Annals of Tourism Research 101 (2023) 103578

Contents lists available at ScienceDirect

Annals of Tourism Research

COMMENTARY

AI, ChatGPT and the university

a r t i c l e i n f o

Article history:
Received 4 April 2023
Received in revised form 15 May 2023
Accepted 17 May 2023
Available online 3 June 2023

Associate editor: Scott McCabe

Introduction

The university has always had a longstanding love–hate relationship with technology. Indeed, universities are often seen as
leaders of technology where the future is ‘created’ through research and design, and then application by industry. As noted in
a recent essay in the Annals of Tourism Research by Nautiyal, Albrecht, and Nautiyal (2023), it is easy to identify many such
technologies that were either developed within the university and moved to society or were developed first by industry but
then translated to support the university. Certainly, an appropriate example is the Internet which was initially developed within
universities for the military and is now an embedded component of its ecosystem; another important technology employed
throughout the university is the computer (and software) itself, whether it is in the form of calculators, mobile phones, laptops,
or supercomputers. However, there are many other technologies whether they are developed within engineering and computer
sciences, medical sciences, and/or the social and physical sciences. With this said, the university has long struggled with the
integration of these technologies within classroom settings. And, indeed, the current discussion of the implications of AI-related
technologies such as ChatGPT highlights the failure of the university to recognize and then adapt them to effectively update
current offerings (Lund et al., 2023).
This article expands the discussion of Nautiyal et al. (2023) by focusing on important AI-related issues facing the university.
Please note that in this essay we use ‘the university’ to refer to processes supporting education, research, and the many applied
relationships between education and society. We also recognize that students often take classes that are directly related to and/or
indirectly exposed to many forms of these technologies. Last, we argue that many of these technologies comprise some form of
‘artificial intelligence’ (AI) which affords us (at the university) the opportunity to develop new ways to improve capacities within
our classrooms and our research.
This complicated relationship between technology and education has been long and deep. For example, it is widely believed
that while the advent of the alphabet contributed to a new age of human evolution, it posed a substantial threat to traditional
education methods. In a recent article in the NY Times titled “What would Plato say about ChatGPT?”, the author argued that
Plato feared the impact of the written word (the new technology) because he believed that it would lead to people without
memory (Tufekci, 2022). More recently (over 50 years ago) and more personally, one author of this essay took biometry classes
as an undergraduate student where he used mechanical calculators to compute the required statistic(s); the professor, like Plato,

⁎ Corresponding author.
URL's:URL: [email protected] (D.R. Fesenmaier), [email protected] (K. Wöber).

https://doi.org/10.1016/j.annals.2023.103578
0160-7383/© 2023 Elsevier Ltd. All rights reserved.
D.R. Fesenmaier and K. Wöber Annals of Tourism Research 101 (2023) 103578

repeatedly stated that we needed to learn all requirements for computation to understand the implications of the particular sta-
tistic. Fast forward 20+ years, students in similar university-based statistics classes were told that the new computing systems
such as Minitab, SAS, SYSTAT, or SPSS would result in much ‘poorer’ students as they would not have the knowledge to do
their calculations. Now, students have to learn to use even more powerful tools (e.g., Qualtrics, R, Python, Snowflake, Tableau,
etc.) to create, calculate and visualize statistics. While the impact of this change is certainly debatable, we argue that this new
technology offers the opportunity to design systems that empower students to learn to ask and answer even more complicated
(and potentially more valuable) questions.

Roles that AI can play

Today's technology including AI-based systems referred to as ‘generative AI’ has been received with many warnings, and in-
deed, many universities worldwide are banning or at least limiting their use. While AI has been around for many years, it appears
that we now have reached the ‘tipping point’ wherein researchers have realized the requirements for building truly ‘intelligent’
systems. ChatGPT is one AI system that has attracted a lot of attention; however, many others exist today, and others are expected
to be made public within the next couple of years. These new systems use deep learning methods and large computational plat-
forms to recognize direct and indirect associations among and between concepts with a focus on ‘generating’ new material,
whether they are text-based (such as essays, poems, and computer programs) or graphic-based (such as art or animated
humans). Importantly, it is believed that these systems can evolve (i.e., learn) independently, and as such, have the ability to
‘think’ creatively.
A brief ‘conversation’ with ChatGPT is presented in Appendix A to highlight one aspect of this technology. In this conversation,
we asked a series of questions regarding theories and research related to tourism experiences with the assumption that ChatGPT
“acts as a university professor trained in theory and methods”. ChatGPT identified five theories it ‘believes’ provide a foundation
for understanding tourism experiences. Note that four of the five theories are reasonably well recognized (i.e., experience econ-
omy, authenticity, service-dominant logic, and flow), but it suggested that Emotional Contagion Theory is a fifth theory that is
not widely recognized in the tourism literature. A follow-up series of questions about this theory shows that the system is
‘thoughtful’ in the sense that somehow ‘understands’ the linkages between tourism-related experiences, emotions, and the social
aspects of travel, and the system offers some very useful references describing research related to tourism. We then questioned
ChatGPT about the other potential theories related to tourism experiences. In particular, we asked directly about Prospect Theory
which is useful in understanding how travelers deal with uncertainty in decision-making. ChatGPT provided general background
information about the theory and then some insightful discussion related to the previous responses related to emotion and
tourism experience. Slight prodding of ChatGPT provided additional theories and recommended readings. For example, the last
question in our ‘interview’ asked how Event Theory by Zachs might fit within this discussion. ChatGPT indicated that Event Theory
can be useful in understanding the various stages of the tourism experience within a trip. However, an important ‘issue’ with
ChatGPT is that it appears to create fake or inaccurate findings which seem plausible; this fake information is called a ‘hallucina-
tion’ within the AI literature. With this in mind, we conducted an analysis of the five references related to Contagion Theory and
tourism research. We found two of the citations actually exist and are somewhat connected to the issues at hand; three other
papers were not found either because they do not exist or they exist somewhere within the ether of the internet.
Of course, one could continue exploring much about tourism and tourism experience, but one might expect that the domain-
specific knowledge required to provide this level of analysis (and recommendation) is reasonably limited; however, having
limited knowledge or making mistakes is a very human trait and does not help to distinguish between AI and human-
generated text. A recent article suggests that it's only a matter of time before these new tools will have reduced their error
rate and improved the quality of their responses (Metz & Collins, 2023).
The implications of these new technologies are profound in much the same way as the alphabet, the printing press, electricity,
and the internet have been. As with these technologies, AI-enabled systems will require us to improve our ability to access all
relevant information and to improve our ability to think more creatively; others believe that AI will also challenge us to recognize
what it means to be human. Reports from university-based discussions have documented the immediate impact of ChatGPT on
students' ability to develop quite detailed literature reviews and pass tests offered by most universities. These reports also
document the responses of professors and students on how these capacities represent opportunities and threats to education.
As professors, we can imagine the challenges posed by ChatGPT wherein it now ‘flips’ the relationships between student, teacher,
and knowledge base to focus more on dyadic activities supporting critical thinking, creativity, and real-world application. Also, it
appears that ChatGPT (and other such systems) can be used by students and industry professionals to gain access to information
that can provide much deeper insight into the questions at hand. Thus, the focus of any exercise, project, or activity is NOT on
simply gaining relevant literature but rather figuring out the correct questions to ask (and in the best order), and further, iden-
tifying the essential elements which address the specific goals of the activity. Indeed, one of the most important affordances of
ChatGPT (for example) is that it offers the opportunity to explore new perspectives and, in turn, to think more creatively
about the issues underlying the respective queries. As such, one can easily imagine how AI-based tools can be used to assist in
university-related activities to redefine the learning process and how the results of queries are shared with others.
However, an important weakness of AI-based systems is their ability to provide incorrect information, disinformation, and/or
misinformation (i.e., hallucinations) in a way that makes it seem ‘real.’ Another significant weakness is that it always remains un-
clear whether all available information has been considered and processed in the best possible way to produce a result. Similar to
how many people today take the search results from Google as a given, with AI-based systems there is an even greater risk that

2
D.R. Fesenmaier and K. Wöber Annals of Tourism Research 101 (2023) 103578

results will not be questioned further because there are no easy ways to falsify them. Thus, the results of any query must be sub-
jected to substantive critical assessment; that is, similar to any study, the researcher is responsible for making certain the infor-
mation provided by the AI-based system is reliable, verifiable, and thorough. It is hoped that unlike in the search engine world
today, we will see a more pluralistic environment with different AI-based systems providing competing results, spurring the pro-
cess of critical evaluation.

Future issues

With this in mind, we need to ask several basic questions from the lens of the university–student–professor: What is
the ‘real’ contribution by the student? What was learned in this exercise? How did the student add critically to the dis-
cussion at hand? Additionally, university professors are asking questions regarding the ethics of using such systems for
homework assignments or exams, and by extension, they are questioning the ethics of use in a literature review of pub-
lished material. While these concerns are understandable, we argue that they are NOT the right questions as they essen-
tially seek to limit the use of such systems rather than encourage their use to create new knowledge. Thus, we argue that
the central question for university professors should focus on creativity, critical thinking, and the extent to which it con-
tributes to progress in our understanding of essential ideas as well as how methods of assessment need to be adapted to
the new environment. Thinking about the state of the university (including teaching and research) 30 years ago, we ar-
gued that the use of online search tools such as Yahoo!, AOL, and Netscape is somehow cheating, or at the very least, cre-
ates the possibility for obtaining inaccurate information and revealing that the student is somehow ‘lazy.’ However, these
changes do not necessarily have to result in a new or more intensive form of digital education, as some might assume, but
can also have the exact opposite effect. More face-to-face teaching, personal interaction, the teaching of values, the re-
naissance of teaching in-depth methodological skills, and oral examinations could be the universities' responses to cur-
rent developments. Would this not be a development to be welcomed?
Beyond the current hype supporting the development of AI and, indeed, others taking dystopian views whereby AI-based systems
will lead to human decay, it seems clear that AI-enabled tools will accelerate changes in society including altering how we learn, how
we ask and answer questions (i.e., do research) and how we communicate. In short, it has the potential to effect a radical change in
university-based education (Mollick & Mollick, 2023). Further, emerging AI-related technologies will help to develop more efficient
systems and to explore new business processes. Examples include the use of AI in medicine to read X-rays, guide many forms of trans-
portation systems, make predictions of highly complex economic trends, and support the development of mobile marketing tools.
However, there are many important questions that we need to ask as we consider the management and future development of AI;
they include: Should we trust systems that claim to detect if a document was written by a content generator? Should we expect AI
providers to demonstrate some moral responsibility to provide features that allow users to check if a document was produced by
them? Should AI vendors be forced to disclose their methods, algorithms, and data sources for how they generated a particular an-
swer? While there are, and will continue to be, many questions about this new technology, we believe that an important first step
is that these AI-related technologies should be made available for free; further, we believe that all publishers of scientific journals
should provide unrestricted access to these systems. Additionally, we believe that systems for governance should be created to
guide the future of its integration within the university and society. And finally, while these systems will inherently pose ‘challenges’
to education, it is essential that the education community continually experiment with this technology to better understand how to
use it to improve education. With these initial steps, we believe that AI-based systems such as ChatGPT offer the potential to greatly
improve education and research, and therefore, the way society views the world.

Declaration of Generative AI and AI-assisted technologies in the writing process

None.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared
to influence the work reported in this paper.

Appendix A. A conversation with ChatGPT

Question
I want you to act as a university professor trained in theory and methodology. I will ask you to discuss an area of study that is
supported by many theories. You will identify and explain five key theories that enable us to understand this particular area of
study. The area of study is tourism experience.
ChatGPT
As a professor of tourism studies, I would be happy to discuss five key theories that enable us to understand the area of tour-
ism experience.

3
D.R. Fesenmaier and K. Wöber Annals of Tourism Research 101 (2023) 103578

1. Experience economy theory: This theory, developed by Pine and Gilmore, suggests that economic value is increasingly being
derived from experiences rather than from products or services. In the context of tourism, this theory suggests that tourists
are seeking unique and memorable experiences, and that these experiences can be created and marketed by businesses in
the tourism industry. This theory has led to the development of a range of new tourism products and services, such as adven-
ture tourism, ecotourism, and cultural tourism, which focus on creating immersive and memorable experiences for tourists.
2. Authenticity theory: This theory, developed by MacCannell, suggests that tourists are seeking authentic experiences that are
genuine and not contrived. In the context of tourism, this theory suggests that tourists are looking for experiences that are
grounded in local cultures, traditions, and history, rather than those that are created purely for tourist consumption. This the-
ory has led to the development of a range of new tourism products and services that focus on providing tourists with authentic
experiences, such as agritourism, culinary tourism, and heritage tourism.
3. Service-dominant logic theory: This theory, developed by Vargo and Lusch, suggests that all economic exchange is fundamen-
tally a service exchange, in which value is co-created by the provider and the consumer. In the context of tourism, this theory
suggests that tourists are active participants in the creation of their own tourism experiences, and that businesses in the tour-
ism industry should focus on co-creating value with tourists rather than simply providing goods and services. This theory has
led to the development of new approaches to tourism marketing and management that emphasize the co-creation of value,
such as customer journey mapping and service design.
4. Flow theory: This theory, developed by Csikszentmihalyi, suggests that people are happiest and most fulfilled when they are
fully engaged in a challenging activity that requires their full attention and skills. In the context of tourism, this theory suggests
that tourists are seeking experiences that challenge them and require them to fully engage with their environment. This theory
has led to the development of new tourism products and services that focus on providing tourists with challenging and
immersive experiences, such as extreme sports tourism, wildlife tourism, and adventure tourism.
5. Emotional contagion theory: This theory, developed by Hatfield, Cacioppo, and Rapson, suggests that emotions are contagious,
and that people can “catch” emotions from others through their facial expressions, body language, and vocal cues. In the con-
text of tourism, this theory suggests that tourists are influenced by the emotions of the people they encounter during their
travels, including other tourists, locals, and tourism industry employees. This theory has led to a greater focus on the emotional
aspects of tourism experiences, including the emotions of tourists and the emotions of the people involved in the tourism in-
dustry. It has also led to the development of new approaches to tourism marketing and management that focus on creating
positive emotional experiences for tourists.

Question
Why is contagion theory important? How has this theory been used in tourism studies?
ChatGPT
The contagion theory is important in tourism studies because it highlights the role of emotions in shaping tourism experiences.
Emotions play a significant role in how tourists perceive and respond to different aspects of their travel experience, such as the
quality of service, the attractiveness of destinations, and the behavior of other tourists and locals. Therefore, understanding how
emotions are transmitted and experienced in tourism contexts can help tourism businesses and destinations better meet the emo-
tional needs and expectations of their customers.
Contagion theory has been used in tourism studies to explore the emotional aspects of tourism experiences. For example, re-
searchers have investigated how emotional contagion occurs among tourists and locals, and how it affects tourists' behavior and
attitudes. Studies have also examined the emotional labor involved in working in the tourism industry, and how emotions are
managed and expressed by tourism employees in their interactions with tourists.
Furthermore, the contagion theory has also been applied in tourism marketing and management. For example, destination
marketing organizations have used emotional appeals in their advertising campaigns to evoke positive emotions among potential
tourists and encourage them to visit. Similarly, tourism businesses have used strategies such as employee training and customer
service protocols to create positive emotional experiences for their customers.
Overall, the contagion theory provides a valuable framework for understanding and managing emotions in the context of tour-
ism, which can ultimately lead to more satisfying and memorable tourism experiences for tourists.
Question
Can you provide specific research studies using contagion theory in tourism studies?
ChatGPT
Certainly, here are a few examples of research studies that have applied the contagion theory to tourism:

1. Hennig-Thurau, Gwinner, Walsh, and Gremler (2004) conducted a study on the emotional contagion effects of service encoun-
ters in the tourism industry. The study found that the emotions expressed by service employees, such as positive affect, friend-
liness, and empathy, could be transmitted to customers, resulting in a positive emotional experience for the customers.
2. Gursoy, Kim, and Uysal (2004) investigated the emotional contagion effects of destination image on tourists' behavior inten-
tions. The study found that positive emotions evoked by destination images were positively related to tourists' intention to
visit, while negative emotions were negatively related to intention to visit.
3. Kim, Kim, and Han (2017) explored the emotional contagion effects of other tourists on tourists' satisfaction and behavioral
intentions in a theme park setting. The study found that the emotional expressions of other tourists, such as excitement and
enthusiasm, were positively related to tourists' satisfaction and intention to revisit the park.

4
D.R. Fesenmaier and K. Wöber Annals of Tourism Research 101 (2023) 103578

4. McCabe and Johnson (2013) examined the emotional labor involved in working in the tourism industry and its impact on
employee well-being. The study found that tourism employees who engaged in deep acting, or genuine emotional expres-
sion, had higher job satisfaction and better mental health outcomes than employees who engaged in surface acting, or faking
emotions.
5. Wang and Li (2016) investigated the emotional contagion effects of social media on tourists' destination choices. The study
found that positive emotions expressed by friends and acquaintances on social media platforms were positively related to tour-
ists' intention to visit a destination, while negative emotions were negatively related to intention to visit.

These studies demonstrate the relevance and utility of contagion theory in understanding the emotional dynamics of tourism
experiences and behaviors.
Question
I would argue that the work by Tversky and Kahneman should be included as a fundamental theory in support of the tourism
experience. Would you agree?
ChatGPT
Yes, I agree that the work of Tversky and Kahneman, particularly their prospect theory, has important implications for under-
standing tourism experience. Prospect theory is a seminal theory in behavioral economics that describes how individuals make
decisions under risk and uncertainty. This theory has been widely applied in various fields, including tourism, to explain how
tourists make choices and evaluate their experiences.
In the context of tourism, prospect theory can help explain how tourists perceive and respond to different aspects of their
travel experience. For example, tourists may be more sensitive to losses than gains when evaluating their travel experience,
which can influence their satisfaction and future behavior. Additionally, prospect theory can help explain why tourists may be
willing to take risks, such as engaging in adventurous activities, in certain situations but not in others.
Moreover, prospect theory can also help inform tourism marketing and management strategies. For example, using framing
techniques, such as emphasizing the potential losses of not visiting a destination, may be more effective in motivating tourists
to book a trip than emphasizing potential gains. Similarly, offering tourists options that are framed as “safe bets” may be more
attractive than options that are framed as risky.
Overall, the prospect theory provides a valuable framework for understanding the decision-making processes and emotional
responses of tourists, which can help tourism businesses and destinations better meet the needs and expectations of their cus-
tomers. Therefore, I agree that the work of Tversky and Kahneman should be included as a fundamental theory in support of tour-
ism experience.
Question
Are there linkages between prospect theory and contagion theory with regard to tourism experiences?
ChatGPT
Yes, there are linkages between prospect theory and contagion theory with regards to tourism experiences. Both theories focus
on the role of emotions in shaping decision-making and behavior in uncertain or ambiguous situations.
Prospect theory explains how individuals evaluate and make decisions under conditions of risk and uncertainty, taking into
account their perceived gains and losses. The theory suggests that people are more sensitive to losses than gains and tend to
make decisions based on how they perceive the outcomes relative to their reference point.
Contagion theory, on the other hand, explains how emotions are transmitted and experienced in social situations, including
tourism experiences. The theory suggests that emotions can be contagious, spreading from one person to another through social
interactions. Emotions can influence tourists' perceptions and evaluations of their travel experiences, as well as their behaviors
and intentions.
In the context of tourism experiences, both prospect theory and contagion theory suggest that emotions play a critical role in
shaping tourists' decision-making and behavior. For example, a tourist's evaluation of a particular travel experience may be influ-
enced not only by their own emotions but also by the emotions of other tourists or locals they encounter. Similarly, a tourist's
willingness to take risks or engage in adventurous activities may be influenced by their emotional state and how they perceive
potential gains and losses.
Therefore, understanding the interplay between prospect theory and contagion theory can help tourism businesses and desti-
nations better manage the emotional dynamics of tourism experiences, improve customer satisfaction, and enhance the overall
tourism experience.
Question
Are there other theories related to emotion that are important to understanding tourism experiences?
ChatGPT
Yes, there are several other theories related to emotions that are important to understanding tourism experiences. Here are
some examples:

1. Affective events theory (AET): AET suggests that emotions are triggered by specific events or situations and that they can have
a direct impact on individuals' attitudes and behaviors. In the context of tourism experiences, AET can help explain how pos-
itive or negative events during a trip can influence tourists' emotional responses, evaluations, and behaviors.
2. Self-perception theory: This theory suggests that people infer their attitudes and emotions based on their own behavior and
the context in which it occurs. In the context of tourism experiences, self-perception theory can help explain how tourists

5
D.R. Fesenmaier and K. Wöber Annals of Tourism Research 101 (2023) 103578

infer their attitudes and emotions towards a destination or activity based on their own behaviors and the social cues they re-
ceive from others.
3. Social identity theory: This theory suggests that individuals derive part of their self-concept from their membership in social
groups and that they are motivated to maintain a positive social identity. In the context of tourism experiences, social identity
theory can help explain how tourists' group membership and identification can influence their emotional responses, evalua-
tions, and behaviors, and how tourism businesses and destinations can leverage this to create positive group experiences.
4. Emotional labor theory: This theory suggests that individuals in service industries, including tourism, may need to manage
their emotions as part of their job requirements. In the context of tourism experiences, emotional labor theory can help explain
how tourism employees manage their emotions to meet customer expectations, and how this can impact their job satisfaction,
well-being, and the overall customer experience.
By applying these theories, tourism researchers and practitioners can gain a better understanding of the emotional dynamics of
tourism experiences, identify factors that contribute to positive or negative emotions, and develop strategies to enhance the emo-
tional aspects of the tourism experience.
Question
Also, Event Theory by JM Zacks seems to be quite important. Do you agree?
ChatGPT
Event segmentation theory, proposed by J. M. Zacks and his colleagues, is an important theory that can help us understand
how people perceive and remember events. This theory posits that events are segmented into discrete units of perception and
action, which are organized around a set of goals or objectives. Each event segment has a specific structure that includes a begin-
ning, middle, and end, and is characterized by specific features and actions.
In the context of tourism experiences, event segmentation theory can be applied to understand how tourists perceive and re-
member their travel experiences. By analyzing the different event segments that make up a tourism experience, researchers can
identify the key factors that contribute to tourists' overall evaluations and memories of their trip.
For example, event segmentation theory can help identify the specific stages or phases of a tourism experience, such as pre-
trip planning, transportation, accommodation, activities, and post-trip reflection. By examining the different features, actions, and
emotions associated with each event segment, researchers can identify the factors that contribute to tourists' satisfaction or dis-
satisfaction with each phase and their overall experience.
Moreover, event segmentation theory can also help inform tourism marketing and management strategies. By understanding
the specific event segments that are most important to tourists, tourism businesses and destinations can design and promote ex-
periences that meet their customers' needs and expectations, enhance their satisfaction, and create positive memories.
Therefore, I agree that event segmentation theory is an important theory for understanding tourism experiences, particularly
for identifying the key factors that contribute to tourists' overall evaluations and memories of their travel experiences.

References

Lund, B. D., Wang, T., Mannuru, N. R., Nie, B., Shimray, S., & Wang, Z. (2023). ChatGPT and a new academic reality: Artificial Intelligence-written research papers and the
ethics of the large language models in scholarly publishing. Journal of the Association for Information Science and Technology, 1–12. https://doi.org/10.1002/asi.24750.
Metz, C., & Collins, K. (2023). 10 ways GPT-4 is impressive but still flawed. New York Times March 15https://www.nytimes.com/2023/03/14/technology/openai-new-
gpt4.html March 15.
Mollick, E., & Mollick, L. (2023). Why all our classes suddenly became AI classes. Harvard Business Education. https://hbsp.harvard.edu/inspiring-minds/why-all-our-
classes-suddenly-became-ai-classes.
Nautiyal, R., Albrecht, J. N., & Nautiyal, A. (2023). ChatGPT and tourism academia. Annals of Tourism Research, 99 (March) pii/S0160738323000178https://www.
sciencedirect.com/science/article/.
Tufekci, Z. (2022). What would Plato say about ChatGPT? New York Times Dec. 15https://www.nytimes.com/2022/12/15/opinion/chatgpt-education-ai-technology.html.

Daniel Fesenmaier is an Adjunct Professor, Modul University. His academic interests are tourism planning and design, traveler behavior, the role and impact of technol-
ogy in tourism, and tourism education.
Karl Wöber is Full Professor and founding president of Modul University Vienna. His main research contributions are in the fields of decision support systems, strategic
marketing, and strategic planning, particularly in the field of city tourism. Karl Wöber is the elected president of the International Academy for the Study of Tourism and
has been a technical advisor to the City Destinations Alliance and the European Travel Commission for more than 20 years.