information is crucial. Chatbots are an artificial intelligence-based technological product that can
provide students with reliable information (Deveci Topal et al., 2021). Students believe that ChatGPT's
ability to respond quickly is one of its positive aspects. In line with the results of Essel et al.'s (2022)
study, students expressed satisfaction with using chatbots and emphasized that chatbots provide
instant feedback. One of the findings of this study is that students emphasized the importance of open-
source artificial intelligence. Open-source AI-based programs are important in terms of providing global
access to education (Mikropoulos & Natsis, 2019). However, it should not be overlooked that open-
source AI may cause access issues, as expressed by the students in this study. In this context, Muramatsu
and Wangmo (2020) stated that students experienced stress due to network interruptions during online
education.
In this study, students have expressed multiple suggestions for the use of chatbots in education
as a result of their experiences processing science lessons with ChatGPT. For example, the students have
indicated that they would like to repeat science lessons using ChatGPT. Furthermore, the students have
expressed a desire to use this method to process other subjects, including Turkish, mathematics, English,
and social studies. In Deveci Topal et al.'s (2021) study, which was conducted using chatbots, students
also expressed a desire to use chatbots in different subjects besides science. This result is consistent
with the findings of Lipko's (2016) study, in which students expressed a desire to use chatbots in
subjects other than computer science. Additionally, chatbots have been used in teaching English (Yıldız,
2022), mathematics (Aygün, 2019; Kabiljagić et al., 2022), and social studies (Son, 2020), as well as in
science instruction (Deveci Topal et al., 2021; Shin & Shin, 2020). It can be assumed that this situation

ChatGPT in ….. Aktay, S., Gök, S., & Uzunoğlu, D.
393
is parallel to the opinions of the students who participated in the study. In this study, students have also
suggested using artificial intelligence to fill in gaps in the lessons and to review lessons. According to
Alanoğlu and Karabatak (2020), AI-based systems can be used to improve students' areas of need. In
other words, students can complete their gaps in the learned topic (Deveci-Topal et al., 2021).
Furthermore, Grudin and Jacques (2019) have indicated that students can review previous topics
through chatbots.
In this study, it was observed that students emphasized acquiring knowledge and processing
lessons as the areas where artificial intelligence can be used. This can be explained by the students' lack
of knowledge and experience in the field of artificial intelligence, which formed the study group.
Additionally, in this study, students identified healthcare as another area where artificial intelligence
can be utilized. According to the students, playing games is also one of the potential uses of artificial
intelligence. In the study conducted by Çam et al. (2021), prospective teachers ranked medical and
educational fields as the areas where artificial intelligence can be used. It is also known that chatbots
are used in the marketing industry (Pokrivcakova, 2019). In addition to these, artificial intelligence is
used in almost every aspect of our lives, such as communication, location finding, accessing information
quickly, entertainment, storage, banking, finance, social media, healthcare, and ensuring institutional
and national security (Çetin and Aktaş, 2021). In light of all these, the use of artificial intelligence in
education has the potential to greatly enhance the effectiveness of education (Zhai, 2022).
Recommedations
Based on this study, some recommendations can be provided for teachers, academics, and
officials from the Turkish Republic Ministry of National Education (MEB) who will use chatbots in
education and teaching processes. These recommendations are listed below:
1. Technological awareness should be developed in children from an early age.
2. The use of artificial intelligence in education should be expanded.
3. An education program depending on artificial intelligence should be prepared for children.
4. In-service training should be given to teachers at the point of teaching lessons through
artificial intelligence.
5. It should be ensured that children use some artificial intelligence software such as chatbots
in the presence of their parents or teachers.
6. The use of chatbots in lessons should be promoted.

Türk Akademik Yayınlar Dergisi (TAY Journal), 2023, 7(2), 378-406
394
References
Alanoğlu, M., & Karabatak, S. (2020). Eğitimde yapay zekâ [Artificial intelligence in education]. In F. Güçlü Yılmaz
& M. Naillioğlu Kaymak (Eds.), Eğitim Araştırmaları-2020 [Educational Research-2020] (pp. 175-186).
EYUDER.
Akdeniz, M., & Özdinç, F. (2021). Eğitimde yapay zeka konusunda türkiye adresli çalışmaların incelenmesi
[Examination of Studies from Turkey on Artificial Intelligence in Education]. Van Yüzüncü Yıl Üniversitesi
Eğitim Fakültesi Dergisi, 18(1), 912-932. https://doi.org/10.33711/yyuefd.938734
Arruda, D., Marinho, M., Souza, E., & Wanderley, F. (2019). A chatbot for goal-oriented requirements modeling. In
S. Misra et al. (Eds.), Computational science and its applications – ICCSA 2019. Lecture notes in computer
science, 11622 (pp. 506–519). Springer.
Arslan, K. (2020). Eğitimde yapay zekâ ve uygulamaları [Artificial Intelligence and Its Applications in Education].
Batı Anadolu Eğitim Bilimleri Dergisi, 11(1), 71-88.
Aygün, E. S. (2019). Problem çözme öğretimine yönelik oyunlaştırılmış uyarlanabilir bir zeki öğretim sisteminin
tasarlanması [Designing a gamified adaptive intelligent teaching system for problem solving teaching].
(Dissertation Number. 567098). [Master of Thesis, Trabzon Univercity], YÖK Thesis Center.
Bahçeci, F., & Gürol, M. (2010). Eğitimde akıllı öğretim sistemleri uygulamalarına yönelik bir model önerisi [A
Model Proposal for Smart Teaching Systems Applications in Education]. Engineering Sciences, 5(2), 121-
128.
Brandtzaeg, P. B. ve Folstad, A. (2017) Why people use chatbots. In I. Kompatsiaris (Ed.). Lecture Notes in Computer
Science: Vol, 10673 (pp. 22-24). Springer.
Bryant, J., Heitz, C., Sanghvi, S., & Wagle, D. (2020). How artificial intelligence will impact K-12 teachers. Retrieved
May, 12, 2020.
Çam, M. B., Çelik, N. C., Turan Güntepe, E. & Durukan, Ü. G. (2021). Öğretmen adaylarının yapay zekâ teknolojileri
ile ilgili farkındalıklarının belirlenmesi [Determining the Awareness of Teacher Candidates About
Artificial Intelligence Technologies]. Mustafa Kemal Üniversitesi Sosyal Bilimler Enstitüsü Dergisi, 18(48),
263-285.
Chen, L., Chen, P., & Lin, Z. (2020). Artificial intelligence in education: A review. Ieee Access, 8, 75264-75278.
https://doi.org/10.1109/ACCESS.2020.2988510
Coppin, B. (2004). Artificial intelligence illuminated. Jones & Bartlett Learning.
Coşkun, F., & Gülleroğlu, H. D. (2021). Yapay zekânın tarih içindeki gelişimi ve eğitimde kullanılması [The
Development of Artificial Intelligence in History and Its Use in Education]. Ankara University Journal of
Faculty of Educational Sciences (JFES), 54(3), 947-966. https://doi.org/10.30964/auebfd.916220
Creswell, J. W. (2021). Nitel Araştırma Yöntemleri. Beş yaklaşıma göre nitel araştırma ve araştırma deseni.
[Qualitative Research Methods: Qualitative Research and Research Design According to Five Approaches].
(M. Bütün ve S. B. Demir, Trans.: 6th ed.). Siyasal. (Original work published 2013).
Çetin, M., & Aktaş, A. (2021). Yapay zeka ve eğitimde gelecek senaryoları [Artificial Intelligence and Future
Scenarios in Education]. OPUS International Journal of Society Researches, 18(Eğitim Bilimleri Özel Sayısı),
4225-4268. https://doi.org/10.26466/opus.911444
Deveci Topal, A., Dilek Eren, C., & Kolburan Geçer, A. (2021). Chatbot application in a 5th grade science course.
Education and Information Technologies, 26(5), 6241-6265. https://doi.org/10.1007/s10639-021-
10627-8
Drigas, A. S., Argyri, K., & Vrettaros, J. (2009). Decade review (1999-2009): artificial intelligence techniques in
student modeling. In Best Practices for the Knowledge Society. Knowledge, Learning, Development and
Technology for All: Second World Summit on the Knowledge Society, WSKS 2009, Proceedings 2 (pp. 552-
564). Springer Berlin Heidelberg.
Essel, H. B., Vlachopoulos, D., Tachie-Menson, A., Johnson, E. E., & Baah, P. K. (2022). The impact of a virtual teaching
assistant (chatbot) on students' learning in Ghanaian higher education. International Journal of
Educational Technology in Higher Education, 19(1), 1-19. https://doi.org/10.1186/s41239-022-00362-6
Fadel, C., Holmes, W., & Bialik, M. (2019). Artificial intelligence in education: Promises and implications for
teaching and learning. The Center for Curriculum Redesign. Boston, MA.
Fryer, L., & Carpenter, R. (2006). Emerging technologies bots as language learning tools. Language Learning
&Technology, 10(3), 8–14 http://llt.msu.edu/vol10num3/emerging/

ChatGPT in ….. Aktay, S., Gök, S., & Uzunoğlu, D.
395
Grudin, G., & Jacques, R. (2019, May 4-9). Chatbots, humbots, and the quest for artificial general intelligence.
[Conference session]. CHI '19: CHI Conference on Human Factors in Computing Systems. Glasgow, Scotland,
UK.
İşler, B., & Kılıç, M. (2021). Eğitimde yapay zekâ kullanımı ve gelişimi [The Use and Development of Artificial
Intelligence in Education]. Yeni Medya Elektronik Dergisi, 5(1), 1-11.
https://dergipark.org.tr/en/pub/ejnm/issue/58097/738221
Kabiljagić, M., Wachtler, J., Ebner, M., & Ebner, M. (2022). Math trainer as a chatbot via system (push) messages for
android. International Journal of Interactive Mobile Technologies (iJIM), 16(17), 75-87.
https://doi.org/10.3991/ijim.v16i17.33351
Jia, J., & Chen, W. (2009). The further development of CSIEC project driven by application and evaluation in English
education. British Journal of Educational Technology, 40(5), 901–918. https://doi.org/10.1111/j.1467-
8535.2008.00881.x
Joseph (2019, 10 Eylül). Artificial Intelligence In Education: Uses And Applications [Blog].
https://robots.net/ai/artificial-intelligence-in-education-uses-and-applications/
Kamita, T., Ito, T., Matsumoto, A., Munakata, T., & Inoue, T. (2019). A chatbot system for mental healthcare based
on SAT counseling method. Mobile Information Systems, 2019.
Kane, D. A. (2016). The role of chatbots in teaching and learning. In E-Learning and the Academic Library: Essays
on Innovative Initiatives (pp. 131-156). McFarland: UC Irvine: Libraries.
https://escholarship.org/content/qt1hs0k71b/qt1hs0k71b.pdf
Kasthuri, E., & Balaji, S. (2021, February). A Chatbot for Changing Lifestyle in Education. In 2021 Third International
Conference on Intelligent Communication Technologies and Virtual Mobile Networks (ICICV) (pp. 1317-
1322). IEEE.
Kayahan, S. (2018, Septemper 12-14). Eğitim uygulamalarında yapay zeka: Bir derleme [Artificial intelligence in
education applications: A review] [Conference session]. 6th International Instructional Technologies and
Teacher Education Symposium. Edirne, Turkey.
Keleş, A., & Aytürk K, B. U. (2002, December). Bilgisayar destekli öğretim ve zeki öğretim sistemleri. [Computer-
assisted teaching and intelligent teaching systems]. VIII. "Türkiye'de Internet" Konferansı. İstanbul, Türkiye.
http://inet-tr. org. tr/inetconf8/bildiri/3. doc
Kış, A. (2019, May 2-4). Eğitimde Yapay Zekâ [Artificial Intelligence in Education]. 14. Uluslararası Eğitim Yönetimi
Kongresi Tam Metin Bildiri Kitabı, 197- 202. Çeşme, İzmir, Turkey.
Kim, K. J., & Han, H. J. (2021). A design and effect of maker education using educational artificial intelligence tools
in elementary online environment. Journal of Digital Convergence, 19(6), 61-71.
https://doi.org/10.14400/JDC.2021.19.6.061
Kim, K., & Park, Y. (2017). A development and application of the teaching and learning model of artificial
intelligence education for elementary students. Journal of The Korean Association of Information
Education, 21(1), 139-149. https://doi.org/10.14352/jkaie.2017.21.1.139
Kolchenko, V. (2018). Can modern AI replace teachers? Not so fast! Artificial intelligence and adaptive learning:
Personalized education in the AI age. HAPS Educator, 22(3), 249-252.
https://doi.org/10.21692/haps.2018.032
Lin, M. P. C., & Chang, D. (2020). Enhancing post-secondary writers’ writing skills with a chatbot. Journal of
Educational Technology & Society, 23(1), 78-92. https://www.jstor.org/stable/26915408
Lincoln, Y. S., & Guba, E.G. (1985). Naturalistic inquiry. Sage.
Lipko, H. (2016). Meet Jill Watson: Georgia Tech’s first AI teaching assistant. Retrieved January 18, 2023, from
https://pe.gatech.edu/blog/meet-jill-watson-georgia-techs-first-ai-teaching-assistant.
Liu, C. C., Liao, M. G., Chang, C. H., & Lin, H. M. (2022). An analysis of children’interaction with an AI chatbot and its
impact on their interest in reading. Computers & Education, 189, 104576.
https://doi.org/10.1016/j.compedu.2022.104576

Türk Akademik Yayınlar Dergisi (TAY Journal), 2023, 7(2), 378-406
396
Meço, G., & Coştu, F. (2022). Eğitimde yapay zekânın kullanılması: Betimsel içerik analizi çalışması [Using Artificial
Intelligence in Education: Descriptive Content Analysis Study]. Karadeniz Teknik Üniversitesi Sosyal
Bilimler Enstitüsü Sosyal Bilimler Dergisi, 12(23), 171-193.
https://dergipark.org.tr/en/pub/sbed/issue/70445/1092727
Mikropoulos, T. A. & Natsis, A. (2011). Educational virtual environments: A ten-year review of empirical research
(1999–2009). Comput. Edu, 56(3), 769–780. https://doi.org/10.1016/j.compedu.2010.10.020
Mokmin, N. A. M., & Ibrahim, N. A. (2021). The evaluation of chatbot as a tool for health literacy education among
undergraduate students. Education and Information Technologies, 26(5), 6033-6049.
https://doi.org/10.1007/s10639-021-10542-y
Muramatsu, K., & Wangmo, S. (2020). Comparison of traditional and online education in bhutan. Kidmore End:
Academic Conferences International Limited, 607-611. https://doi.org/10.34190/EEL.20.085.
Nghi, T.T., Phuc, T.H. & Thang, N.T. (2019). Applying Al chatbot for teaching a foreign language: An empirical
research. International Journal of Scientific & Technology Research, 8(12) 897-902.
Osetskyi, V., Vitrenko, A., Tatomyr, I., Bilan, S., & Hirnyk, Y. (2020). Artificial intelligence application in education:
Financial implications and prospects. Financial and Credit Activity Problems of Theory and Practice, 2(33),
574-584.
Ottenbreit-Leftwich, A., Glazewski, K., Jeon, M., Hmelo-Silver, C., Mott, B., Lee, S., & Lester, J. (2021, March). How do
Elementary Students Conceptualize Artificial Intelligence? In Proceedings of the 52nd ACM Technical
Symposium on Computer Science Education (pp. 1261-1261).
Özkol, İ., Doğan, K., & Köseali, G. (2019). EBYS Uygulamalarında Yapay Zekâ Destekli Chatbot (Sohbet Robotu)
Kullanımı. In B. Yalçınkaya, M.A. Ünal, B. Yılmaz & F. Özdemirci (Eds.) Bilgi Yönetimi ve Bilgi Güvenliği
[Information Management and Information Security] (pp. 229-250). E- Arşiv.
Palasundram, K., Sharef, N. M., Nasharuddin, N., Kasmiran, K., & Azman, A. (2019). Sequence to sequence model
performance for education chatbot. International Journal of Emerging Technologies in Learning (İJET),
14(24), 56-68. https://www.learntechlib.org/p/217029/.
Patton, M. Q. (2018). Nitel araştırma ve değerlendirme yöntemleri [Qualitative Research and Evaluation Methods]
(M. Butun & S.B. Demir, Trans eds.: 2th ed.). Pegem Akademi. (Original work published in 2014).
Pokrivcakova, S. (2019). Preparing teachers for the application of AI-powered technologies in foreign language
education. J. Lang Cultural Edu, 7(3), 135–153. https://doi.org/10.2478/jolace-2019-0025
Ryu, M., & Han, S. (2017). Image of artificial intelligence of elementary students by using semantic differential scale.
Journal of The Korean Association of Information Education, 21(5), 527-535.
https://doi.org/10.14352/jkaie.2017.21.5.527
Saldana, J. (2019). Nitel araştırmalar için kodlama el kitabı. [Coding Handbook for Qualitative Research] (A. Tufekci
& N. Sad, Trans eds.: 2th ed.). Pegem Akademi. (Original work published in 2015).
Shin, W. S. (2020). A case study on application of artificial intelligence convergence education in elementary
biological classification learning. Journal of Korean Elementary Science Education, 39(2), 284-295.
http://dx.doi.org/10.15267/keses.2020.39.2.284
Shin, W. S., & Shin, D. H. (2020). A study on the application of artificial intelligence in elementary science education.
Journal of Korean elementary science education, 39(1), 117-132.
Son, W. S. (2020). Development of SW education class plan using artificial intelligence education platform: focusing
on upper grade of elementary school. Journal of The Korean Association of Information Education, 24(5),
453-462. https://doi.org/10.14352/jkaie.2020.24.5.453
Teng, X. (2019, April). Discussion about artificial intelligence’s advantages and disadvantages compete with
natural intelligence. In Journal of Physics: Conference Series (Vol. 1187, No. 3, p. 032083). IOP Publishing.
Uzun, Y., Tümtürk, A. Y., & Öztürk, H. (2021, Nowember 1-3). Günümüzde ve Gelecekte Eğitim Alanında Kullanılan
Yapay Zekâ [Artificial Intelligence Used in Education Today and in the Future] [Conference session]. 1st
International Conference on Applied Engineering and Natural Sciences. Konya, Turkey.
Vincent-Lancrin, S. and R. van der Vlies (2020), "Trustworthy artificial intelligence (AI) in education: Promises and
challenges", OECD Education Working Papers, No. 218, OECD Publishing.
https://doi.org/10.1787/a6c90fa9-en.
Yildiz, Y. (2022). An Examination of the Experiences of Turkish ELLs about the Chatbot Apps to Learn English.
Canadian Journal of Language and Literature Studies, 2(5), 32-41. https://doi.org/10.53103/cjlls.v2i5.59

ChatGPT in ….. Aktay, S., Gök, S., & Uzunoğlu, D.
397
Yin, J., Goh, T. T., Yang, B., & Xiaobin, Y. (2021). Conversation technology with micro-learning: The impact of
chatbot-based learning on students’ learning motivation and performance. Journal of Educational
Computing Research, 59(1), 154-177. https://doi.org/10.1177/0735633120952067
Zhai, X. (2022). ChatGPT user experience: Implications for education. Department of Mathematics, Science, and
Social Studies Education AI4STEM Education Center University of Georgia Athens, United States. Available
at SSRN 4312418. http://dx.doi.org/10.2139/ssrn.4312418

Türk Akademik Yayınlar Dergisi (TAY Journal), 2023, 7(2), 378-406
398
BIOGRAPHICAL NOTES
Contribution Rate of Researchers
Author 1: 34%
Author 2: 33%
Author 3: 33%
Conflict Statement
There is no material or individual organic connection with the people or institutions involved in the
research and there is no conflict of interest in the research.

ChatGPT in ….. Aktay, S., Gök, S., & Uzunoğlu, D.
399
Genişletilmiş Türkçe Özet
http://www.tayjournal.com
https://dergipark.org.tr/tr/pub/tayjournal
Eğitimde ChatGPT
Giriş
İçinde bulunduğumuz çağ hız ve değişim çağı olarak nitelendirilebilir. Bu hız ve değişim en çok
teknolojik gelişmelerde kendini göstermektedir. Teknolojik gelişmeler eğitim alanına da birçok yeni
boyut kazandırmaktadır. Bu bağlamda öğrenme ortamlarının yapay zeka alanında yapılan
çalışmalardan etkilendiği görülmektedir (Kayahan, 2018). Yapay zeka, insan benzeri zekaya sahip
bilgisayar ve makineler (Coppin, 2004) üretmeyi amaçlayan bir çalışma alanı ve bunun sonucunda
ortaya çıkan yenilik ve gelişmelerdir (Chen vd., 2020). İnsanda zeka, düşünme, akıl yürütme, yargılama
ve sonuç çıkarma yeteneklerinin tamamı olarak tanımlanabilir. Bu yeteneklerin makinelerce
yapılabilmesine yönelik çalışmaların bir ürünü olarak yapay zeka ortaya çıkmıştır. Başka bir söylemle
yapay zeka insan beyninin problem çözme, karar verme tahmin ve çıkarım yapma gibi zihinsel
becerilerini taklit etmeyi amaçlamıştır (Arslan, 2020; Drigas vd., 2009; Kış, 2019).
Yapay zekanın eğitimde kullanılan teknikleri ise chatbot (sohbet robotu), uzman sistemler ve
akıllı öğretim sistemleri olarak sıralanmaktadır (Meço ve Coştu, 2022). Eğitimde yapay zeka ileri
öğrenme teknolojileri kavramının altında yer alan akıllı öğretim sistemlerinin bir parçasıdır. Akıllı
öğretim sistemleri neyi, nasıl ve kime öğreteceğini bilen, yapay zeka teknolojisinden yararlanan
bilgisayar programları olarak tanımlanabilir. Başka bir söylemle akıllı öğretim sistemleri karmaşık
konuları bir insan öğreticiyi model alarak öğreten ve bire bir öğretim fırsatı sunan yazılım
programlarıdır (Akdeniz ve Özdinç, 2021; Bahçeci ve Gürol, 2010; Kayahan, 2018; Keleş, 2002). Akıllı
öğretim sistemleri yapay zekanın eğitimdeki en yaygın uygulamasıdır. Bu sistemde öğrenci ilerledikçe
sistem otomatik olarak zorluk seviyesini ayarlamaktadır (Fadel vd., 2019). Bu bağlamda Carnegie
Learning, Jill Watson, EBA ADES, iTalk2Learn gibi platformlar yapay zeka tabanlı akıllı eğitim
sistemlerine örnek olarak verilebilir (Çetin ve Aktaş, 2021).

Türk Akademik Yayınlar Dergisi (TAY Journal), 2023, 7(2), 378-406
400
Öğrencilerin bireysel gelişimine katkı sağlayan sohbet robotları da eğitim alanında
yaygınlaşmaya başlamıştır (Arruda vd., 2019; Deveci Topal vd., 2021; Nghi vd., 2019). Alex, ELIZA,
megaHAL, PARRY, ANTswers, A.L.I.C.E. ve GPT-3 gibi sohbet robotları geliştirilmiştir (Kane, 2016;
Osetskyi, 2020; Özkol vd., 2019). Geliştirilen sohbet robotlarından biri de ChatGPT’dir. ChatGPT’yi diğer
yapay zeka tabanlı sohbet robotlarından ayıran en önemlim özelliklerden biri yaratıcılık gerektiren
aktivitelerde daha başarılı olmasıdır. ChatGPT, openAI tarafından geliştirilen genel amaçlı bir sohbet
robotudur. Bu robotun eğitim üzerindeki etkileri bilinmemektedir. ChatGPT’nin kapasitesi öğrenme
hedefleri, öğrenme etkinlikleri ve ölçme değerlendirme uygulamalarında değişikliklere yol
açabileceğinden bu alanlardaki etkisinin çok büyük olacağı düşünülmektedir (Zhai, 2022).
Alanyazın incelendiğinde yapay zekanın eğitimde kullanımını konu alan birçok çalışmanın
(Aygün, 2019; Kabiljagic vd., 2022; Kim ve Han, 2021; Kim ve Park, 2017; Ottenbreit- Leftwich vd., 2021;
Ryu ve Han, 2017; Shin, 2020; Shin ve Shin, 2020; Shin vd., 2018; Son, 2020) olduğu görülse de yapay
zeka tabanlı sohbet robotlarının eğitimde kullanımını konu alan çalışmaların yeterli olmadığı
söylenebilir. Bu noktada lisans öğrencileriyle (Essel vd., 2022; Mokmın ve Ibrahim, 2021), ortaokul
öğrencileriyle (Deveci Topal, 2021), ilkokul öğrencileriyle (Kabiljagic vd., 2022) ve yabancı dil
eğitiminde (Yıldız, 2022) çalışmalar yapılmıştır. Ancak ChatGPT sohbet robotunun öğretimde
kullanıldığı herhangi bir çalışmaya rastlanmamıştır.
ChatGPT’nin eğitimde kullanımını ortaya koymayı amaçlayan bu çalışmanın öğretmenlere yapay
zekanın sınıfta uygulanması noktasında katkı sağlayacağı, eğitimde yapay zeka kullanımına ilişkin
farkındalık kazandıracağı, araştırmacılara yapay zeka tabanlı sohbet robotunun kullanımına yönelik
yeni bir bakış kazandıracağı, MEB yetkililerine sohbet robotlarının sınıf içinde kullanılabileceğine ilişkin
olumlu bir örnek sunacağı düşünülmektedir.
Yöntem
Araştırma nitel bir araştırma olarak tasarlanmış ve yürütülmüştür. Nitel araştırma, bireylerin
veya grupların sosyal veya bireysel sorunlarına ilişkin araştırma sorularının incelenmesini içeren
yorumlayıcı çerçevelerin kullanıldğı bir araştırma yöntemidir (Creswell, 2013/2021). Araştırmanın
çalışma grubu, 2022-2023 eğitim-öğretim yılında AYDIN ili Köşk içesindeki bir devlet okulunda okuyan
15 4. sınıf öğrencisinden oluşmaktadır. Araştırmanın çalışma grubu uygun örnekleme yöntemi ile
belirlenmiştir. Uygun örnekleme hızlı ve kullanışlı bir örnekleme stratejisidir (Patton, 2014/2018). Bu
nedenle çalışma grubu ikinci araştırmacının çalıştığı devlet ilkokulundaki öğrenciler arasından
belirlenmiştir.
Nitel araştırmalarda veriler görüşmeler, gözlemler ve belgelerle toplanmaktadır (Creswell,
2013/2021; Patton, 2014/2018). Bu nedenle, araştırmadaki veriler yarı yapılandırılmış görüşmelerle
elde edilmiştir. Görüşme formu araştırmacılar tarafından geliştirilmiş ve form, temel eğitim alanındaki
iki uzmanın görüşleri doğrultusunda düzeltmeler yapılarak tamamlanmıştır.
Katılımcılar, araştırmacı ve katılımcılardan başka kimsenin görüşme oturumlarına katılmasına
izin verilmeyeceği konusunda bilgilendirilmiştir. Yine katılımcılar, elde edilen materyalin gelecekte
yapılacak araştırmalarda şifrelenmiş bir biçimde sunulacağına ikna edilmiştir. Yarı yapılandırılmış
görüşmelerden önce, her ebeveynden, çocuğunun çalışmaya katılmalarına izin verdiklerini belirten
bilgilendirilmiş bir onay formu alınmıştır. Araştırmacıların ulaştığı sonuçları desteklemek ve

ChatGPT in ….. Aktay, S., Gök, S., & Uzunoğlu, D.
401
öğrencilerin bakış açılarını yansıtmak için görüşme metinlerinden örnek alıntılar da çalışmaya dahil
edilmiştir (Lincoln ve Guba, 1985).
Araştırmada, ilkokul 4. sınıf Fen Bilimleri programına dahil olan geri dönüşüm konusu, yapay
zeka temelli bir yazılım olan ChatGPT aracılığıyla haftada iki saat, üç hafta boyunca ele alınmıştır.
ChatGPT’nin uygulam sürecinden sonra gönüllü katılımcılarla görüşmek için randevular
düzenlenmiştir. Görüşmeler, katılımcılarla kararlaştırılan tarihlerde okul yöneticileri tarafından tahsis
edilen sessiz bir odada (veli toplantı odası) yüz yüze gerçekleştirilmiştir. Ayrıca, görüşme sırasında ses
kaydedici cihaz kullanılıştır. Görümelerin kaydedileceğine ilişkin öğrenci velilerinden onay alınmıştır.
Bu çalışmada yapılan yarı yapılandırılmış görüşmelerin ses kayıtları Microsoft Word ortamına
akarılmıştır. Bu aşamadan sonra, transkript edilen veriler okunmuş ve not alınarak ön keşif yapılmıştır.
Sonra elde edilen veriler içerik analizi yöntemi ile çözümlenmiştir (Patton, 2014/2018; Saldana,
2015/2019). Yine veriler araştırmacı tarafından ayrı ayrı kodlanmıştır ve daha sonra bu kodlarla ilgili
temalar oluşturulmuştur. Tüm bu süreç " NVIVO for Windows " adı verilen nitel araştırma analiz yazılımı
ile gerçekleştirilmiştir. Veri analizi sürecinde endüktif bir yaklaşım benimsenmiştir ve analiz süreci
buna göre gerçekleştirilmiştir (Lincoln ve Guba, 1985).
Bulgular
Bu çalışmada, yarı yapılandırılmış görüşmeler yoluyla elde edilen veriler üzerinde içerik analizi
yapılmış ve bu analiz sonucunda 5 tema ve bu temalara ilişkin kodlara ulaşılmıştır. Ulaşılan temalar (1)
ChatGPT’ye yönelik genel düşünceler, (2) ChatGPT ile yapılan öğretimin olumlu yönleri, (3) ChatGPT ile
yapılan öğretimin olumsuz yönleri, (4) ChatGPT ile yapılan öğretime yönelik öneriler ve (5) Öğrenciler
perspektifinden ChatGPT’nin kullanım alanlarıdır.
ChatGPT’ye yönelik genel düşünceler teması incelendiğinde, öğrencilerin ChatGPT ile ders
işlemeyi beğendiği ve farklı bulduğu görülmüştür. Bununla birlikte ChatGPT ile Fen dersi işleme süreci
öğrencileri mutlu etmiştir. Ayrıca bu öğretim süreci öğrenciler açısından eğlenceli bulunmuştur.
ChatGPT ile yapılan öğretimin olumlu yönleri teması incelendiğinde öğrencilerin çoğunun bu yöntemle
yürütülen derslerin akdemik başarılarına olumlu yönde etki edeceğini ifade ettiği görülmüştür.
Öğrenciler ChatGPT’den doğru, anlaşılır, kısa ve net cevaplar alınabildiğini belirtmiştir. Ek olarak
öğrenciler ChatGPT’den bilgi edinmelerini bu sohbet robotunun olumlu özelliklerinden biri olarak
vurgulamıştır. Öğrenciler ChatGPT’nin metin yazabilmesini, şiir yazabilmesini, istenilen her soruya
cevap verebilmesini, sorulara hemen yanıt vermesini ve açık kaynak kodlu olmasını da bu yapay zeka
temelli yazılımın olumlu özellikleri olarak sıralamıştır.
ChatGPT ile yapılan öğretimin olumsuz yönleri teması incelendiğinde, öğrencilerin çoğunun
erişim sorununa dikkat çektiği görülmüştür. Ek olarak öğrenciler, ChatGPT’nin uygunsuz çerik
oluşturabileceği, kitapların yerini alamayacağı, radyasyona maruz kalınması ve sadece anlatım yapması
gibi olumsuz yönlerini de ifade etmiştir. ChatGPT ile yapılan öğretime yönelik öneriler teması
incelendiğinde ise öğrencilerin çoğunun tekrar ChatGPT ile Fen dersi işlemeyi önerdiği görümüştür.
Ayrıca öğrenciler Fen dersinin yanı sıra Sosyal Bilgiler, Matematik, Türkçe ve İngilizce derslerinde de
ChatGPT ‘yi kullanmayı önermiştir. Öğrenciler açısından ChatGPT sadece ders işleme amacıyla değil
derslerdeki eksiklerin tamalanması ve işlenen konuların tekrarında da kullanılmalıdır. Başka bir açıdan
öğrenciler, ChatGPT’nin ihtiyaç harici kullanılmaması gerektiği üzerinde durmuştur. ChatGPT’ye yönelik

Türk Akademik Yayınlar Dergisi (TAY Journal), 2023, 7(2), 378-406
402
önerilerden biri de sohbet robotlarının kullanıcıların yaşını dikkate alarak yanıtlarını gözden
geçirmesidir.
Öğrenciler perspektifinden yapay zekanın kullanım alanları teması incelendiğinde, öğrenciler
deneyimlerine ve yapay zekaya ilişkin bilgi düzeylerine paralel olarak yapay zekanın ders işleme ve bilgi
edinmek amacıyla kullanıldığını ifade etmiştir. Ek olarak öğrenciler yapay zekanın oyun oynamak ve
düşünceleri ifade etmek amacıyla da kullanıldığını belirtmiştir. Öğrencilerin az bir kısmının yapay
zekanın sağlık alnında kulanıldığını ifade etmesi gözden kaçırılmamalıdır.
Tartışma ve Sonuç
Bu çalışmada öğrenciler ChatGPT ile Fen dersi işlemeyi beğenmiş ve bu süreci eğlenceli
bulmuştur. Ek olarak ChatGPT'yi Fen öğretimi için kullanmak öğrencileri mutlu etmiştir. Ayrıca
öğrenciler, ChatGPT gibi yapay zeka tabanlı bir yazılımla öğrenmenin farklı olduğunu ifade etmiştir. Bu
bulgular, eğitimde sohbet robotu kullanımına odaklanan birçok çalışmanın sonuçlarıyla örtüşmektedir
(Chen vd., 2020; Deveci-Topal vd., 2021; Essel vd., 2022; Fryer ve Carpenter, 2006; Liu vd. ., 2022). Bu
çalışmaların sonuçları incelendiğinde, öğrencilerin sohbet robotları ile etkileşimli olarak tasarlanan ve
uygulanan öğretim süreçlerini beğendikleri, ilginç buldukları ve keyif aldıkları görülmektedir. Benzer
şekilde, Nghi ve diğerlerinin (2019) İngilizce eğitimi üzerine yaptığı çalışmada, öğrenciler chatbotları
eğlenceli ve heyecan verici bulmuşlardır. Başka bir bakış açısıyla, Ryu ve Han (2017), öğrencilerin
zihnindeki yapay zeka imajlarını incelemiş ve öğrencilerin yapay zekayı yeni ve heyecan verici olarak
tanımladıklarını bulmuştur. Araştırmacılar ve öğretmenler için yapay zekanın çocuk eğitiminde
kullanılması ilgi çekici ve motive edicidir (Akdeniz, 2019; Kamite vd., 2019; Jia ve Chen, 2009; Keleş ve
Aytürk Keleş, 2002; Kolchenko, 2018). Ayrıca özellikle eğitimde sohbet robotları kullanıldığında
çocukların süreçten olumlu etkilendikleri görülmektedir.
Bu çalışmada öğrenciler ChatGPT'yi öğretimde kullanmanın akademik başarılarına katkı
sağlayacağına inandıklarını ifade etmişlerdir. Bu bulgu, chatbotlarla yapılan birçok çalışma ile tutarlıdır.
Örneğin, Lin ve Chang (2020) tarafından psikoloji öğrencilerinin tez yazmalarını kolaylaştırmak için
geliştirilen birsohbet robotu, öğrencilerin tez yazma yerliliklerini önemli ölçüde arttırmıştır. Chen ve
diğerleri (2020) ise yabancı dil öğrenimi için tasarlanmış bir sohbet robotunun öğrenci başarısını
önemli ölçüde artırdığı sonucuna ulaşmıştır. Essel ve diğerleri (2022) de öğretim asistanı olarak bir
chatbot ile etkileşim kurmanın öğrencilerin akademik performansını olumlu yönde etkilediğini gösteren
bir çalışma yürütmüştür. Ancak, Yin ve diğerleri (2020), üniversite öğrencileriyle yaptıkları çalışmada,
chatbot uygulamalarının deney grubunun içsel motivasyonuna olumlu etkilerinin olduğunu ancak
deney ve kontrol gruplarının performansları arasında çok fazla fark olmadığını belirlemiştir. Genel
olarak eğitimde yapay zeka kullanımının öğrencilerin akademik başarılarına katkı sağladığı
belirtilmiştir (Grudin ve Jacques, 2019; İşler ve Kılıç, 2021; Kim ve Han, 2021; Meço ve Coştu, 2022;
Pokrivcakova, 2019). Sohbet robotlarının eğitimde kullanılmasının öğrenciler açısından birçok olumlu
yönü bulunmaktadır. Örneğin, ChatGPT doğru, anlaşılır, özlü ve net yanıtlar sağlar. Bunun nedeni,
ChatGPT'nin bilgileri doğru, verimli, sistematik ve bilgilendirici bir şekilde işleme yeteneğidir (Zhai,
2022). Kasturi ve Balaji'ye (2021) göre, sohbet robotları kullanıcı sorularını doğru bir şekilde
yanıtlayabilmektedir. Ayrıca Deveci Topal ve diğerleri (2021), 5. sınıf Fen Bilimleri dersinin "Maddenin
Halleri" ünitesi için özel olarak programlanmış bir sohbet robotu uygulaması kullanıldıkları
çalışmalarında, öğrenciler sohbet robotuna her türlü soruyu sorabilmiş ve sohbet robotu da öğrencilerin
tüm sorularını cevaplamıtır. Bu çalışmada öğrenciler ChatGPT'den derslere ve diğer alanlara ilişkin

ChatGPT in ….. Aktay, S., Gök, S., & Uzunoğlu, D.
403
bilgiler edinebileceklerini belirtmişlerdir. Bu bulgu, Brandtzaeg ve Folstad'ın (2017) insanların neden
sohbet robotu kullandığını belirlemeyi amaçlayan çalışmasının sonuçlarıyla tutarlılık göstermektedir.
Brandtzaeg ve Folstad'a (2017) göre insanların (16-55 yaş arası) sohbet robotlarını tercih etme
sebeplerinden biri bilgi edinmektir. Kısacası chatbotlar çoğunlukla kullanıcılara bilgi sağlamak ve
sorularını hızlı bir şekilde cevaplamak için kullanılmaktadır (Uzun vd., 2021).
Bu çalışmada öğrenciler, ChatGPT ile Fen dersi işleme deneyimleri sonucunda, eğitimde sohbet
robotu kullanımına yönelik birçok öneride bulunmuşlardır. Örneğin, öğrenciler Fen derslerini tekrar
ChatGPT kullanarak işlemek istediklerini belirtmiştir. Ayrıca öğrenciler bu yöntemi Türkçe, matematik,
İngilizce ve sosyal bilgiler gibi diğer dersleri işlemek için de kullanmak istediklerini ifade etmiştir.
Deveci Topal ve diğerlerinin (2021) çalışmasında öğrenciler, sohbet robotlarını Fen dersi dışında farklı
konularda da kullanmak istediklerini belirtmiştir. Bu sonuç, Lipko'nun (2016) öğrencilerin bilgisayar
bilimi dışındaki konularda da sohbet robotlarını kullanmak isteklerini ifade ettiği çalışmanın
bulgularıyla da tutarlıdır. Ayrıca sohbet robotlarının Fen öğretiminde (Deveci Topal vd., 2021; Shin ve
Shin, 2020), İngilizce öğretiminde (Yıldız, 2022), matematik öğretiminde (Aygün, 2019; Kabiljagić vd.,
2022) ve Sosyal Bilgiler (Son, 2020) öğretiminde kullanıldığı görümektedir. Bu durumun araştırmaya
katılan öğrencilerin görüşleri ile paralellik gösterdiği düşünülebilir. Ayrıca bu çalışmada öğrenciler
yapay zekanın derslerdeki boşlukları doldurmak ve dersleri gözden geçirmek için kullanılmasını da
önermiştir.
Bu çalışmada öğrencilerin yapay zekanın kullanılabileceği alanlar olarak bilgi edinme ve dersleri
işleme konularına vurgu yaptıkları görülmüştür. Bu durum çalışma grubunu oluşturan öğrencilerin
yapay zeka alanındaki bilgi ve deneyim eksiklikleri ile açıklanabilir. Ayrıca bu çalışmada öğrenciler
sağlık hizmetlerini yapay zekanın kullanılabileceği bir diğer alan olarak tanımlamışlardır. Öğrencilere
göre oyun oynamak da yapay zekanın potansiyel kullanım alanlarından biridir. Çam ve diğerlerinin
(2021) yürüttüğü çalışmada öğretmen adayları, yapay zekanın kullanılabileceği alanlar olarak tıp ve
eğitimi işaretetmiştir. Sohbet robotlarının pazarlama sektöründe de kullanıldığı bilinmektedir
(Pokrivcakova, 2019). Bunların yanı sıra iletişim, yer bulma, bilgiye hızlı erişim, eğlence, depolama,
bankacılık, finans, sosyal medya, sağlık, kurumsal ve ulusal güvenliğin sağlanması gibi hayatımızın
hemen her alanında yapay zeka kullanılmaktadır (Çetin ve Aktaş, 2021). Tüm bunların ışığında eğitimde
yapay zeka kullanımı, eğitimin etkinliğini büyük ölçüde artırma potansiyeline sahiptir (Zhai, 2022).
Öneriler
Bu çalışmadan hareketle sohbet robotlarını öğretim süreçlerinde kullanacak öğretmenlere,
akademisyenlere ve Türkiye Cumhuriyeti Milli Eğitim Bakanlığı (MEB) yetkililerine bazı önerilerde
bulunulabilir. Bu öneriler aşağıda sıralanmıştır:
1. Çocuklarda küçük yaşlardan itibaren teknolojik farkındalık geliştirilmelidir.
2. Yapay zekanın eğitimde kullanımı yaygınlaştırılmalıdır.
3. Çocuklar için yapay zeka konusunda bir eğitim programı hazırlanmalıdır.
4. Öğretmenlere derslerin yapay zeka ile işlenmesi noktasında hizmet içi eğitimler verilmelidir.
5. Çocukların, ebeveynlerinin veya öğretmenlerinin rehberliğinde yapay zeka yazılımlarını kullanmaları
sağlanmalıdır.
6. Derslerde sohbet robotlarının kullanımı teşvik edilmelidir.

Türk Akademik Yayınlar Dergisi (TAY Journal), 2023, 7(2), 378-406
404
APPENDIX 1.
Photos of The Application Process

ChatGPT in ….. Aktay, S., Gök, S., & Uzunoğlu, D.
405

Türk Akademik Yayınlar Dergisi (TAY Journal), 2023, 7(2), 378-406
406