The current issue and full text archive of this journal is available on Emerald Insight at:

https://www.emerald.com/insight/1741-5659.htm

Adoption and
What drives students toward usage of
ChatGPT? An investigation of the ChatGPT

factors influencing adoption and

usage of ChatGPT
Chandan Kumar Tiwari and Mohd. Abass Bhat Received 21 April 2023
Revised 5 August 2023
College of Economics and Business Administration, University of Technology and Accepted 12 August 2023
Applied Sciences, Muscat, Oman
Shagufta Tariq Khan
Department of Logistics, Tourism and Service Management,
German University of Technology (GUtech), Muscat, Oman, and
Rajaswaminathan Subramaniam and Mohammad Atif Irshad Khan
College of Economics and Business Administration,
University of Technology and Applied Sciences, Muscat, Oman

Abstract
Purpose – The purpose of this paper is to identify the factors determining students’ attitude toward using
newly emerged artificial intelligence (AI) tool, Chat Generative Pre-Trained Transformer (ChatGPT), for
educational and learning purpose based on technology acceptance model.
Design/methodology/approach – The recommended model was empirically tested with partial least
squares structural equation modeling using 375 student survey responses.
Findings – The study revealed that students have a favorable view of the instructional use of
ChatGPT. Usefulness, social presence and legitimacy of the tool, as well as enjoyment and motivation,
contribute to a favorable attitude toward using this tool in a learning environment. However, perceived
ease of use was not found to be a significant determinant in the adoption and utilization of ChatGPT by
the students.
Practical implications – This research is intended to benefit enterprises, academic institutions and
the global community by offering light on how students perceive the ChatGPT service in an educational
setting. Furthermore, the application enhances confidence and interest among learners, leading to
improved literacy and general awareness. Eventually, the outcome of this research will help AI
developers to improve their product and service delivery, as well as benefit regulators in regulating the
usage of AI-based bots.
Originality/value – Due to its novelty, the current research on AI-based ChatGPT usage in the
education sector is rather restricted. This study provides the adoption aspects of ChatGPT, a new AI-
based technology for students, thereby contributing significantly to the existing research on the
adoption of advanced education technologies. In addition, the literature lacks research on the adoption
of ChatGPT by students for educational purposes; this study addresses this gap by identifying adoption
determinants of ChatGPT in education.
Keywords ChatGPT, Technology acceptance model, Adoption, Education, Hedonic motivation,
Perceived social presence
Paper type Research paper Interactive Technology and Smart
Education
© Emerald Publishing Limited
1741-5659
Disclosure statement: No potential conflict of interest was reported by the author(s). DOI 10.1108/ITSE-04-2023-0061
ITSE 1. Introduction
Artificial intelligence (AI) appeared to have a transformative effect on business, health and
society as an emerging field of computer science (Çelik, 2023; Ooge et al., 2022; Vrbka and
Rowland, 2020). The origin of technologies like AI and Chatbots can be traced back to the
1950s, when scientists began exploring the concept of AI (King and ChatGPT, 2023; Oravec,
2019). The term “artificial intelligence” is defined as the capability of a system to correctly
interpret external data, to learn from such data and to apply such learnings to achieve
specific goals and tasks through flexible adaptation (Haenlein et al., 2019). AI is used to
describe computers and processes that mimic human intelligence (Tlili et al., 2023). With the
proliferation of AI technology, its applications in education have increased, with the
potential to supply personalized learning, dynamic assessments and meaningful
interactions in blended learning environments (Zhang and Aslan, 2021). However, it has
been argued that as AI becomes more and more advanced, it will ruin education and dull
down the population as more and more people rely on computers for information and
research. Researchers have further suggested that AI solutions could serve as viable
replacements for crucial faculty, staff and administrative roles (Zawacki-Richter et al., 2019;
Keiper et al., 2023). In addition, E-learning resources such as video conferencing, learning
management systems, gamification, virtual reality, AI-driven adaptive platforms,
collaboration tools and mobile apps provided a variety of educational advantages (Sankey
and Marshall, 2023) but also has drawbacks like connectivity issues, the potential for
distraction, a lack of in-person dynamics and high costs (Ali et al., 2023). By adapting
content to user needs and delivering data-driven insights, AI integration improves
personalization and efficiency. To guarantee a balanced and successful educational
experience, human engagement, teacher preparation and ethical considerations are still
essential. AI’s influence is anticipated to increase as technology develops, necessitating
ongoing efforts to solve privacy issues and preserve educators’ primary role (Nguyen et al.,
2023). An AI language model called Chat Generative Pre-Trained Transformer (ChatGPT)
has shown improvements in natural language processing (NLP) and offered individualized
educational support, but it also has drawbacks including occasionally producing inaccurate
responses and being sensitive to input phrasing (Hariri, 2023). Future advancements might
have made e-learning tools easier to use and more integrated with AI, all the while
addressing bias and data privacy issues with AI model constraints.
ChatGPT, a recently released AI tool for public testing, is generating buzz for its ability
to generate creative jokes and explain scientific topics. ChatGPT is the only one example of a
new sort of AI that has the potential to become the next significant general-purpose
technology. ChatGPT is a type of conversational interaction between GPT users introduced
by Open AI in November 2022. Due to the dialogue structure, ChatGPT can respond to next
questions, admit its mistakes, debate erroneous premises and deny inappropriate requests.
It is programmed to reply to user input in a conversational situation with human-sounding
text. The interface was created using a GPT-3.5 series model that completed its training at
the beginning of the year 2022. ChatGPT and GPT 3.5 were both taught using the AI
supercomputing capacity of Microsoft Azure. The ChatGPT model derives from the
InstructGPT model which is configured to follow an instruction and supply further
information in response to a prompt (ChatGPT, 2022; Cotton et al., 2023). According to
OpenAI (2022), ChatGPT system has been trained on a massive data set of human
conversations and can generate responses to a wide range of topics and cues. It is capable of
accurate text translation, producing marketing copy, summarizing reports and news, and
coding, among the other things. It can comprehend the meaning of lengthy texts and
producing accurate responses (Elbanna and Armstrong, 2023; Lee, 2023). The Chatbot can
be used for a variety of functions, including customer service, content development and Adoption and
language translation, creating responses in a range of languages (Keiper et al., 2023). The usage of
ChatGPT is accessible via the OpenAI Application Programming Interface, which also
enables developers to integrate the Chatbot into their own applications and systems (Open
ChatGPT
AI, 2022).
Since its inception, in a short span of time ChatGPT has garnered immense attention
from corporations and academics in the internet world. The folks around the world are using
this revolutionary technology majorly in education and research. The community is seeking
answer to the queries raised in medical science (Sharma and Thakur, 2023; Jahanshahi et al.,
2022) as well as in engineering (Qadir, 2022). One of the most widely discussed advancement
of ChatGPT is accessibility. Anyone can use the tool for learning and problem-solving.
However, being accessible to everyone can even present a threat as the tool can be used to
seek answer to support malicious intentions in the society. Furthermore, the tool is not up to
date and could not answer the questions concerning recent trends post-2021. Nevertheless,
the challenges highlighted here, ChatGPT is useful for education and learning (Dwivedi
et al., 2023; Lee, 2023). Due to its novelty, the current research on AI-based ChatGPT usage
in the education sector is rather restricted. This study provides the adoption aspects of
ChatGPT, a new AI-based technology for students, thereby contributing significantly to the
existing research on the adoption of advanced education technologies. In addition, the
literature lacks research on the adoption of ChatGPT by students for educational purposes;
this study addresses this gap by identifying adoption determinants of ChatGPT in
education.

2. Literature review and development of hypotheses

Research into Chatbots has grown in the academic literature alongside advances in AI
systems. For a long time, Chatbots relied on NLP to interpret user questions and match them
to the most relevant answers in the database. To solve NLP issues in real time and deliver
immediate responses to the clients, Chatbots have embraced language models (Dwivedi
et al., 2023; Cotton et al., 2023; Kushwaha and Kar, 2021). Researchers in the field of service
commonly believe that despite the rapid development of AI technology, the application of AI
in service delivery will not be successful or even feasible without a high level of customer
acceptance of using AI (Belanche et al., 2020; Lu et al., 2019). Some studies (Balakrishnan
et al., 2021; Lim and Zhang, 2022) relied on technology acceptance model (TAM) (Davis et al.,
1989), which proposed that users’ intention to use technology is driven by their perceived
usefulness of the technology and perceived ease of use. The TAM is one of the most
prominent models of technology acceptance, with two primary factors driving a person’s
intention to use new technology-perceived ease of use and perceived usefulness (Giovanis
et al., 2012). Because of the TAM’s simplicity, it is easy to modify or expand; thus, it has been
used in many fields of research (Min et al., 2021; Song et al., 2021). During the rapid
technological progress of the 2000s, the utility of TAM became more widely recognized.
According to the TAM requirement, the variables influencing digital banking adoption in
the Kingdom of Saudi Arabia were taken from the Global Financial Index Database (2017),
namely, perceived ease of use and perceived usefulness.
TAM has been studied in a variety of situations and contexts, including university
education, online learning and secondary education (Shroff et al., 2011; Wojciechowski and
Cellary, 2013). TAM, as previously stated, proposes four constructs: perceived usefulness
(PU), perceived ease-of-use (PEOU), attitude toward use (ATT) and behavioral intention to
use (BI). Furthermore, the model has been expanded to include “trust” as a new variable. As
a result, it is necessary to bring up the issues and challenges associated with TAM to
ITSE determine future research directions and needs. As a result, several studies have suggested
that additional variables be included in TAM. However, the previous studies proposed
extensions, perceived credibility, perceived social presence, hedonic motivation and
students’ attitude. As a result, this study incorporated these variables into TAM to assess
their compatibility.

2.1 Perceived usefulness

Perceived usefulness refers to an individual’s subjective perception of and belief in the
efficacy of implementing specific information technologies or work processes (Lin et al.,
2007). The TAM theory is based on the idea that perceived usefulness and perceived ease of
use are the most important influencing factors in predicting users’ attitudes and intentions
to use technology (Davis, 1989). The degree to which a person believes that using a
technology will improve his or her performance is defined as perceived usefulness (Davis
et al., 1989; Davis, 1993; De Luna et al., 2019). Bashir and Madhavaiah (2015) used the TAM
to study internet banking adoption in India and discovered that perceived usefulness has a
direct impact on the adoption of online banking technology. Other studies by Çelik (2008)
and Chong et al. (2013) have validated the positive effect of perceived usefulness on digital
banking adoption. The study thus proposes the following hypothesis:

H1. Perceived usefulness has a positive influence on the attitude of students to use
ChatGPT for educational purposes.

2.2 Perceived ease of use

The most important and widely proposed antecedent in the evaluation of mobile payment
adoption is perceived ease of use (Warsono et al., 2023). Both perceived ease of use and
perceived usefulness have been proposed as the two main factors determining new
technology acceptance (Belanche et al., 2019), and both are important and reliable predictors
of users’ attitude and intention toward new technologies. Davis (1989) proposed, in the
TAM, perceived ease of use has a positive influence on perceived usefulness and indirectly
affects intention to use by improving attitude toward new technologies. Some studies have
investigated the role of perceived ease of use in the adoption of mobile payments (Abdul-
Halim et al., 2022). Many studies have shown that the TAM is an effective psychometric tool
for assessing consumers’ acceptance of technology, and it has been used to explain and
predict the use of various information technologies (Abbasi et al., 2022). Several studies have
found that perceived ease of use and perceived usefulness are the important predictors of
long-term use (Peng and Lai, 2012). Furthermore, previous research has found that perceived
ease of use is strongly linked to future and current usage, as well as attitude toward users’
acceptance of a system (Davis, 1989). This demonstrates how simple it is to use the
ChatGPT and how useful it is perceived to be. Therefore, in this background the following
hypothesis is proposed:

H2. Perceived ease of use has a positive influence on the attitude of students to use
ChatGPT for educational purposes.

2.3 Perceived credibility

The concept of trust is “a person’s confidence in his or her anticipation of what the other
person will do based on their previous interactions, despite the fact that the other person’s
behaviour cannot be guaranteed” (Gefen, 2000). Suh and Han (2002) were the first to use
“trust” as a factor affecting the adoption of internet banking. Alsajjan and Dennis (2010), in a Adoption and
comparative study of the UK and Saudi Arabia on user acceptance of internet banking, usage of
underlined the need to include trust as a robust component of the TAM. According to their
research, trust increases digital banking uptake positively. Yousafzai et al. (2009) underlined
ChatGPT
that the capacity of a bank to engage in digital banking depends on the trust of its
consumers. Many studies using the TAM to study the factors influencing the adoption of
digital banking have found that consumers’ trust is positively correlated with their
willingness to adopt online banking (Bashir and Madhavaiah, 2015). Trust is a
multidisciplinary term that draws from economics, marketing, sociology, psychology,
organizational behavior, strategy and information systems (Mukherjee and Nath, 2003).
Consumers will be reluctant to adopt new technology if they lack confidence in it, which is
essential for any financial transaction (Howcroft et al., 2002). Using an extended TAM to
investigate adoption of online banking information systems, Kim et al. (2022) found that
trust had a significant positive effect on perceived usefulness and behavioral intention.
According to available literature, trust is an essential factor in determining technology
adoption behavior (Van Pinxteren et al., 2019). Recent research (Chi et al., 2021) defined trust
in interaction with AI robots as “a belief that customer’ interaction with AI social robots can
result in positive service outcomes” and developed the Social Service Robot Interaction
Trust (SSRIT) framework, which suggests that customers’ trust in human–robot interaction
is a higher-order construct resulting from the evaluation of multiple factors. According to
the SSRIT framework (Chi et al., 2021), consumers’ trust in human–robot interaction might
directly influence their desire to interact with robots. Perceived credibility is an emotional
condition in which individuals are inspired to have faith in others by their pleasing activities
(Singh and Sinha, 2020). Trust fosters the formation of enduring customer relationships and
is crucial for the adoption of new technology (Partel et al., 2019). Liebana-Cabanillas et al.
(2018) discovered that trust is crucial for establishing customer interactions and bolstering
system security. Several studies have shown that drivers’ trust in technology (whether
general or specific to driver support systems) is a crucial factor that may affect their decision
to use driver support systems (Kabra et al., 2023). This underlines the significance of the
trustworthiness of ChatGPT. The analysis generates the following hypothesis based on the
evaluated literature:

H3. Perceived credibility has a positive influence on the attitude of students to use
ChatGPT for educational purposes.

2.4 Perceived social presence

It is the perception that societal pressure exists to perform or refrain from performing a
specific behavior (Ashraf, 2018). The word relates to the social aspect of process analysis.
Individuals are more likely to comply with social expectations if they anticipate that others
will behave similarly. Alternately, it is fair for individuals to be less likely to engage in certain
behavior if social norms prohibit it (La Barbera and Ajzen, 2020). So, individuals are more
likely to engage in activities if they perceive them to be socially desirable. Numerous
empirical research has demonstrated a link between subjective norms and individuals’ actual
action intentions (Ashraf, 2019). Subjective norms are the perceived social pressures that
originate from key referents, such as family and friends, to do a specific action (Ajzen, 1991).
Moreover, according to De Luna et al. ((2019), compliance (subjective norms), internalization
(group norms) and identification (social identity) are the three dimensions of social influence.
Social impact is an influential factor in technology adoption. Chang et al. (2022) hypothesized
that social variables could be significant predictors of mobile technology adoption. Moreover,
ITSE trust is a psychological motivator of dependent behaviors and the basis of human–computer
interaction. Users’ confidence in technology may influence their attitudes and intentions
regarding its use (Kim et al., 2023). This is because the social presence will encourage
individuals to use ChatGPT. Thus, we propose the following research hypothesis:

H4. Perceived social presence has a positive influence on the attitude of students to use
ChatGPT for educational purposes.

2.5 Hedonic motivation

Hedonic motivation relates to the perceived enjoyment of engaging with AI robots in the
service delivery setting (Sehabuddin and Oktarina, 2022). When people are motivated by
hedonic wants, they are more likely to acquire faith in technology, according to a significant
corpus of technology study (Baabdullah, 2018). This is especially true given that hedonic
desire is one of the primary motivators for an action that needs the usage of a technology
(Baabdullah, 2018). According to the SSRIT framework (Chi et al., 2021), customers gain
confidence in human–robot interaction based on whether the usage of robots will match
their service expectations. According to previous research (Mishra et al., 2022), novelty is a
component of hospitality services, and social service robots encourage novel services (De
Kervenoael et al., 2020; Chang et al., 2023). Consequently, students with a strong hedonic
motivation are more likely to trust ChatGPT because of their demand for novelty, resulting
in a greater level of trust. The following hypothesis is proposed based on the reviewed
literature:

H5. Hedonic motivation has a positive influence on the attitude of students to use
ChatGPT for educational purposes.

2.6 Students attitude

An attitude is a person’s inclination to respond positively or negatively to an object, person,
institution or event, or to any other distinguishable component of the individual’s
environment (Ajzen, 1989). Attitude consists of behavioral, cognitive and affective
components (De Luna et al., 2019). Both the TAM (Davis, 1989) and the theory of planned
behavior (Ajzen, 1991) consider attitude to be an important construct that determines the
user’s behavior or intent to adopt a certain technology. Moreover, attitudes have been a
consistent determinant of mobile payment uptake (De Luna et al., 2019). According to Gupta
and Arora (2017), attitudes have a beneficial effect on the uptake of mobile banking services.
Many thinkers have argued that attitude consists of affective and cognitive components
(Kabra et al., 2023; Montano and Kasprzyk, 2015). Davis (1985) asserted that perceived
usefulness and perceived simplicity of use have beneficial effects on attitudes toward
information technology. In other words, when users’ opinions of a technology’s usefulness
and usability improve, their positive attitudes about adopting that technology are more
likely to increase. The TAM has been used to forecast and explain a wide range of
information technologies, and the hypothesized links have been largely supported (Larue
and Watling, 2021). Adu-Gyamfi et al. (2022) used TAM to explain drivers’ perceptions of
the onboard monitoring system. In addition, research indicates that perceived usefulness
can strongly influence the intention to continue using e-government systems (Hamid et al.,
2016), attitudes regarding the use of the internet for online commerce (Sahoo et al., 2022) and
the behavioral intention to use driver support systems (Rahman et al., 2018). Thus, attitude
plays a significant part for any revolutionary technology such as ChatGPT. Therefore, the Adoption and
following hypothesis is proposed by this study: usage of
H6. Students’ attitude positively influences intention to adopt ChatGPT for educational ChatGPT
purposes.

3. Material and methods

It is novel research that satisfies scholarly demands and examines ChatGPT usage by
students in an original manner. Using the TAM and context-specific constructs including
perceived usefulness (PEN), perceived ease of use (PEU), perceived social presence (PSP),
perceived credibility (PEC), hedonic motivation (HEM), attitude (AAT) and intent to adopt
ChatGPT, we present the proposed model (Figure 1). This paper demonstrates how
university and higher education officials can comprehend the perspectives of students when
using ChatGPT in education. The ChatGPT adoption factors will drive developers to
construct ChatGPT based on the needs of the students.

3.1 Sample characteristics

As shown in Table 1, our sample consists of 85.87% Omani residents and only 14.13% non-
Omani. Most of the respondents (24%) in our sample are mainly younger population, i.e.
between 18 and 22 years followed by 22.13% between 28 and 32 years and 20.27% between
23 and 27 years, 21.6% are below 18 years. However, only 12% are found above 32 years.
Majority of the respondents (53.6%) are females, and only 46.6% are male population. As far
as educational qualification of the respondents is concerned, majority are advance diploma
holders (24.81%) followed by bachelors (22.67%), diploma holders (21.07%), secondary
education (14.93%), masters (9.67%), others (6.13%) and only 1.33% respondents hold a
PhD. A large proportion of our sample has a specialization of business (30.13%) followed by
engineering (22.4%), information technology (19.2%), applied sciences (13.6%), humanities
(9.33%) and others (5.33%).

3.2 Sampling and data collection

A technique of nonprobability purposive sampling was used to obtain the responses. The target
audience for this study is ChatGPT-savvy millennials born between the middle of the 1990s and
the beginning of the 2000s. Also, millennials are seen as digitally savvy age group with
heightened trend awareness (Aslam et al., 2022; Moreno et al., 2017). Including millennials is
prompted by the AI industry’s need to pay attention to younger generations, who are more
likely to be early adopters of emerging technologies (Chattaraman et al., 2019; Szymkowiak et al.,
2021). In accordance with the methodology of Han et al. ((2021) and Wang and Chen (2019), a

Figure 1.
Research model for
the present study
ITSE Demographic variables Frequency Percentage

Age
Under 18 81 21.6
18–22 90 24
23–27 76 20.27
28–32 83 22.13
Above 32 45 12
Total 375 100
Gender
Male 174 46.4
Female 201 53.6
Total 375 100
Ethnicity
Omani 322 85.87
Non-Omani 53 14.13
Total 375 100
Education level
High School/Senior Secondary 56 14.93
Diploma 79 21.07
Advance Diploma 93 24.81
Bachelor 85 22.67
Master 34 9.67
PhD 05 1.33
Others 23 6.13
Total 375 100
Specialization
Business 113 30.13
Engineering 84 22.4
Information technology 72 19.2
Applied sciences 51 13.6
Humanities 35 9.33
Others 20 5.33
Table 1. Total 375 100
Description of the
respondents Source: Primary data

video link was provided to respondents (2019). These are the YouTube videos that demonstrate
how to use ChatGPT. The respondents were invited to complete the questions after viewing the
films. The purpose of the video is to explain to respondents how ChatGPT operates. Students
attending various public and private institutions and colleges in Oman were the intended
respondents. The poll was built using Google Documents, and connections to the site were
placed on the social media accounts of many universities so that students could access it from
January through February 2023, the beginning of the second semester of the academic year
2022–2023.
Based on the data provided by most of the management and social science studies, the
majority of academics believe that a sample size of 200–500 respondents is sufficient
(Siddiqui, 2013). The sample size can also be decided by the number of questions in the
questionnaire; for each question, 5–10 replies are acceptable (Henseler et al., 2015; Bhat et al.,
2022). Upon the completion of the survey, 375 questionnaires were collected. Thirty-five of
the surveys were excluded from the final analysis because respondents did not complete the Adoption and
questionnaires in full. All completed survey questionnaires were examined for any missing usage of
information. The items were evaluated using a five-point Likert scale, with 1 signifying
strong disagreement and 5 denoting strong agreement.
ChatGPT

3.3 Research instrument

To obtain validation of research instrument and scales, it is vital to use the apt tool to elicit
the appropriate data from respondents. In research, the methodical development, selection
and validation of scales culminates in the construction of instruments judged suitable for
data gathering. The authors of the current study modified the constructions of past studies
using the same methodology. The constructs are generated using measurements from prior
TAM-related investigations. The following constructs are explained using operational
definitions that have been revised considering the current body of literature (see Table 2).
After selecting and developing the measuring items, three subject matter experts were
engaged to guarantee their correctness and validity. After getting positive feedback from
subject matter experts, the authors undertook pretesting with 30 users of ChatGPT. The
questionnaire was greatly enhanced and restructured based on the findings of the
preliminary testing. In addition, the external and internal validity of the measures were
evaluated. The poll questions were developed using both English and Arabic (bilingually).
The study constructed a survey questionnaire with two primary components for data
collection: the first section consists of demographic information about ChatGPT, including
age, gender, education, specialization, nationality and experience, while the second section
consists of the construct components obtained from prior studies (see adapted sources in
Table 2).

4. Analysis and results

The proposed model and its hypotheses were examined in this study using the partial least
squares (PLS) approach for a few reasons. For predictive research models, the PLS is
advised, with a focus on theory development (Liengaard, 2021). Given that this study was
the first attempt to build a theoretical model that investigates the elements impacting
adoption and usage among students (Figure 2), PLS was selected because it is suitable for
exploratory research (Hair et al., 2020). In addition, the structural equation modeling (SEM)
method has grown in acceptance and is commonly used in management and educational
research (Bhat et al., 2021; Bhat and Khan, 2023). In addition, PLS-SEM is becoming more
and more popular because of its capacity to model latent constructs even for nonnormality
and small to medium-sized samples (Fernandes and Oliveira, 2021). All the variables are
created using reflective measurement models. The variation in item measurements can be
explained by variations in the constructs because latent constructs exist independently of
the measures used. To identify how the observed items were loaded on the model’s
constructs, the measurement model was first evaluated when the SmartPLS was used for
the study. The evaluation of the structural model may enable the testing of hypotheses by
analyzing the connections between the constructs.
For data analysis, the statistical software packages SPSS 24 and SMART PLS 3 were
used, primarily for testing hypotheses and building structural models (Sarstedt and Cheah,
2019). Internal reliability and validity (i.e. convergent and discriminant validity) analysis are
both involved in measurement model analysis. The internal reliability of each variable was
investigated using composite reliability (CR) and Cronbach’s alpha. The construct validity
of the statements used to evaluate the variables was examined directly at the outset of the
research. Values for Cronbach’s alpha coefficient are calculated for this. Reliability is the
ITSE Type of
No. variable Definition Indicators References

1 Perceived Degree to which a person believes the use Useful in education Davis (1989), Shen
usefulness of a technology improves his/ her Enhances the quality et al. (2022),
performance of learning Venkatesh et al.
Accomplishes tasks (2012)
more quickly
Enhances learning
effectiveness
2 Perceived ease Degree to which a person believes that Easy to use Davis (1989), Shen
of use the usage of a particular technology is Easy to master et al. (2022)
easy and without effort Uncomplicated and
less mental efforts
Interaction is clear
and understandable
3 Perceived The perception that there is personal, Feeling of being Gefen and Straub
social sociable and sensitive human contact in involved with a (2004)
presence the medium person
Feeling of
communicating with
an intelligent agent
Feeling accompanied
by an intelligent
being
Sense of human
sensitivity
4 Perceived The perception of dealing with a reliable, Trustworthy Gong and Nass
credibility trustworthy, secure and confidential Reliable (2007)
agent Safe
Confidential
5 Hedonic The fun or pleasure derived from using Enjoyable Shen et al. (2022),
motivation technology Entertaining Venkatesh et al.
Fun (2012)
6 Attitude An individual’s positive or negative Positive attitude Davis et al. (1989),
toward feelings (evaluative affect) about toward usage Shen et al. (2022)
ChatGPT performing the target behavior Makes learning
interesting
Positive attitude
toward learning
Positive general
opinion
7 Behavioral Measure of the strength of one’s intention Intention to use in Davis et al. (1989),
intention to to perform a specified behavior future Shen et al. (2022)
use ChatGPT Predicting to use for
learning experience
Frequent use
Attempt to use for
Education
Table 2.
Research instrument Source: Developed by authors based on literature
 Adoption and
usage of
ChatGPT

Figure 2.
Measurement model

term used to describe the features’ consistency and stability (Shaked et al., 2020; Khan et al.,
2020). In constructs with CR more than 0.70 and Cronbach’s alpha greater than 0.60, internal
consistency is substantially more common (Clayson, 2020). In addition, this confirms that all
structures have attained the required consistency (Table 3). Furthermore, the convergent
validity was examined using the factor loadings and the average variance extracted (AVE).
Table 3 demonstrates that all variable items’ factor loadings are more than 0.60, except for
one item related to perceived social presence, which was removed from further research.
According to Henseler et al. ((2016) and Shrestha (2021), all the AVE values for each variable
are more than 0.5, demonstrating that the variables’ convergent validity has been met.
Fornell and Larcker (1981) introduced the criterion for measuring discriminant validity,
while Franke and Sarstedt (2019) argued that the square root of AVE should be higher than
the correlation between the two components in the model. This study’s latent variables all
fulfilled that requirement, proving discriminant validity. Table 4 shows that all the AVEs
for the construct are greater than the squared interconstruct correlations and the Cronbach’s
alpha values, showing that the measurement scale is legitimate. The high level of validity of
the measurement scale is shown by a high Cronbach’s alpha coefficient. This outcome
demonstrates that discriminant validity was achieved.
A PLS-based SEM program was used to assess the study’s hypotheses. To assess the
statistical significance of the weights of the subconstructs and path coefficients, a
bootstrapping approach of 3,000 iterations was used (Ali et al., 2017). PLS does not produce
measures of overall quality of fit. To evaluate the model fit, Tenenhaus et al. ((2005)
ITSE Factor Cronbach’s Composite Average variance
Construct Items loading alpha reliability (CR) extracted (AVE) R2

Perceived usefulness PEN1 0.771 0.751 0.84 0.569

PEN2 0.775
PEN3 0.663
PEN4 0.803
Perceived ease of use PEU1 0.793 0.842 0.894 0.679
PEU2 0.851
PEU3 0.811
PEU4 0.844
Perceived social presence PSP1 0.792 0.734 0.849 0.652
PSP2 0.836
PSP3 0.794
Perceived credibility PEC1 0.751 0.809 0.876 0.639
PEC2 0.844
PEC3 0.739
PEC4 0.861
Hedonic motivation HEM1 0.767 0.656 0.814 0.593
HEM2 0.807
HEM3 0.734
Attitude toward ChatGPT AAT1 0.781 0.823 0.883 0.655 0.629
AAT2 0.884
AAT3 0.815
AAT4 0.752
Intention to adopt ChatGPT IACG1 0.767 0.714 0.822 0.536 0.421
IACG2 0.741
IACG3 0.715
IACG4 0.704
Average scores 0.617 0.525
Table 3. AVE*R2 0.323
Summary of the HAVE*R2 (GoF) 0.568
results of the
assessments Source: Smart PLS output

Intention Perceived
Perceived Hedonic to adopt Perceived Perceived social
Construct usefulness Attitude motivation ChatGPT credibility ease of use presence

Perceived usefulness 0.754

Attitude 0.603 0.809
Hedonic motivation 0.402 0.564 0.771
Intention to adopt ChatGPT 0.514 0.649 0.455 0.732
Perceived credibility 0.589 0.715 0.475 0.631 0.799
Perceived ease of use 0.750 0.699 0.45 0.567 0.767 0.824
Table 4. Perceived social presence 0.639 0.686 0.65 0.636 0.673 0.723 0.808
Fornell–Larcker
Criterion Source: SmartPLS output

introduced a diagnostic tool called the goodness-of-fit (GoF) index. The geometric mean of
the typical communality and the typical R2 are used to calculate the GoF (for endogenous
constructs). The following cutoff values were published by Hoffmann and Birnbrich (2012)
for evaluating the outcomes of the GoF analysis: GoFmedium ¼ 0.25, GoFlarge ¼ 0.36 and
GoFsmall ¼ 0.1. As demonstrated in Table 3, the model used in this investigation produced Adoption and
a GoF value of 0.568, indicating a very strong model fit. To identify the problem of usage of
multicollinearity, we also look at the variance inflation factor (VIF). The findings showed
that all the constructs’ VIF values, which ranged between 1.751 and 3.941, did not go above
ChatGPT
the permitted level of 5.0 (Liang and Shiau, 2018; Bhat and Tariq, 2022) (see Table 5). This
suggested that multicollinearity was not a major issue for this investigation.
The relationships inferred from the structural model’s postulated relationships were
tested after the measurement model and GoF. The findings of the analysis are shown in
Figure 2. The explanatory strength of the predictor variable(s) on the corresponding
construct is indicated by the corrected R2 values in Figure 2. The two dependent constructs,
attitude toward ChatGPT (R2 ¼ 0.629) and intention to adopt ChatGPT (R2 ¼ 0.421), both
reach the necessary threshold, as shown in Table 3. Based on the R2 values of 0.67, 0.33 or
0.19 respectively, Chin et al. (2008) categorized the endogenous latent variables as
considerable, moderate or weak in terms of model validity. Our findings indicate that
ChatGPT attitudes and adoption intentions can both be classified as moderate (R2 ¼ 0.629
and 0.421, respectively). In addition, the predictive relevance (Q2) and effect size (F2) of the
predicting constructs were examined to measure the model’s predictive validity (Peng and
Lai, 2012). All five of the predicting (exogenous) components have highly significant
impacts, as shown in Table 5. The predictive sample reuse method (Q2) was also used (Chin
et al., 2008). All Q2 values are positive (shown in Table 5), demonstrating the model’s
appropriate predictive relevance (Peng and Lai, 2012). Finally, the impact of the independent
constructs on the dependent variables was examined using a standardized path
examination to evaluate the hypothesized associations. Five hypotheses, i.e. H1 and H3–H6,
had positive route coefficients and were significant at the p ¼ 0.001 level, according to the
results (shown in Table 6 and Figure 2). These hypotheses are therefore supported.
However, H2 (Perceived ease of use > Attitude) was not supported.

5. Discussion
This research investigated the influence of adoption and usage of recently introduced AI-
based tool called ChatGPT among students for education based on TAM. We included the
factors, namely, perceived usefulness, perceived ease of use, perceived credibility, perceived
social pressure, hedonic motivation along with attitude and intention to use ChatGPT by the
student community. The results of our quantitative survey supported the proposed model.
H1, perceived usefulness, was found to have positive impact on the attitude of the
students using ChatGPT. This was further supported by studies involving the use of AI-
based learning tools in the previous literature (Pillai et al., 2023; Shen et al., 2022; Chocarro
et al., 2021). The students found the tool useful in education as it enhances the quality of

Constructs Q2 F2 Collinearity statistics (VIF)

Perceived usefulness 0.014 2.394

Attitude 0.384 0.727
Hedonic motivation 0.056 2.751
Intention to adopt ChatGPT 0.207 1
Perceived credibility 0.11 2.649
Perceived ease of use 0.02 3.94
Table 5.
Perceived social presence 0.019 3.091 Blindfolding process:
F2, Q2 and VIF
Source: SmartPLS output values
ITSE Standard
Original deviation t-statistics
Construct sample (b) (STDEV) (jb/STDEVj) p-values 2.50% 97.50%

H1 Perceived usefulness 0.311 0.096 3.244 0.000 0.004 0.252

! Attitude
H2 Perceived ease of use 0.172 0.097 1.769 0.078 0.025 0.367
! Attitude
H3 Perceived credibility 0.328 0.089 3.704 0.000 0.147 0.486
! Attitude
H4 Perceived social 0.247 0.088 2.874 0.005 0.421 0.306
presence ! Attitude
H5 Hedonic motivation 0.391 0.055 7.101 0.000 0.375 0.287
! Attitude
H6 Attitude ! Intention 0.649 0.044 14.854 0.000 0.565 0.726
Table 6. to adopt ChatGPT
Structural estimates:
path coefficients Source: SmartPLS output

their learning. This even assists them in accomplishing the academic task quickly and
effectively. The students from varied background could seek answers to different queries
from the AI tool to enhance their knowledge on different matters. This will make them more
informed about the concepts and developments in the field of their interest.
However, our proposed hypothesis H2 perceived ease of use and attitude was found to be
insignificant among students for learning and education. This finding is in line with Abdul-
Halim et al. ((2022) and Abbasi et al. ((2022). Students are finding it difficult to use, and they
have doubts about the tool making them skillful. Sometimes the users found the tool to be
complex and involving a lot of mental effort. This could be due to challenges faced by the
students like heavy traffic on the website while raising queries as experienced in the past
few months. As the tool is the latest technological breakthrough in information science, most
of the users are excited to explore the use causing heaving traffic to the Open AI server.
Furthermore, the responses provided by the ChatGPT are not clear and understandable by
the user often. This may demotivate some students to use those who look for a quick fix to a
problem. Nevertheless, the developer of the tool has taken note of this issue and thereby
strengthening the user interface (Open AI, 2022).
The study further revealed that attitude to use ChatGPT was found to be significant and
positively influenced by perceived creditability (H3) (Kim et al., 2022; Kabra et al., 2023).
Students find it trustworthy and reliable tool for learning and education. They resort to the
tool for different questions and solve problems. They further consider the tool is safe to use
and found that the responses provided shall be confidential. This denotes that the data
provided, or questions asked to the bot will be confidential and at the same time students
can trust the answers given to be genuine and confidential. Students are found to be
involved with ChatGPT as they were communicating with an intelligent agent. This is
evident through our findings supporting H4 where perceived social presence is a significant
factor positively influencing attitude of students toward ChatGPT, as same results were
found by Chang et al. (2022) and Kim et al. (2023). The feeling of being accompanied by an
intelligent being makes the user’s attitude positive. Furthermore, it has been observed that
the tool has a sense of human sensitivity promoting students’ engagement with the bot.
In yet another finding, hedonic motivation has played a major role in developing positive
attitude among students’ user for ChatGPT, as similar was found by Sehabuddin and
Oktarina (2022), Mishra et al. (2022) and Chang et al. (2023). The respondents find the tools Adoption and
quite simulative and entertaining. They found the responses provided by the tool to be usage of
entertaining. Nevertheless, the technology has taken everyone as a storm as there exists a
game element present for the users where you feel excited while receiving the answers from
ChatGPT
an artificial brain. Among the respondents, the overall attitude is found to be positive and
significant. The student community welcomes this innovation with a quite robust outlook.
They are confident of using ChatGPT for learning and education and found the tool quite
interesting and informative. This could be the positive sign for the education technology
companies where their robust innovation can really bring results for the intended users.
Finally, when it came to measuring the behavioral intention to use the learning-based AI
tool, the student respondents shown positive intent stating that they will use ChatGPT for
studies in the future (Kabra et al., 2023; Adu-Gyamfi et al., 2022; Sahoo et al., 2022). They
have decided to use the bot frequently to enhance their learning experience in the times to
come. This favors that the tool will certainly have a bright future and will be the
revolutionary innovation for the students studying in schools and colleges.
Overall, the study observed the positive attitude among students for adoption of
ChatGPT for education and learning. The usefulness, social presence, credibility of the tool
along with the fun and motivation are variables creating the positive behavioral intention to
use this tool in education and learning environment.

6. Implications of the study

The present study will have numerous implications for the global community, corporations
and academicians. First, the several factors highlighted in the study will make the
technology more accessible to the masses. The usefulness, social presence, credibility of the
tool along with the fun and motivation are the variables creating the positive behavioral
intention to use this tool in education and learning environment. There would be enhanced
literacy and general awareness among people as they continue to pursue AI assistance for
seeking answers as the tool is useful for education and learning as highlighted by this
research. This will have a revolutionary impact on the learning and development of human
brains as the information presented through computer interaction is more comprehensive.
Second, the motivation gained by the users while interacting with ChatGPT is quite
optimistic. At the same time, the users are finding the source of answer provided by the bot
to be credible. The tool has acquired widespread popularity and is being discussed in both
online and offline social settings. Thus, students are confident of using this tool. These
characteristics as outlined by our study can even benefit other community as well beyond
students. Executives, managers, customers, shareholders and other important stakeholders
associated with the corporations would also be more informed on different matters of their
interest. They can get answers to the questions on marketing, finance, accounting,
economics, operations, human resource development and any other field generated by
ChatGPT. They can also resort to answers to the problems faced in daily operations if not to
complex dealings. Third, the abundance of information generated by ChatGPT may be of
interest to academics. This could necessitate a search for a vast quantity of academic
content by the researchers. Even researchers and scientists may find it convenient to collect
data, answer research queries and compose reports and theses. Due to the bot’s interactivity,
this would facilitate education and learning for the wider population and academic
community. This may assist colleges and universities in producing graduates with the
strong abilities required by outside organizations. ChatGPT can be effectively incorporated
into the classroom to automate mundane tasks and improve students’ educational
experiences to increase the efficiency and promote adaptive learning.
ITSE 7. Conclusion and future research directions
Using quantitative method, we investigated the users’/students’ attitude toward ChatGPT
adoption. The suggested framework expands the TAM through perceived social presence.
Using the TAM and context-specific constructs such as PEN, PEU, PSP, PEC, HEM, AAT
and intention to adopt ChatGPT, we presented a model that explains the behavioral
intention of students to adopt and use ChatGPT for educational and learning purposes. The
model has been empirically studied and statistically validated, as it sufficiently explains
62.9% of attitudes and 42.1% of intentions to adopt ChatGPT. The research reveals that the
student population is motivated to use the ChatGPT tool and that they found it to be
beneficial and credible in educational contexts.
Nevertheless, alike any other technology, the ChatGPT will have some serious challenges.
The government, companies and educational institutes must regulate the usage of ChatGPT for
ethical and right reasons. Otherwise, it would be detrimental in the learning and development
of youth and individuals. Factual errors, the encouragement of bias, a lack of in-depth
comprehension and safety concerns are just some of ChatGPT’s drawbacks that should be kept
in mind, even after updates. It has been argued that there is a possibility that ChatGPT could be
trained on data which are not free from errors and the training process could have some gaps.
This results into inaccurate output generated by the bot. Students and learners seeking for any
information could get the wrong results affecting their learning and career advancement (Lee,
2023). For the researcher’s community, it has been found that the system produces fake
citations thereby negatively impacting the literature review process (Cooper, 2023). This can
affect the research and publications for authors and scholars (Perkins, 2023). The abundance of
useful information for users on a click would at the same time discourage innovation and
curiosity to solve problems by human brains as the artificial brains would be at work. Limitless
use of artificial brains can slow down the growth of human brains at least in children and
young adults. To overcome such challenges, the technology companies and governments have
a role to play in creating how much answers the AI tools like ChatGPT should provide and how
much remains for the human minds to solve on their own.
The present research is limited by the sample size and universe which is restricted to a
particular geography of Oman. Moreover, it measures the intention to adopt and use ChatGPT by
the students in education which is altogether a novel idea. The collected data is also based on the
students’ responses, which is a result of experience gained in a short span of time. The Open AI
tool ChatGPT is the latest innovation in AI and has seen quite robust response from the
community worldwide. Therefore, it will be worthwhile to watch how the tools progress with the
passage of time? Whether it enhances students learning or make them technology dependent?
Whether it is a boon or a bane for the learners’ community? Whether ChatGPT can be a
disruptive technology changing the landscape for information seekers? Researchers can address
these questions. Future scholars can also consider studies with larger sample sizes in different
countries. Moreover, the resultant impact of using ChatGPT among learners, scholars and
academicians may be examined in the schools, colleges or similar other organizational settings.

References
Abbasi, G.A., Sandran, T., Ganesan, Y. and Iranmanesh, M. (2022), “Go cashless! Determinants of
continuance intention to use E-wallet apps: a hybrid approach using PLS-SEM and fsQCA”,
Technology in Society, Vol. 68, p. 101937, doi: 10.1016/j.techsoc.2022.101937.
Abdul-Halim, N.A., Vafaei-Zadeh, A., Hanifah, H., Teoh, A.P. and Nawaser, K. (2022), “Understanding
the determinants of e-wallet continuance usage intention in Malaysia”, Quality and Quantity,
Vol. 56 No. 5, pp. 3413-3439, doi: 10.1007/s11135-021-01276-7.
Adu-Gyamfi, G., Song, H., Nketiah, E., Obuobi, B., Adjei, M. and Cudjoe, D. (2022), “Determinants of Adoption and
adoption intention of battery swap technology for electric vehicles”, Energy, Vol. 251, p. 123862,
doi: 10.1016/j.energy.2022.123862.
usage of
Ajzen, I. (1989), “Attitude structure and behavior”, in Pratkanis, A.R., Breckler, S.J. and Greenwald, A.
ChatGPT
G. (Eds), Attitude Structure and Function, Lawrence Erlbaum Associates, pp. 241-274.
Ajzen, I. (1991), “The theory of planned behavior”, Organizational Behavior and Human Decision
Processes, Vol. 50 No. 2, pp. 179-211.
Ali, M., Wood-Harper, T. and Wood, B. (2023), “Understanding the technical and social paradoxes of
learning management systems usage in higher education: a sociotechnical perspective”, Systems
Research and Behavioral Science, doi: 10.1002/sres.2945.
Ali, F., Hussain, K., Konar, R. and Jeon, H.M. (2017), “The effect of technical and functional quality
on guests’ perceived hotel service quality and satisfaction: a SEM-PLS analysis”, Journal of
Quality Assurance in Hospitality and Tourism, Vol. 18 No. 3, pp. 354-378, doi: 10.1080/
1528008X.2016.1230037.
Alsajjan, B. and Dennis, C. (2010), “Internet banking acceptance model: cross-market examination”,
Journal of Business Research, Vol. 63 Nos 9/10, pp. 957-963, doi: 10.1016/j.jbusres.2008.12.014.
Ashraf, M.A. (2018), “Use of bounded rationality theory to understand participation of women in
Islamic microfinance”, Enterprise Development and Microfinance, Vol. 29 No. 3, pp. 186-208,
doi: 10.3362/1755-1986.18-00005.
Ashraf, M.A. (2019), “Islamic marketing and consumer behavior toward halal food purchase in
Bangladesh: an analysis using SEM”, Journal of Islamic Marketing, Vol. 10 No. 3, pp. 893-910,
doi: 10.1108/JIMA-03-2018-0051.
Aslam, W., Ahmed Siddiqui, D., Arif, I. and Farhat, K. (2022), “Chatbots in the frontline: drivers of
acceptance”, Kybernetes, doi: 10.1108/K-11-2021-1119.
Baabdullah, A.M. (2018), “Consumer adoption of mobile social network games (M-SNGs) in Saudi
Arabia: the role of social influence, hedonic motivation and trust”, Technology in Society, Vol. 53,
pp. 91-102, doi: 10.1016/j.techsoc.2018.01.004.
Balakrishnan, J., Dwivedi, Y.K., Hughes, L. and Boy, F. (2021), “Enablers and inhibitors of AI-powered
voice assistants: a dual-factor approach by integrating the status quo bias and technology
acceptance model”, Information Systems Frontiers, doi: 10.1007/s10796-021-10203-y.
Bashir, I. and Madhavaiah, C. (2015), “Consumer attitude and behavioural intention towards internet
banking adoption in India”, Journal of Indian Business Research, Vol. 7 No. 1, pp. 67-102,
doi: 10.1108/JIBR-02-2014-0013.
Belanche, D., Casalo, L.V. and Flavian, C. (2019), “Artificial intelligence in FinTech: understanding
robo-advisors adoption among customers”, Industrial Management and Data Systems, Vol. 119
No. 7, pp. 1411-1430, doi: 10.1108/IMDS-08-2018-0368.
Belanche, D., Casalo, L.V., Flavian, C. and Schepers, J. (2020), “Service robot implementation: a
theoretical framework and research agenda”, The Service Industries Journal, Vol. 40 Nos 3/4,
pp. 203-225, doi: 10.1080/02642069.2019.1672666.
Bhat, M.A. and Khan, S.T. (2023), “Determinants of accounting students’ decision to pursue career as
ACCA-certified accountants: a case study of Omani students”, Management and Sustainability:
An Arab Review, doi: 10.1108/MSAR-09-2022-0043.
Bhat, M.A., Khan, S.T. and Rainayee, R.A. (2021), “Assessment of perceived labor market conditions in
employees’ turnover intention model–mediation and moderation analyzes”, PSU Research
Review, (ahead-of-print).
Bhat, M.A. and Tariq, S. (2022), Impact of Job Burnout on Performance: A Study among Hospital
Employees of J&K, India, BIMTECH Business Perspectives, pp. 1-17.
Bhat, M.A., Gomero, G.D. and Khan, S.T. (2022), “Antecedents of savings behaviour among rural
households: a holistic approach”, FIIB Business Review, doi: 10.1177/23197145221110281.
ITSE Çelik, H. (2008), “What determines Turkish customers’ acceptance of internet banking?”, International
Journal of Bank Marketing, Vol. 26 No. 5, pp. 353-370, doi: 10.1108/02652320810894406.
Celik, I. (2023), “Towards Intelligent-TPACK: an empirical study on teachers’ professional knowledge
to ethically integrate artificial intelligence (AI)-based tools into education”, Computers in Human
Behavior, Vol. 138, p. 107468, doi: 10.1016/j.chb.2022.107468.
Chang, C.J., Hsu, B.C.Y. and Chen, M.Y. (2022), “Viewing sports online during the COVID-19 pandemic:
the antecedent effects of social presence on the technology acceptance model”, Sustainability,
Vol. 14 No. 1, p. 341, doi: 10.3390/su14010341.
Chang, Y.-W., Hsu, P.-Y., Chen, J., Shiau, W.-L. and Xu, N. (2023), “Utilitarian and/or hedonic shopping –
consumer motivation to purchase in smart stores”, Industrial Management and Data Systems,
Vol. 123 No. 3, pp. 821-842, doi: 10.1108/IMDS-04-2022-0250.
ChatGPT (2022), “ChatGPT: optimizing language models for dialogue”, OpenAI, available at: https://
openai.com/blog/chatgpt/ (accessed 31 January 2023).
Chattaraman, V., Kwon, W.S., Gilbert, J.E. and Ross, K. (2019), “Should AI-based, conversational digital
assistants employ social-or task-oriented interaction style? A task-competency and reciprocity
perspective for older adults”, Computers in Human Behavior, Vol. 90, pp. 315-330, doi: 10.1016/j.
chb.2018.08.048.
Chi, O.H., Jia, S., Li, Y. and Gursoy, D. (2021), “Developing a formative scale to measure consumers’
trust toward interaction with artificially intelligent (AI) social robots in service delivery”,
Computers in Human Behavior, Vol. 118, p. 106700, doi: 10.1016/j.chb.2021.106700.
Chin, W.W., Peterson, R.A. and Brown, S.P. (2008), “Structural equation modeling in marketing: some
practical reminders”, Journal of Marketing Theory and Practice, Vol. 16 No. 4, pp. 287-298,
doi: 10.2753/MTP1069-6679160402.
Chocarro, R., Cortiñas, M. and Marcos-Matas, G. (2021), “Teachers’ attitudes towards chatbots in
education: a technology acceptance model approach considering the effect of social language,
bot proactiveness, and users’ characteristics”, Educational Studies, Vol. 49 No. 2, pp. 1-19,
doi: 10.1080/03055698.2020.1850426.
Chong, A.Y.L., Chan, F.T., Goh, M. and Tiwari, M.K. (2013), “Do interorganisational relationships
and knowledge-management practices enhance collaborative commerce adoption?”,
International Journal of Production Research, Vol. 51 No. 7, pp. 2006-2018, doi: 10.1080/
00207543.2012.701776.
Clayson, P.E. (2020), “Moderators of the internal consistency of error-related negativity scores: a meta-analysis
of internal consistency estimates”, Psychophysiology, Vol. 57 No. 8, p. e13583, doi: 10.1111/psyp.13583.
Cooper, G. (2023), “Examining science education in ChatGPT: an exploratory study of generative
artificial intelligence”, Journal of Science Education and Technology, Vol. 32 No. 3, doi: 10.1007/
s10956-023-10039-y.
Cotton, D.R., Cotton, P.A. and Shipway, J.R. (2023), “Chatting and cheating: ensuring academic integrity
in the era of ChatGPT”, Innovations in Education and Teaching International, pp. 1-12,
doi: 10.1080/14703297.2023.2190148.
Davis, F.D. (1985), “A technology acceptance model for empirically testing new end-user information
systems: theory and results”, Doctoral dissertation, Massachusetts Institute of Technology.
Davis, F.D. (1989), “Perceived usefulness, perceived ease of use, and user acceptance of information
technology”, MIS Quarterly, Vol. 13 No. 3, pp. 319-340, doi: 10.2307/249008.
Davis, F.D. (1993), “User acceptance of information technology: system characteristics, user perceptions
and behavioral impacts”, International Journal of Man-Machine Studies, Vol. 38 No. 3,
pp. 475-487, doi: 10.1006/imms.1993.1022.
Davis, F.D., Bagozzi, R.P. and Warshaw, P.R. (1989), “User acceptance of computer technology: a
comparison of two theoretical models”, Management Science, Vol. 35 No. 8, pp. 982-1003,
doi: 10.1287/mnsc.35.8.982.
de Kervenoael, R., Hasan, R., Schwob, A. and Goh, E. (2020), “Leveraging human-robot interaction in Adoption and
hospitality services: incorporating the role of perceived value, empathy, and information sharing
into visitors’ intentions to use social robots”, Tourism Management, Vol. 78, p. 104042,
usage of
doi: 10.1016/j.tourman.2019.104042. ChatGPT
De Luna, I.R., Liebana-Cabanillas, F., Sanchez-Fernandez, J. and Muñoz-Leiva, F. (2019), “Mobile payment is
not all the same: the adoption of mobile payment systems depending on the technology applied”,
Technological Forecasting and Social Change, Vol. 146, pp. 931-944, doi: 10.1016/j.techfore.2018.09.018.
Dwivedi, Y.K., Kshetri, N., Hughes, L., Slade, E.L., Jeyaraj, A., Kar, A.K., Baabdullah, A.M.,
Koohang, A., Raghavan, V., Ahuja, M., Albanna, H., Albashrawi, M.A., Al-Busaidi, A.S.,
Balakrishnan, J., Barlette, Y., Basu, S., Bose, I., Brooks, L., Buhalis, D., Carter, L., Chowdhury,
S., Crick, T., Cunningham, S.W., Davies, G.H., Davison, R.M., De, R., Dennehy, D., Duan, Y.,
Dubey, R., Dwivedi, R., Edwards, J.S., Flavian, C., Gauld, R., Grover, V., Hu, M.-C., Janssen,
M., Jones, P., Junglas, I., Khorana, S., Kraus, S., Larsen, K.R., Latreille, P., Laumer, S., Malik, F.
T., Mardani, A., Mariani, M., Mithas, S., Mogaji, E., Nord, J.H., O’Connor, S., Okumus, F.,
Pagani, M., Pandey, N., Papagiannidis, S., Pappas, I.O., Pathak, N., Pries-Heje, J., Raman, R.,
Rana, N.P., Rehm, S.-V., Ribeiro-Navarrete, S., Richter, A., Rowe, F., Sarker, S., Stahl, B.C.,
Tiwari, M.K., van der Aalst, W., Venkatesh, V., Viglia, G., Wade, M., Walton, P., Wirtz, J. and
Wright, R. (2023), “Opinion paper: ‘So what if ChatGPT wrote it?’ Multidisciplinary perspectives on
opportunities, challenges and implications of generative conversational AI for research, practice
and policy”, International Journal of Information Management, Vol. 71, p. 10264, doi: 10.1016/j.
ijinfomgt.2023.102642.
Elbanna, S. and Armstrong, L. (2023), “Exploring the integration of ChatGPT in education:
adapting for the future”, Management and Sustainability: An Arab Review, doi: 10.1108/
MSAR-03-2023-0016.
Fernandes, T. and Oliveira, E. (2021), “Understanding consumers’ acceptance of automated
technologies in service encounters: drivers of digital voice assistants adoption”, Journal of
Business Research, Vol. 122, pp. 180-191, doi: 10.1016/j.jbusres.2020.08.058.
Fornell, C. and Larcker, D.F. (1981), “Structural equation models with unobservable variables and
measurement error: algebra and statistics”, Journal of Marketing Research, Vol. 18 No. 3,
doi: 10.1177/002224378101800313.
Franke, G. and Sarstedt, M. (2019), “Heuristics versus statistics in discriminant validity testing: a comparison
of four procedures”, Internet Research, Vol. 29 No. 3, pp. 430-447, doi: 10.1108/IntR-12-2017-0515.
Gefen, D. (2000), “E-commerce: the role of familiarity and trust”, Omega, Vol. 28 No. 6, pp. 725-737,
doi: 10.1016/S0305-0483(00)00021-9.
Gefen, D. and Straub, D.W. (2004), “Consumer trust in B2C e-Commerce and the importance of social
presence: experiments in e-Products and e-services”, Omega, Vol. 32 No. 6, pp. 407-424, doi: 10.1016/j.
omega.2004.01.006.
Giovanis, A.N., Binioris, S. and Polychronopoulos, G. (2012), “An extension of TAM model with IDT
and security/privacy risk in the adoption of internet banking services in Greece”, EuroMed
Journal of Business, Vol. 7 No. 1, pp. 24-53, doi: 10.1108/14502191211225365.
Global financial index database (2017), available online: https://microdata.worldbank.org/index.php/
catalog/3324
Gupta, A. and Arora, N. (2017), “Consumer adoption of m-banking: a behavioral reasoning theory perspective”,
International Journal of Bank Marketing, Vol. 35 No. 4, pp. 733-747, doi: 10.1108/IJBM-11-2016-0162.
Gong, L. and Nass, C. (2007), “When a talking-face computer agent is half-human and half-humanoid:
human identity and consistency preference”, Human Communication Research, Vol. 33 No. 2,
pp. 163-193, doi: 10.1111/j.1468-2958.2007.00295.x.
Haenlein, M., Kaplan, A., Tan, C.W. and Zhang, P. (2019), “Artificial intelligence (AI) and
management analytics”, Journal of Management Analytics, Vol. 6 No. 4, pp. 341-343, doi:
10.1080/23270012.2019.1699876.
ITSE Hair, J.F., Jr, Howard, M.C. and Nitzl, C. (2020), “Assessing measurement model quality in PLS-SEM
using confirmatory composite analysis”, Journal of Business Research, Vol. 109, pp. 101-110,
doi: 10.1016/j.jbusres.2019.11.069.
Hamid, A.A., Razak, F.Z.A., Bakar, A.A. and Abdullah, W.S.W. (2016), “The effects of perceived
usefulness and perceived ease of use on continuance intention to use e-government”, Procedia
Economics and Finance, Vol. 35, pp. 644-649, doi: 10.1016/S2212-5671(16)00079-4.
Han, N., Chen, J., Xiao, G., Zhang, H., Zeng, Y. and Chen, H. (2021), “Fine-grained cross-modal alignment
network for text-video retrieval”, Proceedings of the 29th ACM International Conference on
Multimedia, pp. 3826-3834, 10.1145/3474085.3475241
Hariri, W. (2023), “Unlocking the potential of ChatGPT: a comprehensive exploration of its applications,
advantages”, Limitations, and Future Directions in Natural Language Processing, arXiv preprint
arXiv:2304.02017.
Henseler, J., Hubona, G. and Ray, P.A. (2016), “Using PLS path modeling in new technology research:
updated guidelines”, Industrial Management and Data Systems, Vol. 116 No. 1, pp. 2-20,
doi: 10.1108/IMDS-09-2015-0382.
Henseler, J., Ringle, C.M. and Sarstedt, M. (2015), “A new criterion for assessing discriminant validity in
variance-based structural equation modeling”, Journal of the Academy of Marketing Science,
Vol. 43 No. 1, pp. 115-135, doi: 10.1007/s11747-014-0403-8.
Hoffmann, A.O.I. and Birnbrich, C. (2012), “The impact of fraud prevention on bank-customer
relationships: an empirical investigation in retail banking”, International Journal of Bank
Marketing, Vol. 30 No. 5, pp. 390-407, doi: 10.1108/02652321211247435.
Howcroft, B., Hamilton, R. and Hewer, P. (2002), “Consumer attitude and the usage and adoption of
home-based banking in the United Kingdom”, International Journal of Bank Marketing, Vol. 20
No. 3, pp. 111-121, doi: 10.1108/02652320210424205.
Jahanshahi, H., Kazmi, S. and Cevik, M. (2022), “Auto response generation in online medical chat
services”, Journal of Healthcare Informatics Research, Vol. 6 No. 3, pp. 344-374, doi: 10.1007/
s41666-022-00118-x.
Kabra, G., Ghosh, V. and Joshi, Y. (2023), “Factors influencing adoption of cloud computing services in
HEIs: a UTAUT approach based on students’ perception”, International Journal of Business
Information Systems, Vol. 42 No. 1, pp. 103-122, doi: 10.1504/IJBIS.2023.128301.
Keiper, M.C., Fried, G., Lupinek, J. and Nordstrom, H. (2023), “Artificial intelligence in sport
management education: playing the AI game with ChatGPT”, Journal of Hospitality, Leisure,
Sport and Tourism Education, Vol. 33, p. 100456, doi: 10.1016/j.jhlste.2023.100456.
Khan, S.T., Bhat, M.A. and Sangmi, M.U.D. (2020), “Impact of microfinance on economic, social, political
and psychological empowerment: evidence from women’s self-help groups in Kashmir valley,
India”, FIIB Business Review, Vol. 12 No. 1, p. 2319714520972905, doi: 10.1177/2319714520972905.
Kim, J., Merrill, K., Jr, and Collins, C. (2023), “Investigating the importance of social presence on
intentions to adopt an AI romantic partner”, Communication Research Reports, Vol. 40 No. 1,
pp. 1-10, doi: 10.1080/08824096.2022.2159800.
Kim, J., Merrill, K., Jr, Xu, K. and Kelly, S. (2022), “Perceived credibility of an AI instructor in online
education: the role of social presence and voice features”, Computers in Human Behavior,
Vol. 136, p. 107383, doi: 10.1016/j.chb.2022.107383.
King, M.R, chatGPT (2023), “A conversation on artificial intelligence, chatbots, and plagiarism in
higher education”, Cellular and Molecular Bioengineering, Vol. 16 No. 1, pp. 1-2, doi: 10.1007/
s12195-022-00754-8.
Kushwaha, A.K. and Kar, A.K. (2021), “MarkBot – a language model-driven chatbot for interactive
marketing in post-modern world”, Information Systems Frontiers, doi: 10.1007/s10796-021-10184-y.
La Barbera, F. and Ajzen, I. (2020), “Control interactions in the theory of planned behavior: rethinking
the role of subjective norm”, Europe’s Journal of Psychology, Vol. 16 No. 3, p. 401.
Larue, G.S. and Watling, C.N. (2021), “Acceptance of visual and audio interventions for distracted Adoption and
pedestrians”, Transportation Research Part F: Traffic Psychology and Behaviour, Vol. 76,
pp. 369-383.
usage of
Lee, H. (2023), “The rise of ChatGPT: exploring its potential in medical education”, Anatomical Sciences
ChatGPT
Education, pp. 1-6, doi: 10.1002/ase.2270.
Liang, C.C. and Shiau, W.L. (2018), “Moderating effect of privacy concerns and subjective norms between
satisfaction and repurchase of airline e-ticket through airline-ticket vendors”, Asia Pacific Journal of
Tourism Research, Vol. 23 No. 12, pp. 1142-1159, doi: 10.1080/10941665.2018.1528290.
Liengaard, B.D., Sharma, P.N., Hult, G.T.M., Jensen, M.B., Sarstedt, M., Hair, J.F. and Ringle, C.M.
(2021), “Prediction: coveted, yet forsaken? Introducing a cross-validated predictive ability test in
partial least squares path modeling”, Decision Sciences, Vol. 52 No. 2, pp. 362-392, doi: 10.1111/
deci.12445.
Liebana-Cabanillas, F., Marinkovic, V., De Luna, I.R. and Kalinic, Z. (2018), “Predicting the determinants of
mobile payment acceptance: a hybrid SEM-neural network approach”, Technological Forecasting
and Social Change, Vol. 129, pp. 117-130, doi: 10.1016/j.techfore.2017.12.015.
Lim, J.S. and Zhang, J. (2022), “Adoption of AI-driven personalization in digital news platforms: an
integrative model of technology acceptance and perceived contingency”, Technology in Society,
Vol. 69, doi: 10.1016/j.techsoc.2022.101965. Article 101965.
Lin, C.H., Shih, H.Y. and Sher, P.J. (2007), “Integrating technology readiness into technology acceptance:
the TRAM model”, Psychology and Marketing, Vol. 24 No. 7, doi: 10.1002/mar.20177.
Lu, L., Cai, R. and Gursoy, D. (2019), “Developing and validating a service robot integration willingness scale”,
International Journal of Hospitality Management, Vol. 80, pp. 36-51, doi: 10.1016/j.ijhm.2019.01.005.
Min, S., So, K.K.F. and Jeong, M. (2021), “Consumer adoption of the uber mobile application: Insights
from diffusion of innovation theory and technology acceptance model”, In Future of Tourism
Marketing, Routledge, pp. 2-15.
Mishra, A., Shukla, A. and Sharma, S.K. (2022), “Psychological determinants of users’ adoption and
word-of-mouth recommendations of smart voice assistants”, International Journal of
Information Management, Vol. 67, p. 102413, doi: 10.1016/j.ijinfomgt.2021.102413.
Montano, D.E. and Kasprzyk, D. (2015), “Theory of reasoned action, theory of planned behavior, and
the integrated behavioral model”, Health Behavior: Theory, Research and Practice, Vol. 70 No. 4,
p. 231.
Moreno, F.M., Lafuente, J.G., Carreon, F.Á. and Moreno, S.M. (2017), “The characterization of the
millennials and their buying behavior”, International Journal of Marketing Studies, Vol. 9 No. 5,
pp. 135-144, doi: 10.5539/ijms.v9n5p135.
Mukherjee, A. and Nath, P. (2003), “A model of trust in online relationship banking”, International
Journal of Bank Marketing, Vol. 21 No. 1, pp. 5-15, doi: 10.1108/02652320310457767.
Nguyen, A., Ngo, H.N., Hong, Y., Dang, B. and Nguyen, B.P.T. (2023), “Ethical principles for
artificial intelligence in education”, Education and Information Technologies, Vol. 28 No. 4,
pp. 4221-4241.
Ooge, J., Stiglic, G. and Verbert, K. (2022), “Explaining artificial intelligence with visual analytics in
healthcare”, Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, Vol. 12
No. 1, p. e1427, doi: 10.1002/widm.1427.
OpenAI (2022), available at: https://openai.com/blog/chatgpt/ (accessed 31 January 2023).
Oravec, J.A. (2019), “Artificial intelligence, automation, and social welfare: some ethical and historical
perspectives on technological overstatement and hyperbole”, Ethics and Social Welfare, Vol. 13
No. 1, pp. 18-32, doi: 10.1080/17496535.2018.1512142.
Partel, V., Kakarla, S.C. and Ampatzidis, Y. (2019), “Development and evaluation of a low-cost and
smart technology for precision weed management utilizing artificial intelligence”, Computers
and Electronics in Agriculture, Vol. 157, pp. 339-350, doi: 10.1016/j.compag.2018.12.048.
ITSE Peng, D.X. and Lai, F. (2012), “Using partial least squares in operations management research: a
practical guideline and summary of past research”, Journal of Operations Management, Vol. 30
No. 6, pp. 467-480, doi: 10.1016/j.jom.2012.06.002.
Perkins, M. (2023), “Academic integrity considerations of AI large language models in the post-
pandemic era: ChatGPT and beyond”, Journal of University Teaching and Learning Practice,
Vol. 20 No. 2, doi: 10.53761/1.20.02.07.
Pillai, R., Sivathanu, B., Metri, B. and Kaushik, N. (2023), “Students’ adoption of AI-based teacher-bots
(T-bots) for learning in higher education”, Information Technology and People, doi: 10.1108/ITP-
02-2021-0152.
Rahman, M.M., Strawderman, L., Lesch, M.F., Horrey, W.J., Babski-Reeves, K. and Garrison, T. (2018),
“Modelling driver acceptance of driver support systems”, Accident Analysis and Prevention,
Vol. 121, pp. 134-147, doi: 10.1016/j.aap.2018.08.028.
Sahoo, D., Harichandan, S., Kar, S.K. and Sreejesh, S. (2022), “An empirical study on consumer motives
and attitude towards adoption of electric vehicles in India: policy implications for stakeholders”,
Energy Policy, Vol. 165, p. 112941, doi: 10.1016/j.enpol.2022.112941.
Sankey, M.D. and Marshall, S.J. (2023), “Perspective chapter: the learning management system of 2028 and
how we start planning for this now”, Higher Education-Reflections from the Field, IntechOpen.
Sarstedt, M. and Cheah, J.H. (2019), “Partial least squares structural equation modeling using
SmartPLS: a software review”, Journal of Marketing Analytics, Vol. 7 No. 3, pp. 196-202, doi:
10.1057/s41270-019-00058-3.
Sehabuddin, A. and Oktarina, N. (2022), “E-Learning adoption; how is students behavior during the covid-
19 pandemic?”, Dinamika Pendidikan, Vol. 17 No. 1, pp. 48-61, doi: 10.15294/dp.v17i1.34367.
Shaked, D., Faulkner, L.M., Tolle, K., Wendell, C.R., Waldstein, S.R. and Spencer, R.J. (2020), “Reliability
and validity of the Conners’ continuous performance test”, Applied Neuropsychology: Adult,
Vol. 27 No. 5, pp. 478-487, doi: 10.1080/23279095.2019.1570199.
Sharma, G. and Thakur, A. (2023), ChatGPT in Drug Discovery, doi: 10.26434/chemrxiv-2023-qgs3k.
Shen, S., Xu, K., Sotiriadis, M. and Wang, Y. (2022), “Exploring the factors influencing the adoption and
usage of augmented reality and virtual reality applications in tourism education within the
context of COVID-19 pandemic”, Journal of Hospitality, Leisure, Sport and Tourism Education,
Vol. 30, p. 100373.DOI:, doi: 10.1016/j.jhlste.2022.100373.
Shrestha, N. (2021), “Factor analysis as a tool for survey analysis”, American Journal of Applied
Mathematics and Statistics, Vol. 9 No. 1, pp. 4-11, doi: 10.12691/ajams-9-1-2.
Shroff, R.H., Deneen, C.C. and Ng, E.M. (2011), “Analysis of the technology acceptance model in
examining students’ behavioural intention to use an e-portfolio system”, Australasian Journal of
Educational Technology, Vol. 27 No. 4, doi: 10.14742/ajet.940.
Siddiqui, K. (2013), “Heuristics for sample size determination in multivariate statistical
techniques”, World Applied Sciences Journal, Vol. 27 No. 2, pp. 285-287.
Singh, N. and Sinha, N. (2020), “How perceived trust mediates merchant’s intention to use a mobile
wallet technology”, Journal of Retailing and Consumer Services, Vol. 52, p. 101894, doi: 10.1016/j.
jretconser.2019.101894.
Song, H., Ruan, W.J. and Jeon, Y.J.J. (2021), “An integrated approach to the purchase decision making
process of food-delivery apps: focusing on the TAM and AIDA models”, International Journal of
Hospitality Management, Vol. 95, p. 102943, doi: 10.1016/j.ijhm.2021.102943.
Suh, B. and Han, I. (2002), “Effect of trust on customer acceptance of internet banking”, Electronic
Commerce Research and Applications, Vol. 1 Nos 3/4, pp. 247-263, doi: 10.1016/S1567-4223(02)
00017-0.
Szymkowiak, A., Melovic, B., Dabic, M., Jeganathan, K. and Kundi, G.S. (2021), “Information technology and
gen Z: the role of teachers, the internet, and technology in the education of young people”, Technology
in Society, Vol. 65, p. 101565, doi: 10.1016/j.techsoc.2021.101565.
Tenenhaus, M., Vinzi, V.E., Chatelin, Y.M. and Lauro, C. (2005), “PLS path modeling”, Computational Adoption and
Statistics and Data Analysis, Vol. 48 No. 1, pp. 159-205, doi: 10.1016/j.csda.2004.03.005.
usage of
Tlili, A., Shehata, B., Adarkwah, M.A., et al. (2023), “What if the devil is my guardian angel: ChatGPT
as a case study of using chatbots in education”, Smart Learning Environments, Vol. 10 No. 1, ChatGPT
p. 15, doi: 10.1186/s40561-023-00237-x.
Van Pinxteren, M.M.E., Wetzels, R.W.H., Rüger, J., Pluymaekers, M. and Wetzels, M. (2019), “Trust in
humanoid robots: implications for services marketing”, Journal of Services Marketing, Vol. 33
No. 4, pp. 507-518, doi: 10.1108/JSM-01-2018-0045.
Venkatesh, V., Thong, J.Y. and Xu, X. (2012), “Consumer acceptance and use of information technology:
extending the unified theory of acceptance and use of technology”, MIS Quarterly, pp. 157-178,
doi: 10.2307/41410412.
Vrbka, J. and Rowland, Z. (2020), “Using artificial intelligence in company management”, Digital Age:
Chances, Challenges and Future 7, Springer International Publishing, pp. 422-429, doi: 10.1007/
978-3-030-27015-5_51.
Wang, Y. and Chen, H. (2019), “The influence of dialogic engagement and prominence on visual product
placement in virtual reality videos”, Journal of Business Research, Vol. 100, pp. 493-502,
doi: 10.1016/j.jbusres.2019.01.018.
Warsono, H., Yuwono, T. and Putranti, I. (2023), “Analyzing technology acceptance model for
collaborative governance in public administration: empirical evidence of digital governance and
perceived ease of use”, International Journal of Data and Network Science, Vol. 7 No. 1, pp. 41-48,
doi: 10.5267/j.ijdns.2022.12.008.
Wojciechowski, R. and Cellary, W. (2013), “Evaluation of learners’ attitude toward learning in ARIES
augmented reality environments”, Computers and Education, Vol. 68, pp. 570-585, doi: 10.1016/j.
compedu.2013.02.014.
Yousafzai, S., Pallister, J. and Foxall, G. (2009), “Multi-dimensional role of trust in internet banking adoption”,
The Service Industries Journal, Vol. 29 No. 5, pp. 591-605, doi: 10.1080/02642060902719958.
Zawacki-Richter, O., Marín, V.I., Bond, M., et al. (2019), “Systematic review of research on artificial
intelligence applications in higher education – where are the educators?”, International Journal
of Educational Technology in Higher Education, Vol. 16 No. 1, p. 39, doi: 10.1186/s41239-019-
0171-0.
Zhang, K. and Aslan, A.B. (2021), “AI technologies for education: recent research and future directions”,
Computers and Education: Artificial Intelligence, Vol. 2, p. 100025, doi: 10.1016/j.caeai.2021.100025.

Corresponding author
Chandan Kumar Tiwari can be contacted at: [email protected]

For instructions on how to order reprints of this article, please visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm
Or contact us for further details: [email protected]