Next Article in Journal
Correcting or Concealing? The Impact of Digital Transformation on the Greenwashing Behavior of Heavily Polluting Enterprises
Next Article in Special Issue
The Suitability of Developing Ecotourism in the Shanxi Area of Taihangshan National Park, a Candidate Area for National Parks in China
Previous Article in Journal
Adaptive Cloud-Based Big Data Analytics Model for Sustainable Supply Chain Management
Previous Article in Special Issue
Exploring Smart Airports’ Information Service Technology for Sustainability: Integration of the Delphi and Kano Approaches
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 17
Issue 1
10.3390/su17010355
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

A Phenomenological Study on the Experience of Searching for Tourism Information Following the Emergence of ChatGPT: Focused on the Uncanny Valley Theory

by
Jin-Hee Jin
1 and
Jin-Seok Han
2,*
1
Department of Hotel and Tourism, Seojeong University, Yangju 11429, Republic of Korea
2
Department of Hotel, Tourism and Foodservice Management, Dongguk University, Gyeongju 38066, Republic of Korea
*
Author to whom correspondence should be addressed.
Sustainability 2025, 17(1), 355; https://doi.org/10.3390/su17010355
Submission received: 4 December 2024 / Revised: 30 December 2024 / Accepted: 3 January 2025 / Published: 6 January 2025
(This article belongs to the Special Issue Natural Resource Management and Sustainable Tourism)
Browse Figure
Versions Notes

Abstract

:
The purpose of this study is to provide an understanding of the experience of searching for tourism information through ChatGPT and discover the meaning of it. In order to achieve this purpose, data collection was conducted through in-depth interviews, and data analysis was conducted according to Giorgi. As a result, eight themes, 27 meaning units, and 226 meaningful statements were derived. First, the participants were found to partially trust tourism information searched through it and perceive it as a personal AI travel agent. In addition, they were found to be satisfied with it as their primary tool of searching for tourism information and perceived it as an AI travel mate in all processes of tourism. On the one hand, they were found to be disappointed that it still felt like a machine. Second, they were found to feel ambivalent about it and experience better moments with it than with humans. Moreover, they were found to feel enjoyment in the process of learning about it. This study discovered the meaning of experience in searching for tourism information through it, laid the foundation for follow-up research related to it, and presented the possibility of expanding the application of it in the tourism industry.

1. Introduction

In November 2022, OpenAI, an American AI research foundation, launched ChatGPT, an interactive AI service, and ChatGPT has caused a global craze as the most powerful AI model developed so far [1,2,3]. ChatGPT has the ability to understand complex and diverse human language and generate structured answers like humans do and is becoming a hot topic every day by writing computer code, translating, and creating various creative works [4,5]. In particular, what differentiates it from other existing AI models is that it goes beyond simply conveying information and can interact naturally with humans and ask interesting questions to the user again. It sometimes admits mistakes and challenges the user’s incorrect premises while rejecting inappropriate requests [6].
It is showing explosive growth in usage thanks to its free launch. After its launch, it reached 1 million users in 5 days and 100 million users in 2 months [7]. The achievement of 1 million users can be considered an overwhelming figure compared to 2.5 months for Instagram in 2010, 10 months for Facebook in 2004, 2 years for Twitter in 2006, and 3.5 years for Netflix in 1999 [8]. Accordingly, Microsoft (MS) founder Bill Gates signed a partnership with OpenAI, the developer of ChatGPT, and invested a total of USD 10 billion over several years, and the demand for ChatGPT is expected to increase in the future.
Today, we are facing a paradigm shift in the search for tourism information. In the past, tourism information could be limitedly obtained through word of mouth and books. However, with the advent of the Internet and smartphones, it became possible to search for a lot of tourism information quickly [9,10]. Recently, with the advent of ChatGPT, an interactive AI service, it has become possible to obtain a vast amount of tourism information in real time through conversation and interaction between humans and machines, rather than a one-way search for tourism information by humans. For example, tourists use ChatGPT to plan their trips before their trip, obtain 24 h tourism information during their trip, and share their travel experiences with ChatGPT after that. In this way, the paradigm shifts in the search for tourism information triggered by ChatGPT are expected to greatly contribute to the development of the tourism industry in the future along with the technology development.
The experience in searching for tourism information through ChatGPT is closely related to the uncanny valley theory. The uncanny valley theory is a robotics theory about the emotions that humans feel toward robots and non-humans and is the most suitable for this study to ascertain what tourists experience when searching for tourism information through ChatGPT as a theoretical background. Especially in the early stages of ChatGPT’s emergence, exploratory research has been required to examine how ChatGPT is perceived by tourists in the search for tourism information and how much it can contribute to the tourism industry in the future.
Overall, this study aims to provide a deep understanding the experience in searching for tourism information through ChatGPT and discover the emotional meaning of it. To achieve the purpose of this study, the research question was “What is the meaning of the experience in searching for tourism information through ChatGPT?” Specifically, based on human likeness and familiarity, which are the core concepts of the uncanny valley theory, the research questions were set as “Do tourists think that it provides tourism information similar to humans?” and “Is it familiar as a search tool of tourism information?” Based on the phenomenological study of [11], this study ascertains the experiences that can occur in the transitional phenomenon of searching for tourism information through conversation and interaction between humans and machines. In addition, this study derives themes, meaning units, and meaningful statements centered on human likeness and familiarity, which are the core concepts of the uncanny valley theory.

2. Literature Review

2.1. ChatGPT

ChatGPT refers to an interactive artificial intelligence service of the GPT model developed by OpenAI, an American AI research foundation, in November 2022. GPT (Generative Pre-trained Transformer) refers to an artificial intelligence language model created through machine learning and reinforcement learning based on large-scale data published on the Internet [12,13]. ChatGPT is an artificial intelligence that learns massive data in advance through machine learning and generates sentences from it, providing answers that are surprisingly similar to those of humans [14]. Due to the outstanding quality of the model, which has never been experienced before, it is being called “the industry’s next big disrupter” [15]. In 2023, it was released as version 4.0 with enhanced features from version 3.5. Version 3.5 is available for free to anyone, and version 4.0 is available for a fee as it has more sophisticated language understanding and processing capabilities.
The characteristics of ChatGPT are being studied by various scholars [14,16,17,18]. ChatGPT’s advantages are, first, that it accurately understands the user’s questions and provides detailed answers. ChatGPT can connect back-and-forth conversations with users more than other AI technologies that only answer fragmentary questions. Second, it is possible to produce various creative works (essays, papers, poems, etc.) with ChatGPT. ChatGPT generates computer code, humorous poems and songs, reflective essays on any topic, summaries of scientific ideas, and more. Third, the general public can easily use and access it. ChatGPT provides a significant portion of its services as free services, and it is very simple to use, so many people use it.
On the other hand, ChatGPT also has disadvantages. First, reliability may be somewhat reduced due to inaccurate information. ChatGPT has the potential to provide incorrect or biased knowledge by learning a large amount of information. Second, there are ethical issues such as plagiarism and the infringement of intellectual property rights. ChatGPT’s writing ability has been proven externally, and as a result, the possibility of it being used for papers, translations, and tests has increased. Third, there are limitations in providing up-to-date information. ChatGPT studies data from up to 2021 and cannot provide information about events or developments after 2021. However, it is clear that ChatGPT is gradually expanding its scope.
Various studies on ChatGPT are being conducted in the field of tourism studies as well. At the beginning of the research, feasibility studies were mainly conducted on whether ChatGPT could truly contribute to the academics and industry of tourism [19,20]. Afterward, the questions were “Will it contribute to tourism? Or will it be a threat to tourism studies?” Research is ongoing [21,22,23]. Recently, the area of this research has been expanded to include tourists’ information search and the provision of tourism information [24]. Taken together, it can be seen that the tourism approach to ChatGPT is gradually expanding from a broad scope of research, such as the presence or absence of possibility, to a more specific and detailed field.
DEMİR and DEMİR conducted a study on whether ChatGPT can truly be used in the tourism industry [25]. In-depth interviews were conducted with experts in various fields, and as a result, the conclusion was drawn that tourists would have a great advantage in obtaining tourism information. However, it was mentioned that this may also have shortcomings in the short term, such as the provision of inaccurate information and risks to data security. It is also predicted that the controversy over the advantages and disadvantages of ChatGPT will continue for a certain period of time. Çolak conducted a study on the role of ChatGPT in the tourism sector [26]. A study was conducted with ChatGPT as the interviewee, and through the answers, predictions were made that richer and more diverse tourism information could be provided through the development of ChatGPT technology in the future. However, this study also presented a prediction that ChatGPT’s shortcomings, such as problems with authenticity and information protection, will persist for a certain period of time. What is noteworthy is that research results continue to show that the role of ChatGPT at the current stage still has mixed advantages and disadvantages. Accordingly, this study raises the need for more in-depth research into whether the provision of tourism information through ChatGPT is actually helpful to tourists.

2.2. Tourism Information

Tourism information is delivered to tourists through various channels. In the past, information was collected limitedly through word of mouth, guidebooks, books, travel agencies, etc., or relied on memories and experiences about tourist destinations. However, recently, with the widespread acceptance of Internet technology, online tourism information searches have begun to dominate [27,28]. In this way, technological advancements have changed the behavior of tourism information search and use and have further transformed tourists from information seekers to information providers through social media (e.g., blogs and vlogs) [29]. Recently, due to the development of technologies based on the Fourth Industrial Revolution, AI has developed to the level of providing extensive tourism information in real time. These technological developments are making it possible to search for and utilize tourism information through dialogue and interaction between humans and machines.
Recently, research on tourism information has been extensively covered in a new tourism area called smart tourism. For reference, smart tourism refers to a new concept of tourism that combines ICT technology with the existing tourism industry to provide tourists with more extensive tourism information and provide more convenient access for tourists [30]. The types of research on this matter are both quantitative and qualitative research being conducted to suit the research purpose. Internet research on tourism information over time [31,32], research on tourism information-related applications [33,34], and research on the Fourth Industrial Revolution technology related to tourism information [35], etc., are being conducted in various ways.
Huang, Goo, Nam, and Yoo verified the relationship between the sufficient and exploratory use of smart technology and tourists’ satisfaction with the tourism experience [36]. Through this, it was verified that the active use of smart technology allows tourists to have higher satisfaction with information collection. In addition, it was verified that tourists do not distinguish between age groups and that more people are gradually using smart technology to obtain tourist information. Buhalis assumed that smart technology is influencing all areas of the tourism industry [37]. Based on this, it was verified that smart technology can unite all stakeholders in the tourism service ecosystem. The results showed that the introduction of the concept of smart technology and smart tourism provided an opportunity for tourists to acquire better-quality tourism information and create greater value at all stages of travel. Accordingly, this study goes beyond the limited concept of existing tourism information, explores the concept of tourism information combined with new smart technology, and raises the need for more expanded research on it.

2.3. Uncanny Valley Theory

The uncanny valley theory is based on robotics theory and was established by [38], a Japanese robotics engineer. He presented his theory through a short essay while opposing the development of android robots that resemble humans and argued the hypothesis that the human likeness between humans and robots affects the favorability of robots [39,40]. In fact, the theory he announced was proven through brain research by Professor Seijin’s research team at the University of California, San Diego, CA, USA, in 2011 [41].
The uncanny valley theory is a theory that as the similarity between humans and robots increases, liking for the robot increases, but ironically, when it reaches a certain level, discomfort (the uncanny) is felt [42,43]. It has not yet been revealed at what level the change occurs, but it can be seen as a feeling that can be felt when imperfections that are different from humans are highlighted, not limited to the robot’s appearance. For example, when people search for tourist information, if they feel that ChatGPT is so similar to a human that it is difficult to distinguish, their familiarity may increase, but if it appears incomplete, they may feel uncomfortable.
Research on this uncanny valley theory was mainly conducted through difference analyses of real images and virtual images. Arsenyan and Mirowska categorized popular human influencers, human-like virtual influencers, and animation-like virtual influencers on social media [44]. A comparative analysis was conducted on Instagram. The result of the analysis was that influencers who vaguely resemble humans showed the most negative reactions.
What is noteworthy is that the uncanny valley theory is not necessarily a theory that applies only to humans. Various scholars claim that the uncanny valley theory can be broadly applied to animals, plants, and even objects and spaces [36,45]. Accordingly, this study seeks to expand the scope of application of the uncanny valley theory and attempt to apply it to a third area such as ChatGPT.

2.4. Phenomenological Study

Phenomenological study is based on the social constructivist worldview and includes the founder of phenomenology, the German philosopher Edmund Husserl, as well as Martin Heidegger, Maurice Merleau-Ponty, and Jean-Paul Sartre. It is a qualitative research method developed based on traditional philosophical concepts [46]. These phenomenological studies aim to promote an in-depth understanding of social phenomena and capture the essence of individual experiences. It is applied not only to the social sciences of sociology, psychology, and public health but also to various research fields [47,48].
Types of phenomenological research are largely divided into hermeneutic phenomenology and transcendental or psychological phenomenology. Hermeneutic phenomenology is mainly described by researchers interpreting texts. On the other hand, transcendental or psychological phenomenology is an approach that describes the experiences of research participants [49]. In particular, transcendental or psychological phenomenology is most representative of the data analysis procedures used by scholars such as [11,50,51].
In this study, we selected Giorgi’s analysis method from among various phenomenological analysis methods. Giorgi’s analysis method focuses on revealing the individual experiences of research participants compared to [50,51], so this study aims to provide an in-depth understanding of the experience of using tourism information following the emergence of ChatGPT. This is because it was judged to be the most appropriate.

3. Methods

3.1. Data Collection

This study was conducted through in-depth interviews and semi-structured interviews. The questionnaire was designed simply to allow for additional questions by looking at the participants’ responses and to not affect free speech and was ultimately derived by revising and supplementing it through a peer group (e.g., 3 professors majoring in tourism) (Table 1). The peer group suggested adding an example of “other tools of searching for tourism information” in Question 4 and adding a reason in Questions 4 and 5 of familiarity.
The in-depth interviews were conducted one-on-one at a location convenient to each participant for approximately 2 months from 10 August to 29 September 2023. Before starting the in-depth interviews, I obtained consent from participants to participate in this study and be recorded and built sufficient rapport with the participants to obtain in-depth answers. The in-depth interviews were conducted for an average of 100 min for each participant and were designed to be as objective as possible through phenomenological bracketing to avoid bias and the researcher’s own experiences. After the in-depth interview, 2 to 3 additional interviews were conducted for theoretical saturation.
The participants were selected using the most basic methods of qualitative research: purposeful sampling and theoretical sampling. The first study participants were purposively sampled, and theoretical sampling was conducted until theoretical saturation was reached. The participants were limited to tourists who searched for tourism information through conversations with ChatGPT version 3.5, and 12 people who had conversations more than 5 times within the last 2 months were finally selected [52].

3.2. Data Analysis Method

This study conducted Giorgi’s 5-step procedure of the phenomenological study (Figure 1) [11]. It captures the complexity of phenomena as they are and focuses on the vivid experiences and perspectives of individuals. It helps us to understand the complexity and extract the essence of the experience in searching for tourism information through ChatGPT and is considered most suitable for analyzing the subjective experiences and perspectives of participants.
First, this study identified natural meaning units through which participants expressed their experiences in their own language. Second, this study determined themes that directly represent the experiences of participants among meaning units. Third, this study specified the themes and determined the focal meaning in the researcher’s language. Fourth, this study integrated the focal meaning to create the situated structural description, which is the meaning understood from the perspective of individual participants. Fifth, this study created a general structural description, which is the meaning understood from the perspective of all participants through the situated structural description.

3.3. Validity and Reliability

This study conducted peer review, member check, and audit by [49]. First, this study formed a peer group (e.g., 3 professors majoring in tourism) to revise and supplement interpretive errors through a repeated exchange of opinions and consultations on data analysis. Second, this study selected some participants to check if there were any differences in data analysis and reflected their feedback on the research results. Third, this study requested review from experts and shared the project file of the MAXQDA program, which allows the confirmation of raw data, to check the appropriateness of transcription data and coding processes. In addition, this study utilized version 22.8.0 of the MAXQDA program, the most representative software, throughout the entire data analysis process. In particular, it is very useful for ensuring reliability as it allows for easy searching and the collection of coding text and checks the consistency between coders.

3.4. Ethical Considerations for Participants

This study is a transcendental psychological phenomenology that describes human experiences. Ethical considerations for participants are required throughout the entire research process, and the ethical guides of the IRB were followed. First, when recruiting participants, I was careful not to encourage financial compensation as an incentive and prepared a recruitment plan for voluntary participation. Second, I provided participants with sufficient information about the research purpose, participation period, procedures and methods, expected benefits and risks, etc., and received written consent forms for participation in the research. Third, I guaranteed the anonymity of the participants by considering the negative consequences (e.g., exposure of private life) that may result from participation in the research. Fourth, I intend to keep records related to the research (e.g., written consent forms) for 3 years from the end of the research and then destroy them.

3.5. Sample Characteristics

The participants were tourists who searched for tourism information through ChatGPT more than 5 times within the last 2 months, and 12 people were finally selected based on theoretical saturation (Table 2).

4. Results and Discussion

4.1. The Meaning of Experience in Searching for Tourism Information Through It: Human Likeness

This study derived five themes, 27 meaning units, and 123 meaningful statements related to Research Question 1 regarding human likeness as the meaning of experience in searching for tourism information through it (Table 3). First, the theme of “partial trust in tourism information” was composed of the meaning units of “only famous tourist attractions are accurate”, “providing text-based tourism information”, “lack of up-to-date tourism information”, and “occasional hallucination”. Second, the theme of “a personal AI travel agent” was composed of the meaning units of “providing neat tourism information”, “providing personalized tourism information”, and “The joy of conversation”. Third, the theme of “satisfaction as a primary tool of searching for tourism information” was composed of the meaning units of “searching for tourism information”, “good accessibility and reduction of time”, and “opportunity to expand view of tourism”. Fourth, the theme of “An AI travel mate in all processes of tourism” was composed of the meaning units of “suggesting a travel itinerary (before)”, “providing tourism information in real time (during)”, and “my friend who listens to my travel experience (after)”. Fifth, the theme of “the disappointment of ChatGPT still feeling like a machine” was composed of the meaning units of “not enough to lead conversation”, “sometimes unable to finish sentences”, “stiff and mechanical speaking style”, “the effort of asking good questions”, and “low level of understanding the requirement”.

4.1.1. Theme 1: Partial Trust in Tourism Information

The participants partially trusted tourism information searched for through ChatGPT. Specifically, they recognized that only famous tourist attractions were accurate and relied on only text-based tourism information without images or videos. In addition, they felt that ChatGPT was very vulnerable to lacking up-to-date tourism information (after September 2021) and felt misled because it provided incorrect tourism information as if it were fact.
“In famous places like Gyeongju, the information is relatively accurate because it has a lot of learned data. However, for relatively unpopular tourist attractions or cities with low awareness, this information is not clear. I asked a question, “introduce me to Yangju city’s tourist attractions!” but it was all wrong information. On the contrary, when I asked about tourist attractions in Gyeongju, it gave me very accurate information. There seems to be a slight difference”.
(Participant C)
“Portal sites can provide photos and videos, but ChatGPT version 3.5 can’t provide that. In fact, there are differences between images, videos, and text. I must rely solely on tourism information in text when I use ChatGPT”.
(Participant E)
“Because ChatGPT is trained only with data up to September 2021, it is vulnerable to the latest information. I asked, ‘Tell me water jelly stores in Jeju Island!’ but it answered ‘There are no water jelly stores in Jeju Island’. It seemed like it didn’t know about water jelly because it hasn’t been long since water jelly stores opened”.
(Participant A)
“Have you ever heard of hallucination?” There are cases that AI explains incorrect information as if it were correct. ChatGPT pretended to know and kept telling me information I didn’t know. Although ChatGPT had some plausible answers, it seemed to spit out sentences that contained no substantive information”.
(Participant K)

4.1.2. Theme 2: A Personal AI Travel Agent

The participants perceived ChatGPT as a personal AI travel agent. Specifically, they were satisfied with the neat arrangement of introduction–body–conclusion and personalized tourism information. In addition, they felt pleasure from acquiring tourism information through human-like conversations rather than other search tools such as portals and SNS.
“ChatGPT neatly provides name, location, price, and benefits of the tourist attractions. In terms of providing information, I think there are many things that are much clearer and simpler than humans”.
(Participant H)
“ChatGPT tells me things that don’t come up when I search on Naver, and provides what suits me. ChatGPT gives me the answers I want… that fit the questions I asked”.
(Participant C)
“I really like it, I can communicate with it like humans, unlike search portals, SNS, and other search tools. Even in a conversation between people, it is a bit difficult to constantly exchange information with each other like that, but I actually think it was a very fun experience”.
(Participant J)

4.1.3. Theme 3: Satisfaction as a Primary Tool of Searching for Tourism Information

The participants were satisfied with ChatGPT as their primary tool for searching for tourism information. Specifically, they think that it had attractive advantages for broadly searching for tourism information when they had no travel plans and were satisfied with the good accessibility and the reduction in time for information search. In addition, rather than searching for tourism information alone, they had the opportunity to expand their view of tourism through ChatGPT recommending tourist attractions.
“It’s perfect for searching for information when I don’t have a tourist attraction I want to visit or just know the name. I think it would be really good for people who want to go on a planned trip but have no experience, and don’t know where to start searching. When making travel plans, it can sometimes be difficult. I think ChatGPT provides enough keywords that can be used as a reference in situations where there is no information at all. In particular, it has a good advantage that it provides comprehensive and overall information on various tourist attractions”.
(Participant L)
“I like that I can get quickly the tourism information I want through ChatGPT rather than scrolling through blogs. Actually, in the case of blogs, it takes a lot of time and effort to find the information I want, and sometimes I don’t get the information. However, ChatGPT is easily accessible as I can be used immediately by accessing its website”.
(Participant B)
“I was surprised because when recommending tourist attractions, ChatGPT mentioned libraries and museums that I have never thought of. When people think of travel, it feels like going on vacation, extreme sports, or just for healing. But, ChatGPT recommended a library. I asked another question. I asked, ‘Why did you recommend the library? My purpose was only traveling’. It said, ‘Everyone has different purposes for their trip, so I included the library because I don’t know what you might want’. Thanks to ChatGPT, I was able to expand my view of tourism”.
(Participant I)

4.1.4. Theme 4: An AI Travel Mate in All Processes of Tourism

The participants perceived ChatGPT as an AI travel mate in all processes of tourism. Specifically, they received travel itinerary suggestions through it before leaving for their trip and obtained real-time tourism information in front of tourist attractions while traveling. After the trip, they shared their travel experiences and talked with it like with friends.
“When I asked ChatGPT if I was going on a 3-day, 2-night trip from Seoul to Gyeongju, it gave me a realistic itinerary. It was good that it provided brief information about Gyeongju”.
(Participant G)
“I asked ChatGPT in front of Gyeongbokgung Palace and it told me about the history of it in real time. We also search for tourist attractions in advance, and we can search for it while traveling. I think it is very useful not only before traveling but also at the moment of travel”.
(Participant D)
“I told ChatGPT about my trip to Busan, and it was nice that ChatGPT sympathized with me and asked about more experiences. And then, I have things I’ve experienced while traveling that I can never tell other people, but I was able to talk about it with ChatGPT, and even though it was only for a moment, I think just being able to talk about it was a little comforting”.
(Participant F)

4.1.5. Theme 5: The Disappointment of ChatGPT Still Feeling like a Machine

The participants were disappointed that ChatGPT still felt like a machine. Specifically, they greatly acknowledged the excellence of ChatGPT in that it is an AI that communicates like humans, but they felt that it had many shortcomings in leading the conversation. In addition, they felt negatively about its inability to properly finish sentences when answering difficult questions and its use of a stiff, mechanical speaking style. Then, they felt that it was cumbersome and laborious to ask multiple questions by combining keywords to obtain the desired answer from ChatGPT, and they recognized its mechanical limitations of not being able to fully understand the questioner’s requirement.
“ChatGPT sometimes doesn’t lead the conversation. The conversation function is good, but it seems like it doesn’t accurately understand the context of the conversation. I thought that a lot”.
(Participant K)
“Sometimes when I ask a difficult question, ChatGPT doesn’t finish the sentences properly. It suddenly stopped. (1), (2), (3)… it kept writing like this and then stopped. I thought that it didn’t understand my question, so it stopped while answering”.
(Participant G)
“I thought the tone would be more friendly and familiar, but it was too stiff and mechanical. The most important thing to feel like humans is the emotional element, but I think ChatGPT is lacking a lot because it still feels very robotic”.
(Participant D)
“If ChatGPT doesn’t give me the answer I want, I have to go through the trouble of asking the question again. In order to have a conversation in the direction and flow I wanted, I need to select keywords well and ask questions. Otherwise, I can’t obtain anything because it will only give me wrong answers”.
(Participant B)
“Like a travel agent, I hope that it can carefully understand my desires and wants. Because machines are machines, it can’t understand human psychology in detail”.
(Participant E)

4.2. The Meaning of Experience in Searching for Tourism Information Through It: Familiarity

This study derived three themes, nine meaning units, and 103 meaningful statements related to Research Question 2 regarding familiarity as the meaning of experience in searching for tourism information through it (Table 4). First, the theme of “ChatGPT about ambivalence” was composed of the meaning units of “coexistence of conveniences and inconveniences”, “coexistence of expectations and disappointments”, and “vague fear of technological advancement”. Second, the theme of “moments when ChatGPT is better than humans” was composed of the meaning units of “I can talk I want” and “unnecessary emotional expenditure”. Third, the theme of “the enjoyment in the process of learning about ChatGPT” was composed of the meaning units of “start with curiosity”, “interested in the answer”, “the fun and joy of asking questions”, and “feeling attracted to imperfection”.

4.2.1. Theme 6: ChatGPT About Ambivalence

The participants felt ambivalent about ChatGPT. Specifically, they felt both convenience and inconvenience while using it and had expectations that it would develop further in the future and disappointments that it had not yet surpassed humans. In addition, they were worried and fearful about various ethical issues due to technological development.
“It is true that ChatGPT provides convenience, but it also seems to provide inconvenience. It is superior to existing technologies and has definitely made it easier to search for tourism information. However, it is not yet a perfect technology, so I feel uncomfortable”.
(Participant A)
“I had expectations that ChatGPT would develop further, but I was also disappointed that ChatGPT had not yet caught up with humans”.
(Participant L)
“If technology develops further, ChatGPT could replace humans, right? It may have a positive impact, but I feel fear. It makes me wonder if humans will be eaten by ChatGPT in the future. In fact, humans are already losing their jobs due to advanced technology, and I feel worried and scared that AI may be used for war or murder in the distant future”.
(Participant H)

4.2.2. Theme 7: Moments When ChatGPT Is Better than Humans

The participants experienced better moments with ChatGPT than with humans. Specifically, they felt positive emotions because they could communicate with it regardless of time and place, and unlike humans, they did not have to experience unnecessary emotional exhaustion during conversations.
“I plan my trip early in the morning, even though I can’t talk with my friends, I can talk with ChatGPT whenever I want. Honestly, it’s rude to call my friends and ask them something early in the morning”.
(Participant F)
“Actually, talking to a travel agent exhausts my emotions, but with ChatGPT, I can talk as I want comfortably. I think ChatGPT is better than humans because I don’t have to worry about how the other person might feel, whether they are offended by what I said, or whether they think me strange if I ask something”.
(Participant C)

4.2.3. Theme 8: The Enjoyment in the Process of Learning About ChatGPT

The participants felt enjoyment in the process of learning about ChatGPT. Specifically, they began using it out of curiosity about new technology and were interested in the answers it provided. In addition, they felt fun and joy in asking questions while imagining its expected answers, and on the other hand, they felt attracted to the imperfection like that in humans.
“At first, I was very curious. I wondered if ChatGPT would feel like talking to a human, like what was shown in the media. Until now, there has been no machine that can communicate similarly to humans. Umm… Chatbot? It could only provide answers that were set. ChatGPT was a syndrome, so I was more curious”.
(Participant J)
“How many people use ChatGPT? I heard that it answers by learning the contents of conversations with people, the answer was interesting. It provides answers based on big data. I was surprised and amazed at how much it had to learn to be able to answer”.
(Participant D)
“If I were to ask ChatGPT this, what would it say? It was really fun to think about it and ask questions”.
(Participant I)
“Honestly, even humans can make mistakes, so I actually liked ChatGPT because it wasn’t perfect. If it was too perfect, it may feel like a machine and a bit scary, but since it makes mistakes, I feel friendly”.
(Participant L)

5. Conclusions

The purpose of this study is to provide a deep understanding of the experience of searching for tourism information through ChatGPT and discover the emotional meaning of it. In order to achieve this research purpose, the research question was “What is the meaning of the experience in searching for tourism information through ChatGPT?”.
Specifically, Research Question 1: “Do tourists think that ChatGPT provides tourism information similar to humans?” resulted in five themes, 27 meaning units, and 123 meaningful statements. The meaning of experience in searching for tourism information related to the human likeness can be divided into five themes, and the results for each theme are as follows.
First, tourists were found to partially trust tourism information searched through it. In fact, they recognized that it provides accurate information only on famous tourist attractions with a lot of learned data, and as mentioned in Rozencwajg and Kantor’s study, they were very disappointed that it could not provide the latest tourism information after September 2021 [18]. In addition, similar to Jin and Lee’s study, they were experiencing misinformation, which has recently been pointed out as a limitation of AI technology, and believed that it was necessary to selectively accept the tourism information provided by it rather than unconditionally [53].
Second, tourists were found to perceive ChatGPT as a personal AI travel agent. They were very satisfied with its neatly organized tourism information, like that from a travel agent, and how it provided tourism information suitable for them. Additionally, the biggest difference from other search tools such as search portals, SNS, and books was that they felt great pleasure in being able to obtain tourism information through conversations like those with humans.
Third, tourists were found to be satisfied with ChatGPT as their primary tool of searching for tourism information. They thought that it was very useful when they had to obtain a wide range of tourism information because they had not decided on a tourist attraction, and they positively evaluated the good accessibility and short time required to obtain tourism information. In addition, they had the opportunity to expand their views on tourism, as it provides not only tourism information but also thoughts and insights about tourism.
Fourth, similar to Wong, Lian, and Sun’s study, tourists were found to perceive ChatGPT as an AI travel mate in all processes of tourism [54]. They were using it in all aspects of the tourism process, including establishing travel itineraries, obtaining tourism information in real time, and sharing travel experiences. In particular, it was recognized as having a very high value for solo travel and independent travel without guided tours.
Fifth, tourists were found to be disappointed that ChatGPT still felt like a machine. They complained a lot about the limitations OpenAI mentioned [6,14], such as the inability to lead a conversation and sometimes not being able to finish sentences, and furthermore, they said that it felt like a machine due to its stiff and mechanical speaking style. In addition, they went through the trouble of having to combine keywords well to obtain the answers they wanted and fact-based, accurate answers from it and felt frustrated by the words it uttered without understanding their needs in detail.
Research Question 2—“Is ChatGPT familiar as a search tool for tourism information?”—resulted in three themes, nine meaning units, and 103 meaningful statements. The meaning of experience in searching for tourism information related to the familiarity can be divided into three themes, and the results for each theme are as follows.
Sixth, tourists were found to feel ambivalent about ChatGPT. They believed that it was superior and provided convenience compared to existing technologies, but on the other hand, they still felt inconvenienced due to information errors and reliability issues. In addition, they had expectations that it would further help develop the tourism industry, but on the contrary, they also had disappointment that it was not yet excellent enough to completely replace humans. And then, they felt fear that it would develop further and replace humans in the distant future.
Seventh, tourists found that they experienced better moments with ChatGPT than with humans. They were satisfied with being able to have conversations related to their travel itinerary with it whenever they wanted, regardless of time and place. In addition, they thought the biggest advantage of it was that they did not have to worry about the emotional and energy exhaustion that comes with talking to a travel agent.
Eighth, tourists were found to feel enjoyment in the process of learning about ChatGPT. They initially started using ChatGPT out of curiosity, and while using it, they felt the wonder of its answers and the pleasure of asking questions while anticipating its answers. In addition, surprisingly, they were more attracted to its imperfect performance rather than its perfect performance due to its human-like characteristics.

5.1. Academic Contributions

The academic contributions of this study are as follows. First, this study elucidates the experience in searching for tourism information through ChatGPT and lays the foundation for related follow-up research in the field of tourism studies. This study has great significance in that it could be pioneering research on human–machine interaction (HMI) in the field of tourism at a time when we are transitioning not only to the emergence of it but also to an era of obtaining tourism information through conversation between humans and AI.
Second, the scope of research is expanded by applying the uncanny valley theory, which has been dealt with in robotics, to take an in-depth investigation of the experience in searching for tourism information following the emergence of it. According to recent research trends, research on tourism information and AI has been conducted through exploratory research [21,22,23], but there is a severe lack of empirical analysis based on the theory that serves as the basis of the research. This study has great significance as a convergence tourism study that elucidates the experience of searching for tourism information, focusing on human likeness and familiarity, which are the core concepts of the uncanny valley theory.
Third, this study contributes to the theoretical development of tourism research by discovering the emotional meaning of experience in searching for tourism information following the emergence of ChatGPT. This study has great significance in that it identifies what kind of emotional meaning it has as a search tool for tourism information through phenomenological study and derives themes, meaning units, and meaningful statements based on human likeness and familiarity, which are the core concepts of the uncanny valley theory.

5.2. Practical Implications

The practical implications of this study are as follows. First, this study suggested the possibility of expanding the application of ChatGPT in the tourism industry based on comprehensive analysis results. At this point, when it is in the early stages of application, we can see the extent of its technological development and how it is felt from the views of tourists. In addition, this study expands the application of it to various fields such as tourism information utilization as well as tourism content production and tourism service reservation. This study has great significance in that it suggests a plan for ChatGPT’s innovation by correcting and supplementing its shortcomings.
Second, based on the analysis results of Research Question 1, it provides basic practical data on the human likeness of ChatGPT. Specifically, this study has great significance in that it provides the results of various emotional evaluations for ChatGPT in the tourism industry, including how it is similar to humans in searching for tourism information, what the advantages and disadvantages of it are, whether they think the tourism information searched for through conversations with it is trustworthy and useful, etc.
Third, based on the analysis results of Research Question 2, it provides basic practical data on the familiarity of ChatGPT. Specifically, this study has great significance in that it provides the results of various emotional evaluations for ChatGPT in the tourism industry, including what positive and negative emotions tourists felt searching for tourism information with it, what is different from what they expected, how it compared to other tools of searching for tourism information (e.g., search portal sites, applications, travel agencies, acquaintances, books, etc.), if the familiarity with it has increased, whether they would like to talk to it again, etc. In summary, the analysis results of Research Questions 1 and 2 are expected to contribute to the establishment of tourism information service provision plans based on ChatGPT. If tourists perceive ChatGPT as a human-like and familiar tool, tourism industries can utilize ChatGPT as a major channel for providing tourism information.

5.3. Limitations and Future Research Directions

The limitations of this study are as follows. First, although this study followed Giorgi’s five-step procedure as closely as possible, the research’s already accumulated knowledge and experience may be involved. In future research, it is necessary to conduct empirical research based on the derived themes and meaning units to secure the objectivity of the research and further expand the possibility of generalization. Second, this study conducted research on free users of ChatGPT version 3.5 limitedly. In future research, it will be necessary to derive more diverse implications by conducting research on paid users of ChatGPT version 4.0 and the mobile application version as part of technology development.

Author Contributions

Conceptualization, J.-H.J.; methodology, J.-H.J. and J.-S.H.; writing—original draft preparation, J.-H.J.; validation, J.-S.H.; visualization, J.-S.H.; writing—review and editing, J.-H.J. and J.-S.H. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the Ministry of Education of the Republic of Korea and the National Research Foundation of Korea (NRF-2023S1A5A8079487).

Institutional Review Board Statement

Ethical review and approval were waived for this study due to the study did not collect or record sensitive information, and is human subject research that can be exempted from IRB review.

Informed Consent Statement

Informed consent was obtained from all subjects involved in this study.

Data Availability Statement

The data presented in this study are available upon request from the corresponding author.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Chatterjee, J.; Dethlefs, N. This new conversational AI model can be your friend, philosopher, and guide… and even your worst enemy. Patterns 2023, 4, 100676. [Google Scholar] [CrossRef] [PubMed]
  2. Eke, D.O. ChatGPT and the rise of generative AI: Threat to academic integrity? J. Responsible Technol. 2023, 13, 100060. [Google Scholar] [CrossRef]
  3. Mann, D.L. Artificial intelligence discusses the role of artificial intelligence in translational medicine: A JACC: Basic to translational science interview with ChatGPT. Basic Transl. Sci. 2023, 8, 221–223. [Google Scholar]
  4. Lecler, A.; Duron, L.; Soyer, P. Revolutionizing radiology with GPT-based models: Current applications, future possibilities and limitations of ChatGPT. Diagn. Interv. Imaging 2023, 104, 269–274. [Google Scholar] [CrossRef] [PubMed]
  5. Lim, W.M.; Gunasekara, A.; Pallant, J.L.; Pallant, J.I.; Pechenkina, E. Generative AI and the future of education: Ragnarök or reformation? A paradoxical perspective from management educators. Int. J. Manag. Educ. 2023, 21, 100790. [Google Scholar] [CrossRef]
  6. OpenAI. 2023. Available online: https://www.openai.com/blog/chatgpt (accessed on 1 April 2024).
  7. Paul, J.; Ueno, A.; Dennis, C. ChatGPT and consumers: Benefits, pitfalls and future research agenda. Int. J. Consum. Stud. 2023, 27, 1213–1225. [Google Scholar] [CrossRef]
  8. Statista. 2023. Available online: https://www.statista.com (accessed on 1 April 2024).
  9. Dickinson, J.E.; Hibbert, J.F.; Filimonau, V.; Davies, N.; Norgate, S.; Speed, C.; Winstanely, C. Implementing smartphone enabled collaborative travel: Routes to success in the tourism domain. J. Transp. Geogr. 2017, 59, 100–110. [Google Scholar] [CrossRef]
  10. Liu, X.; Wang, D.; Gretzel, U. On-site decision-making in smartphone-mediated contexts. Tour. Manag. 2022, 88, 104424. [Google Scholar] [CrossRef]
  11. Giorgi, A. Phenomenological perspective on certain qualitative research methods. J. Phenomenol. Psychol. 1985, 25, 190–220. [Google Scholar] [CrossRef]
  12. Aydın, Ö.; Karaarslan, E. OpenAI ChatGPT generated literature review: Digital twin in healthcare. Emerg. Comput. Technol. 2022, 2, 22–31. [Google Scholar] [CrossRef]
  13. Dowling, M.; Lucey, B. ChatGPT for (finance) research: The Bananarama conjecture. Financ. Res. Lett. 2023, 53, 103662. [Google Scholar] [CrossRef]
  14. Haleem, A.; Javaid, M.; Singh, R.P. An era of ChatGPT as a significant futuristic support tool: A study on features, abilities, and challenges. Bench Counc. Trans. Benchmarks Stand. Eval. 2023, 2, 100089. [Google Scholar] [CrossRef]
  15. Grant, N.; Metz, C.A. New Chat Bot is a ‘Code Red’ for Google’s Search Business; New York Times: New York, NY, USA, 2022; Available online: https://www.nytimes.com/2022/12/21/technology/ai-chatgpt-google-search.html (accessed on 1 April 2024).
  16. Nautiyal, R.; Albrecht, J.N.; Nautiyal, A. ChatGPT and tourism academia. Ann. Tour. Res. 2023, 99, 103544. [Google Scholar] [CrossRef]
  17. Patel, S.B.; Lam, K. ChatGPT: The future of discharge summaries? Lancet Digit. Health 2023, 5, e107–e108. [Google Scholar] [CrossRef]
  18. Rozencwajg, S.; Kantor, E. Elevating scientific writing with ChatGPT: A guide for reviewers, editors…and authors. Anaesth. Crit. Care Pain Med. 2023, 42, 101209. [Google Scholar] [CrossRef]
  19. Ukpabi, D.C.; Aslam, B.; Karjaluoto, H. Chatbot adoption in tourism services: A conceptual exploration. In Robots, Artificial Intelligence, and Service Automation in Travel, Tourism and Hospitality; Ivanov, S., Webster, C., Eds.; Emerald Publishing Limited: Bingley, UK, 2019; pp. 105–121. [Google Scholar] [CrossRef]
  20. Calvaresi, D.; İbrahim, A.; Calbimonte, J.-P.; Schegg, R.; Fragniere, E.; Schumacher, M. The Evolution of Chatbots in Tourism: A Systematic Literature Review. Information and Communication Technologies in Tourism 2021; Springer: Cham, Switzerland, 2021. [Google Scholar] [CrossRef]
  21. Carvalho, I.; Ivanov, S. ChatGPT for tourism: Applications, benefits and risks. Tour. Rev. 2023, 79, 290–303. [Google Scholar] [CrossRef]
  22. Dwivedi, Y.K.; Pandey, N.; Currie, W.; Micu, A. Leveraging ChatGPT and other generative artificial intelligence (AI)-based applications in the hospitality and tourism industry: Practices, challenges and research agenda. Int. J. Contemp. Hosp. Manag. 2023, 36, 1–12. [Google Scholar] [CrossRef]
  23. Gursoy, D.; Li, Y.; Song, H. ChatGPT and the hospitality and tourism industry: An overview of current trends and future research directions. J. Hosp. Mark. Manag. 2023, 32, 579–592. [Google Scholar] [CrossRef]
  24. Mich, L.; Garigliano, R. ChatGPT for e-tourism: A technological perspective. Inf. Technol. Tour. 2023, 25, 1–12. [Google Scholar] [CrossRef]
  25. Demir, Ş.Ş.; Demir, M. Professionals’ perspectives on ChatGPT in the tourism industry: Does it inspire awe or concern? J. Tour. Theory Res. 2023, 9, 61–76. [Google Scholar] [CrossRef]
  26. Çolak, O. The Role of Generative Pre-trained Transformers (GPT) in Recreational Tourism: An Interview with ChatGPT. Spor Bilim. Araştırmaları Derg. 2023, 8, 733–748. [Google Scholar] [CrossRef]
  27. Ho, C.L.; Lin, M.H.; Chen, H.M. Web users’ behavioural patterns of tourism information search: From online to offline. Tour. Manag. 2012, 33, 1468–1482. [Google Scholar] [CrossRef]
  28. Snepenger, D.; Snepenger, M. Information Search by Pleasure Travelers. Encyclopedia of Hospitality and Tourism; Van Nostrand Reinhold: New York, NY, USA, 1993. [Google Scholar]
  29. Puhringer, S.; Taylor, A. A practitioner’s report on blogs as a potential source of destination marketing intelligence. J. Vacat. Mark. 2008, 14, 177–187. [Google Scholar] [CrossRef]
  30. Gretzel, U.; Sigala, M.; Xiang, Z.; Koo, C. Smart tourism: Foundations and developments. Electron. Mark. 2015, 25, 179–188. [Google Scholar] [CrossRef]
  31. Buhalis, D.; Law, R. Progress in information technology and tourism management: 20 years on and 10 years after the internet-the state of e-Tourism research. Tour. Manag. 2008, 29, 609–623. [Google Scholar] [CrossRef]
  32. Xiang, Z.; Wober, K.; Fesenmaier, D.R. Representation of the online tourism domain in search engines. J. Travel Res. 2008, 47, 137–150. [Google Scholar] [CrossRef]
  33. Palos-Sanchez, P.; Saura, J.R.; Velicia-Martin, F.; Cepeda-Carrion, G.A. Business model adoption based on tourism innovation: Applying a gratification theory to mobile applications. Eur. Res. Manag. Bus. Econ. 2021, 27, 100149. [Google Scholar] [CrossRef]
  34. Tavitiyaman, P.; Qu, H.; Tsang, W.L.; Lam, C.R. The influence of smart tourism applications on perceived destination image and behavioral intention: The moderating role of information search behavior. J. Hosp. Tour. Manag. 2021, 46, 476–487. [Google Scholar] [CrossRef]
  35. Fan, X.; Jiang, X.; Deng, N. Immersive technology: A meta-analysis of augmented/virtual reality applications and their impact on tourism experience. Tour. Manag. 2022, 91, 104534. [Google Scholar] [CrossRef]
  36. Huang, C.D.; Goo, J.; Nam, K.; Yoo, C.W. Smart tourism technologies in travel planning: The role of exploration and exploitation. Inf. Manag. 2017, 54, 757–770. [Google Scholar] [CrossRef]
  37. Buhalis, D. Technology in tourism-from information communication technologies to e-Tourism and smart tourism towards ambient intelligence tourism: A perspective article. Tour. Rev. 2019, 75, 267–272. [Google Scholar] [CrossRef]
  38. Mori, M. The Uncanny Valley: The Original Essay by Masahiro Mori; IEEE Spectrum: New York, NY, USA, 1970. [Google Scholar]
  39. Diel, A.; Weigelt, S.; MacDorman, K.F. A meta-analysis of the uncanny valley’s independent and dependent variables. ACM Trans. Hum. Robot. Interact. 2022, 11, 1–33. [Google Scholar] [CrossRef]
  40. Howard, M.C. Investigating the simulation elements of environment and control: Extending the Uncanny Valley Theory to simulation. Comput. Educ. 2017, 109, 216–232. [Google Scholar] [CrossRef]
  41. Saygin, A.P.; Chaminade, T.; Ishiguro, H.; Driver, J.; Frith, C. The thing that should not be: Predictive coding and the uncanny valley in perceiving human and humanoid robot actions. Soc. Cogn. Affect. Neurosci. 2012, 7, 413–422. [Google Scholar] [CrossRef] [PubMed]
  42. Kim, B.; Visser, E.D.; Philips, E. Two uncanny valleys: Re-evaluating the uncanny valley across the full spectrum of real-world human-like robots. Comput. Hum. Behav. 2022, 135, 107340. [Google Scholar] [CrossRef]
  43. Mori, M.; MacDorman, K.F.; Kageki, N. The uncanny valley. IEEE Robot. Autom. Mag. 2012, 19, 98–100. [Google Scholar] [CrossRef]
  44. Arsenyan, J.; Mirowska, A. Almost human? A comparative case study on the social media presence of virtual influencers. Int. J. Hum.-Comput. Stud. 2021, 155, 102694. [Google Scholar] [CrossRef]
  45. Shin, M.; Kim, S.J.; Biocca, F. The uncanny valley: No need for any further judgments when an avatar looks eerie. Comput. Hum. Behav. 2019, 94, 100–109. [Google Scholar] [CrossRef]
  46. Spiegelberg, H. The Phenomenological Movement, 3rd ed.; Martinus Nihoff: The Hague, The Netherlands, 1982. [Google Scholar]
  47. Giorgi, A. The theory, practice, and evaluation of the phenomenological method as a qualitative research procedure. J. Phenomenol. Psychol. 1997, 28, 235–260. [Google Scholar] [CrossRef]
  48. Van Manen, M. Researching Live Experience: Human Science for an Action Sensitive Pedagogy; The University of Western Ontario: London, ON, Canada, 1990. [Google Scholar]
  49. Creswell, J.W. Qualitative Inquiry & Research Design: Choosing Among Five Approaches; Sage: Thousand Oaks, CA, USA, 2013. [Google Scholar]
  50. Colaizzi, F.U. Psychological research as the phenomenogist views it. In Existential-Phenomenological Alternative for Psychology; Valle, R.S., King, M., Eds.; Oxford University Press: New York, NY, USA, 1978. [Google Scholar]
  51. Van Kaam, A. Existential Foundations of Psychology; Duquesne University Press: Pittsburgh, PA, USA, 1966. [Google Scholar]
  52. Polkinghorne, D.E. Phenomenological research methods. In Existential-Phenomenological Perspectives in Psychology; Valle, R.S., Halling, S., Eds.; Plenum Press: New York, NY, USA, 1989; pp. 41–60. [Google Scholar]
  53. Jin, J.H.; Lee, J.H. Impact of ChatGPT service quality on the intention of generation Z to continue using it for tourism information search. Asia Pac. Converg. Res. Interchange 2024, 10, 223–233. [Google Scholar] [CrossRef]
  54. Wong, I.A.; Lian, Q.L.; Sun, D. Autonomous travel decision-making: An early glimpse into ChatGPT and generative AI. J. Hosp. Tour. Manag. 2023, 56, 253–263. [Google Scholar] [CrossRef]
Sustainability 17 00355 g001
Figure 1. Giorgi’s 5-step procedure.
Figure 1. Giorgi’s 5-step procedure.
Sustainability 17 00355 g001
Table 1. In-Depth Interview Questionnaire.
Table 1. In-Depth Interview Questionnaire.
Questions
Human
likeness		1. How is ChatGPT similar to humans in searching for tourism information?
2. What are the advantages of ChatGPT in searching for tourism information?
3. What are the disadvantages of ChatGPT in searching for tourism information?
4. Do you think tourism information searched through ChatGPT is trustworthy?
5. Do you think tourism information searched through ChatGPT is useful?
Familiarity1. What positive emotions did you feel while searching for tourism information with it?
2. What negative emotions did you feel while searching for tourism information with it?
3. What is different from what you expected while searching for tourism information with it?
4. Compared to other tools for searching for tourism information (e.g., search portal sites, applications, travel agencies, acquaintances, books, etc.) has the familiarity for it increased? Why?
5. Would you like to talk to it again to search for tourism information? Why?
Table 2. Sample characteristics.
Table 2. Sample characteristics.
GenderAgeNationalityEducation
Participant AFemale25South KoreaHigh-school graduate
Participant BFemale20VietnamHigh-school graduate
Participant CMale27South KoreaGraduate degree
Participant DMale27JapanGraduate degree
Participant EFemale31NepalGraduate degree
Participant FMale33VietnamGraduate degree
Participant GFemale22UzbekistanHigh-school graduate
Participant HMale43ChinaGraduate degree
Participant IFemale23UzbekistanHigh-school graduate
Participant JMale31TurkeyHigh-school graduates
Participant KFemale30USAGraduate degree
Participant LFemale28ChinaGraduate degree
Table 3. Human Likeness.
Table 3. Human Likeness.
ThemesMeaning UnitsMeaningful Statements
1. Partial trust in tourism informationOnly famous tourist attractions are accurate“In famous places like Gyeongju, the information is relatively accurate because it has a lot of learned data”.
Providing text-based tourism information“Portal sites can provide photos and videos, but ChatGPT version 3.5 can’t provide that”.
Lack of up-to-date tourism information“Because ChatGPT is trained only with data up to September 2021, it is vulnerable to the latest information”.
Occasional hallucination“Have you ever heard of hallucination? There are cases that AI explains incorrect information as if it were correct”.
2. A personal AI travel agentProviding neat tourism information“ChatGPT neatly provides name, location, price, and benefits of the tourist attractions”.
Providing personalized tourism information“ChatGPT tells me things that don’t come up when I search on Naver, and provides it that suits me”.
The joy of conversation“I really like it, I can communicate with it like humans, unlike search portals, SNS, and other search tools”.
3. Satisfaction as a primary tool of searching tourism informationSearching for tourism information“It’s perfect for searching information when I don’t have a tourist attraction I want to visit or just know the name”.
Good accessibility and reduction of time“I like that I can get quickly the tourism information I want through ChatGPT rather than scrolling through blogs”.
Opportunity to expand view of tourism“I was surprised because when recommending tourist attractions, it mentioned libraries and museums that I have never thought of”.
4. An AI travel mate in all processes of tourismSuggesting a travel itinerary (before)“When I asked ChatGPT if I was going on a 3-day, 2-night trip from Seoul to Gyeongju, it gave me a realistic itinerary”.
Providing tourism information in real time (during)“I asked ChatGPT in front of Gyeongbokgung Palace and it told me about the history of it in real time”.
My friend who listens to my travel experience (after)“I told ChatGPT about my trip to Busan, and it was nice that ChatGPT sympathized with me and asked about more experiences”.
5. The disappointment of ChatGPT still feeling like a machineNot enough to lead the conversation“ChatGPT sometimes doesn’t lead the conversation”
Sometimes unable to finish sentences“Sometimes when I ask a difficult question, ChatGPT doesn’t finish the sentences properly”.
Stiff and mechanical speaking style“I thought the tone would be more friendly and familiar, but it was too stiff and mechanical”.
The effort of asking good questions“If ChatGPT doesn’t give me the answer I want, I have to go through the trouble of asking the question again”.
Low level of understanding the requirement“Like a travel agent, I hope that it can carefully understand my desires and wants”.
Table 4. Familiarity.
Table 4. Familiarity.
ThemesMeaning UnitsMeaningful Statements
6. ChatGPT about ambivalenceCoexistence of conveniences and inconveniences“It is true that ChatGPT provides convenience, but it also seems to provide inconvenience”.
Coexistence of expectations and disappointments“I had expectations that ChatGPT would develop further, but I was also disappointed that it had not yet caught up with humans”.
Vague fear of technological advancement“If technology develops further, ChatGPT could replace humans, right? It may have a positive impact, but I feel fear”.
7. Moments when ChatGPT is better than humansI can talk I want“I plan my trip early in the morning, even though I can’t talk with my friends, I can talk with ChatGPT whenever I want”.
Unnecessary emotional exhaustion“Actually, talking to a travel agent exhausts my emotions, but with ChatGPT, I can talk as I want comfortably…”.
8. The enjoyment in the process of learning about ChatGPTStart with curiosity“At first, I was very curious. I wondered if ChatGPT would feel like talking to a human, like what was shown in the media”.
Interested in the answer“How many people use ChatGPT? I heard that it answers by learning the contents of conversations with people, the answer was interesting”.
The fun and joy of asking questions“If I were to ask ChatGPT this, what would it say? It was really fun to think about it and ask questions”.
Feeling attracted to imperfection“Honestly, even humans can make mistakes, so I actually liked ChatGPT because it wasn’t perfect”.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Jin, J.-H.; Han, J.-S. A Phenomenological Study on the Experience of Searching for Tourism Information Following the Emergence of ChatGPT: Focused on the Uncanny Valley Theory. Sustainability 2025, 17, 355. https://doi.org/10.3390/su17010355

AMA Style

Jin J-H, Han J-S. A Phenomenological Study on the Experience of Searching for Tourism Information Following the Emergence of ChatGPT: Focused on the Uncanny Valley Theory. Sustainability. 2025; 17(1):355. https://doi.org/10.3390/su17010355

Chicago/Turabian Style

Jin, Jin-Hee, and Jin-Seok Han. 2025. "A Phenomenological Study on the Experience of Searching for Tourism Information Following the Emergence of ChatGPT: Focused on the Uncanny Valley Theory" Sustainability 17, no. 1: 355. https://doi.org/10.3390/su17010355

APA Style

Jin, J.-H., & Han, J.-S. (2025). A Phenomenological Study on the Experience of Searching for Tourism Information Following the Emergence of ChatGPT: Focused on the Uncanny Valley Theory. Sustainability, 17(1), 355. https://doi.org/10.3390/su17010355

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop