Trusting ChatGPT Usage in Personalized Travel Planning: The Moderating Role of Privacy and Data Security

Abstract

ChatGPT has become one source of personalized recommendations used by travelers in travel planning contexts. This research aims to examine the ChatGPT usage dimensions and its impact on travelers’ level of trust in ChatGPT recommendations and their behavioral intention of future use. It also tests the moderating effect of privacy and data security on the relationship between travelers’ trust and behavioural intention. Using PLS-SEM to analyze collected data of travelers from the UAE and Oman, and using the multi-group analysis technique, findings revealed a positive impact of ChatGPT usage dimensions on travelers’ trust and behavioral intention. Interestingly, findings revealed a negative moderating effect of privacy and data security on travelers’ trust in ChatGPT recommendations and behavioral intention of usage. Theoretical and practical contributions of the results are also offered.

Keywords

ChatGPT travel planning trust behavioral intention privacy data security

Introduction

The travel and tourism industry, like other industries, has witnessed a remarkable transformation due to the discovery of artificial intelligence (Zlatanov and Popesku, 2019). Xu et al.’s. (2024) findings confirmed that many prominent applications have emerged that have helped the travel industry, the most prominent of which is the provision of personalized travel suggestions through conversational agents such as ChatGPT. Based on Kannan (2024), this could considerably enhance the experiences of travelers. In this vein, when AI-driven services customize suggestions based on individual tastes, they can revolutionize trip planning and enjoyment. However, it is essential to address travellers’ concerns about data security and privacy, along with other factors that impact their decisions.

Previous studies have then stressed that the perceived relevance, credibility, usefulness, and intelligence of the ChatGPT-based personalized travel recommendations have a considerable influence on travellers’ trust and their behavioural intentions are positively influenced (Ali et al., 2023; Li and Lee, 2024; Shin and Kang, 2023). Further, prior research has elaborated on the importance of comprehending how artificial intelligence (AI) works in providing the right clues in the field of traveling services, which entails the tendency, validity, usefulness, and enhanced complexity of the recommendations (Kim et al., 2024a). Pioneering studies have been devoted to these facets by Ali et al. (2023), which empirically examined AI impact on the travellers’ trust and post- behavioural intentions. The practical implications of the affordance-actualization theory and the ad credibility model were employed to gain useful understanding of travellers’ engagement with AI-related services (Ali et al., 2023; Yhee et al., 2024).

Although past research investigations have established a strong basis, there are still significant gaps that have not been addressed. Prior research mostly overlooked the crucial elements of privacy and data security, which are of utmost importance to numerous people when engaging with AI systems (Ioannou et al., 2021). Moreover, the comparison investigation trailed off in merely one cultural environment and thus a significant lack of understanding of the possible differences of these components in other areas emerged. Therefore, the main goal of the present research is to redefine these shortcomings by including privacy and data security as a new variable in the conceptual model. Given the growing importance of these factors, the research objective can be defined more precisely as comprehending the effects of these factors on trust and behavioural intentions of the travellers who are using ChatGPT for travelling. Furthermore, the scope of the study expands the geographical framework by comparing the research findings from the United Arab Emirates and Oman. This approach provides a more thorough cultural viewpoint on the researched subject. By expanding upon the theoretical framework put forward by Ali et al. (2023), we want to provide a more comprehensive understanding of the elements that influence travellers’ confidence in utilizing AI services.

Therefore, the primary objectives of our study can be categorized into three distinct parts. The main objective of our study is to assess how the relationship between privacy and data security, as a moderating factor, influences the level of confidence and behavioural intentions of travellers towards ChatGPT’s tailored travel recommendations. Moreover, we examine and compare the aforementioned procedures in two culturally distinct regions, specifically the United Arab Emirates and Oman, to determine any significant differences. In addition, offering concrete practical uses and suggestions for future research, so making a significant contribution to the continuing discourse on artificial intelligence in the tourism industry. Striving to accomplish these goals will aim to offer substantial knowledge for both scholars and professionals in the field. This will strengthen the theoretical basis and provide practical support for the advancement of more dependable AI-powered travel services.

This article is arranged as follows: it underlines the importance of scientific research in this important industry in both the United Arab Emirates and Oman and offers an outline of the tourist scenario in both countries. Subsequently, the conceptual framework and suppositions are formulated, followed by a summary of the investigative techniques and results. Ultimately, the findings are examined, and definitive inferences are made.

Literature review and hypotheses development

The affordance-actualization theory and the ad credibility model were found to be useful in understanding the travellers’ interactions with AI-related services. According to the affordance-actualization theory that addresses the interaction between the user and specific characteristics of the technology, it becomes easier to explain how much attention the travellers pay on the capabilities and potentials of the AI services in travel activities (Wang et al., 2018). Affordance is described in this theory as users’ perceived usefulness of technology where the affordance is instantiated through interaction (Ali et al., 2023). At the same time, the ad credibility model focuses on the credibility of advertisements and reveals how the believability of recommendations affects the AI travelling acceptance and utilization (Choi and Rifon, 2002; Yhee et al., 2024).

Therefore, the integration of these models would also help extend the knowledge of the specialists regarding the impact of the AI travel recommendations such as the suggestions given by ChatGPT for the trust of the travellers and their decision-making ability (Ali et al., 2023). This two-part perspective, therefore, deals with the technological utilization of AI while also incorporating a concern for credibility, which provides a sound method of analyzing traveller interaction in today’s digital environment. These models have been described on the basis of recent studies, which assert that with the advancement of AI, its believability factor and its so-called ‘usefulness’ factor will have the major influence controlling the customers’ behaviour and thereby improving their travel experience (Ioannou et al., 2021; Kim et al., 2024b).

In accordance with the findings of Kakhki et al. (2024), the affordance-actualization theory provides a complete framework for understanding how travellers interact with ChatGPT by recognizing and utilizing its features, which include intelligence, usefulness, and relevance. The realization of affordances, which are defined as the features of technology that may be put into action, occurs when individuals recognize and engage with the functions of technology. Travelers utilize ChatGPT’s customized advice to optimize their travel planning experience, hence cultivating trust. Similarly, the ad credibility model underscores the importance of trustworthiness and reliability in user interactions with AI (Elbaz et al., 2024). Therefore, this model highlights how credible recommendations, characterized by accuracy and transparency, increase users’ confidence and likelihood of adoption. Together, these theories complement each other, providing a holistic perspective on the factors influencing trust and behavioral intentions toward ChatGPT in travel planning.

According to the study, passengers’ trust should be raised by perceived value, credibility, utility and intelligence of travel recommendations. This trust, consequently, influences the behavioral intentions of such people, thereby increasing their likelihood of acting on suggestions like booking places of interest. Kannan (2024) and Camilleri (2024) confirm this view arguing that such trust in AI recommendations brings better engagement and satisfaction. In addition, privacy and data security are investigated as a moderator variable of the model. The main idea is that people are willing to read and stick to recommendations and ensure that personal data is safe. Further supporting this trust is the ability of clear information presented concerning the privacy measures, which in turn serves to improve the influence on behavioral intentions of the personalization of recommendation systems and the outcomes which are presented in the Figure 1.

Figure 1.

The study theoretical framework.

Value and trust

Relevance, or value, is identified as the competence of a generative AI to produce answers or perform action that is highly appropriate and useful relative to the user’s questions and the context of the dialogue. A pertinent answer is one that clearly answers the user’s question or enquiry and grants utility and accuracy to the user or an organization (Følstad and Taylor, 2021). In this vein, it is very important to notice how recommendations given to travellers are very helpful in gaining their trust. Nowadays, people desire to be treated differently, and therefore recommendations regarding traveller’s preferences can make them feel unique. This translates into the fact that recommendations that are made to travellers are trusted since the other person can be relied on to have a self-interest in making sure that the client gets the best. Camilleri (2024) evidence supports the notion that people will be pleased and loyal if given specific ideas since generalized tips are less practical and satisfy needs better. Furthermore, when recommendations are accepted as suitable and appropriate to the travellers, they will have more trust in the service provider (Shi et al., 2021). This is because customized solutions are usually viewed as effectively suggesting that the supplier has something special done for their clients. Chi et al. (2022) goes further to explain that since the recommendations are generated using some complex algorithms, there is an added confirmation of the fact that the recommendations are accurate, thus building confidence. Hence, it can be evident that the concept of adaptive travel offering cannot be overemphasized when it comes to shaping interpersonal patrons’ trust, thus enhancing travelers’ loyalty and satisfaction. Therefore, we propose that:

H1. The AI value has a significant and positive impact on travelers’ trust.

Credibility and trust

Another key determinant of the travellers’ trust is the credibility of the recommendations that are given based on the travellers’ personalities. Fundamentally, the credibility of generative AI like ChatGPT’s tailored travel recommendations is based on trust, accuracy, dependability, openness, and proven knowledge, hence increasing users’ likelihood of depending on and acting upon the recommendations given (Ali et al., 2023; Kim et al., 2024b). In this regard, if recommended suggestions are considered trustworthy, it is more probable that travellers will follow those recommendations, which in turn affects their satisfaction and loyalty. This is evident in current research. According to Chi et al. (2022), when travellers are convinced that the recommendations made for them are accurate, they are more likely to take the recommendations thus improving their travel experience. This trust is usually created based on the continuity of the correct and credible prediction of the suggestions given. Regarding this aspect, algorithms and AI can contribute to the increase of the credibility of such recommendations. Through inferring the traveller’s needs and the behaviours they have exhibited in previous decisions, these technologies present recommendations that are quite relevant, which further solidifies trust Wang (2024). When travellers grasp that the recommendations underneath are based on data and include personalized information, the information’s credibility rises. Along with this line of thought, Kim et al. (2024b) supplement that it is also important to be clear on how such recommendations are arrived at also contributes to the credibility of the recommendations provided. Business travellers care about the background of suggestions they use, preferring to follow those with clear stated reasons, making them feel safe and content with their choosing of travel services. Thus,

H2. The AI credibility has a significant and positive impact on travelers’ trust.

The perceived usefulness and trust

Shi et al. (2021) and Li and Lee (2024) found that travellers who receive recommendations that are fit to their individual interests are more satisfied and trusting, because those travellers’ needs are perceived to be acknowledged and addressed. However, the prospect and preciseness of such recommendations are very significant in building trust. Kumar et al. (2024) discovered that it is easier for travellers to take action on recommended information especially where the information provides recency and affinity based on travellers past behaviour and interests. Also, the element of utilizing data analytics and AI in the provision of the recommendations adds to their value. According to Kim et al. (2021), with the precision of the recommended information improving the reliability of the system, the traveller’s confidence in arriving at the right decision is enhanced thereby improving the joy felt when travelling. Therefore, we hypothesize the following:

H3. The perceived usefulness has a significant and positive impact on travelers’ trust.

The intelligence of AI and trust

Smart suggestions are based on the use of state-of-the-art computations and statistical analysis to predict preference and behaviour of each and every traveller. Recent investigations support this effect, as Hamid et al. (2021) pointed out that intelligence plays a role in travellers’ decision-making processes when it comes to recommendations since such suggestions foster travellers’ sense of relevance and recognition. Furthermore, accuracy and relevance delivered by intelligent systems are one of the main factors that influence trust. Elbaz et al. (2021) have noted that the travellers’ trust is higher if the recommendations are in line with the travellers’ previous actions and preferences. This intelligence makes the suggestions more relevant and, therefore, more useful and credible to the recipient. Additionally, the manner in which such intelligent recommendations are produced can also be made clear for improving credibility. Kim et al. (2021) observe that any time travellers have the consciousness of how these recommendations work with the help of AI and data analytics, the trust in such recommendations improves. Therefore, the intelligence incorporated in recommended places to visit by travel partners has a positive effect on travellers’ trust, hence, improving their flow of travel plans. Hence:

H4. The perceived usefulness has a significant and positive impact on travelers’ trust.

Trust and behavioral intentions

Trust is an indispensable component of the formation of a traveller’s perception and decision-making process regarding the proposed recommendations. Each time travellers result in the customized recommendations provided by ChatGPT, there is a positive behavioural disposition. Recent studies support this impact. Based on the information provided by Ma et al. (2024), the user’s trust in AI’s recommendation directly correlates with the probability of making a booking decision with rather high figures. Hence, the quality of the decisions that ChatGPT is offering to its users is the key to establishing this trust. Filieri et al. (2021) observed that the guest who receives accurate recommendation feels more confident and, therefore, makes a greater intent to book and travel. In addition, the conversational style of ChatGPT can enhance this trust as the model will be able to respond immediately and depending on the client’s preference. Furthermore, since the usual logical operations which comprise, AI are transparent, the reliability of such recommendations is high. According to Wang (2024), users feel more committed to the fact that the suggestions given by ChatGPT have been derived from data hence have the confidence to undertake the suggestions hence making the ChatGPT more engaging and satisfying to the end users. Thus, we propose that:

H5. Travelers’ trust in AI such as ChatGPT has a positive influence on travelers’ behavioral intentions.

The moderating role of privacy and data security

Personal travel recommendation entails the offering of travel advice based on traveler’s data and there is nothing as crucial as the level of trust users have for their data to be used in the process. For instance, Kannan (2024) suggested that the interaction with personalized AI-generated recommendations is most effective when the travellers are confident that their data is safe; this shows that data safety influences travellers’ compliance with the recommendations of the AI systems in question. Moreover, the impact of privacy’s concerns on trust, in turn, positively or negatively, influences the link between trust and behavioural intention. Camilleri (2024) noted that it increases the rate of travellers’ behavioural engagement, such as booking the recommended travel options, once they are convinced that their privacy is protected. On the other hand, any doubts about the protection of users’ data undermine this trust and reduce the effectiveness of personalized offers in terms of influencing intentions to behave accordingly. Also, the openness of data processing increases the level of trust. It was found by Jeong and Shin (2020) that the clarity concerning the privacy policy and measures taken to enhance data security positively impacts traveller’s behavioural intentions. Hence, strong privacy and data protection are necessary to regulate the roles and opportunities of trust and behavioural intention corresponding to ChatGPT’s travel advice among travellers. On the other hand, Parasuraman and Colby (2015) found that AI privacy and data safety have a negative moderation influence in the link between trust and behavioural intention to use AI in travel planning. This may result because of the loss, misuse, and lack of clarity about data management by AI tools, thereby causing travellers’ anxiety (Ali et al., 2023). The misuse or unauthorized access to personal data. In this vein, privacy and data safety play a significant negative role in determining the willingness to use technology. Thus:

H5. Privacy and data security moderate the relationship between travellers’ trust and behavioural intentions towards ChatGPT’s tailored travel recommendations.

Research method

The present study employs the quantitative method to test five hypotheses that conceptualize the relationships between ChatGPT adoption factors (i.e., relevance, credibility, usefulness, and intelligence) and travellers’ trust and behavioural intention to use ChatGPT travel recommendations considering the moderation role of privacy and data security (Figure 1). The Study uses an e-survey to collect data from travellers using ChatGPT for travel planning purposes in the United Arab Emirates and Oman. The survey is mainly adapted from Ali et al. (2023) and Hassani and Silva (2023) (Appendix 1). Based on a five-point Likert scale, four items were used to measure ChatGPT recommendation relevance, four items for its credibility, four items for usefulness, five items for intelligence, three for trust, three for behavioural intention, and four for privacy and data security. A filter question was used to identify travellers who use ChatGPT for their travel planning in the UAE and Oman. A convenient sample of travellers in the UAE and Oman was accessed for data collection purposes between May and September 2024. The form was piloted on 50 travellers and the corrected item-total-correlation showed a rigorous construct validity. Format and rewording of some terminologies are considered in the final version. 255 and 251 valid responses were collected from the UAE and Oman travellers respectively. The widely used technique in tourism research, PLS-SEM (Abou-Shouk et al., 2021), was used for data analysis and hypothesis testing purposes. The Measurement scale was validated against convergent and discriminant validity and reliability criteria. The structural model, then, is run via the multi-group analysis technique to reveal the results of the two samples (i.e., UAE and Oman).

The UAE and Oman contexts

The purpose of this research is to improve the understanding of the effects that AI has on tourism by studying its effects on travellers’ trust and their intentions to engage in tourism activity, considering aspects that relate to privacy and security of data. These insights will be useful for governmental authorities and other participants in the UAE and Omani markets, as they will contribute to the development of more reliable and attractive services promoted by AI technologies in the tourism sphere.

The UAE is the top country in receiving tourists in the Arab region and is ranked as a top scorer in ICT readiness in 2019 (Abou-Shouk et al., 2024). The country has transformed the tourism industry into totally digitalized and to benefit from modern technologies to improve travel experience and position the UAE as a leader in using technology in the innovative tourism sector. The Oman country is situated on the southeastern side of the Arabian Peninsula; it has a rich cultural background and various landscapes, and therefore the country is very attractive for tourist (Elbaz et al., 2023; Salem et al., 2023). Furthermore, tourism has also grown substantially in Oman due to the governmental plan called Vision 2040, which aims at Oman diversification from the oil-exporting nation (Alkathiri et al., 2021). Currently, Oman is incorporating the past cultural feel with advanced technology like AI-based services for an enhanced tourist experience (Nazir et al., 2023). Past research has noted the need to embrace the use of AI in Oman’s tourist industry to meet the needs of current generation travellers, not forgetting the issue of data protection and security (Ramanathan and Meyyappan, 2019). In the same vein, the study by Alsahafi et al. (2023) shows that Oman can improve visitors’ happiness and build commitment among them by applying artificial intelligence to delivering relevant travel recommendations.

Findings

Sample profile

Results revealed that 278 (55%) of respondents are males versus 228 (45%) of females. 115 (22.7%) of respondents are aged 18–25 years old, 144 (28.5%) are aged 26–35 years, 124 (24.6%) are aged 36–45 years, 77 (15.3%) are aged 46–55 years, and (45) 8.9% are aged more than 55 years old. 299 (59%) of respondents are married, 101 (20%) are single, 106 (21%) are others (i.e., widows, divorced, separated).

Measurement model

Table 1 shows the validity and reliability of the measurement model. Loading of the scale items are all above 0.7. Cronbach’s alpha and composite reliability for all constructs are greater than 0.7 showing a reliable scale. VIF values of all constructs are less than 3.3 reflecting non-critical problems of multicollinearity. Average full collinearity VIF is 3.06 ensuing the absence of common method biased. Harmon’s one-factor test was also conducted and revealed variance value of 48.19% which confirms the absence of common method bias. AVEs values for all constructs are above the cut-off value of 0.5 meaning that convergent validity of the scale is evident.

Table 1.

Scale validity and reliability.

Constructs
Relevance
REL1	REL2	REL3	REL4	α	CR	AVE	VIF
0.764	0.853	0.861	0.837	0.810	0.898	0.688	2.92
Credibility
CRE1	CRE2	CRE3	CRE4	α	CR	AVE	VIF
0.835	0.853	0.834	0.873	0.889	0.912	0.721	3.02
Usefulness
USF1	USF2	USF3	USF4	α	CR	AVE	VIF
0.839	0.820	0.847	0.824	0.878	0.900	0.693	3.19
Intelligence
INT1	INT2	INT3	INT4	INT5	α	CR	AVE	VIF
0.820	0.768	0.753	0.800	0.834	0.855	0.896	0.633	3.29
Trust
TRU1	TRU2	TRU3	α	CR	AVE	VIF
0.877	0.866	0.870	0.896	0.904	0.759	3.04
Behavioural intentions
BEI1	BEI2	BEI3	α	CR	AVE	VIF
0.854	0.857	0.852	0.865	0.890	0.730	3.01
Privacy and data security
PDS1	PDS2	PDS3	PDS4	α	CR	AVE	VIF
0.881	0.870	0.856	0.844	0.893	0.921	0.745	2.98

Table 2 Depicts that square roots of AVEs are greater than correlations among constructs are greater referring to the existence of construct validity of the scale.

Table 2.

Scale discriminant validity.

Constructs	REL	CRE	USF	ITL	TRS	BIN	PRS
Relevance (REL)	(0.829)
Credibility (CRE)	0.752	(0.849)
Usefulness (USF)	0.803	0.715	(0.832)
Intelligence (ITL)	0.725	0.720	0.735	(0.769)
Trust (TRS)	0.815	0.804	0.787	0.703	(0.871)
Behavioural intention (BIN)	0.716	0.789	0.796	0.719	0.746	(0.854)
Privacy and security (PRS)	0.720	0.821	0.805	0.735	0.790	0.827	(0.863)

Table 3 Further confirms the existence of construct validity where HTMT2 ratios are less than the threshold of 0.85.

Table 3.

HTMT2 ratios.

Constructs	REL^a	CRE	USF	ITL	TRS	BIN
Credibility (CRE)	0.758
Usefulness (USF)	0.784	0.803
Intelligence (ITL)	0.795	0.841	0.821
Trust (TRS)	0.810	0.830	0.809	0.843
Behavioural intention (BIN)	0.768	0.835	0.838	0.832	0.818
Privacy and security (PRS)	0.828	0.842	0.845	0.824	0.831	0.839

^aREL: relevance.

Structural model

Figure 2 illustrates that traveller’s trust in ChatGPT is predicted by its relevance (β = 0.24, p < .01, and H1 is accepted), credibility (β = 0.22, p < .01, and H2 is accepted), usefulness (β = 0.25, p < .01, and H3 is accepted), and intelligence (β = 0.20, p < .01, and H4 is accepted). These four predictors explain 75% of traveller’s trust in using ChatGPT in travel planning concerns. These findings explain that travellers believe in the benefits of ChatGPT personalized recommendation for travel, they believe in its credibility, efficiency and usefulness, and its intelligence, knowledge, and competence.

Figure 2.

Structural model of ChatGPT adoption and travel behavioural intentions.

In turn, traveller’s trust in ChatGPT personalized recommendations for travel is affecting their behavioural intentions of using ChatGPT for travel purposes and recommending it to others (β = 0.58, p < .01, and H5 is accepted). Traveller’s trust explains 60% of the variance in traveller’s behavioural intention of adopting ChatGPT for travel recommendation aspects. ChatGPT relevance, credibility, usefulness, and intelligence are indirectly affecting ChatGPT behavioural intention via traveller’s trust construct leading to the partial mediation of traveller’s trust. As for the sum of indirect effects, ChatGPT’s behavioural intention is indirectly affected by ChatGPT’s relevance (β = 0.14, p < .01), credibility (β = 0.13, p < .01), usefulness (β = 0.15, p < .01), and intelligence (β = 0.12, p < .01). the effect size of total effects is medium.

Furthermore, it is revealed that data privacy and security is negatively moderating the relationship between traveller’s trust and traveller’s behavioural intention of adopting and recommending ChatGPT for travel contexts (β = −0.26, p < .01 and H6 is supported). This result is interesting as despite the positive perceptions of travellers about ChatGPT’ personalized recommendations for travel and their trust in it, they have a critical concern of data privacy and security which in turn decrease their behavioural intention to adopt it for travel purposes.

Multi-group analysis

Comparing the perceptions of travellers from the UAE and Oman, Table 4 shows to the differences between paths of effect in the research model. Results highlight the significant differences between the perceptions of the two samples. Despite the substantive effect of ChatGPT’s relevance, credibility, usefulness, and intelligence on traveller’s trust, and travellers’ trust on behavioural intentions, results highlighted the significant differences in effect between the two samples. The effect of ChatGPT’s trust on behavioural intention in the UAE’s sample is greater than its peer in the Oman’s sample. While Oman’s sample is greater in the effect of perceptions of ChatGPT’s negative moderation effect of data privacy and security. These differences could be explained because the UAE is a top scorer of ICT and has transferred the tourism industry to be totally digitized and travellers benefit from the AI technologies and have usage experience of these technologies compared to Oman travellers, and therefore they have their own judgment and perceptions on ChatGPT’s trust that are different from Oman travellers. In addition, when it comes to data privacy people behave differently and that explains the significant negative effect between UAE and Oman travellers.

Table 4.

Multi-group analysis according to travelers’ country (UAE-Oman).

Samples	UAE-Oman
Constructs	Paths	Std. Er.	Sig.
ChatGPT’s relevance → travelers’ trust	0.29–0.21	0.086	0.198
ChatGPT’s credibility → travelers’ trust	0.18–0.22	0.087	0.318
ChatGPT’s usefulness → travelers’ trust	0.19–0.30	0.086	0.104
ChatGPT’s intelligence → travelers’ trust	0.27–0.18	0.086	0.128
Travelers’ trust → travelers’ behavioural intention	0.51–0.24	0.084	<0.001
Data privacy and security moderation	(0.25)–(0.62)	0.083	<0.001

Discussion of findings

The present study aims to examine the influence of ChatGPT adoption factors on the behavioural intention of tourists’ usage of ChatGPT for travel recommendations considering the trust of travellers as a mediation and data privacy and security as a moderator of the relationship between travellers’ trust and ChatGPT behavioural intention. Based on data collected from the UAE and Oman, travellers found that ChatGPT is relevant, credible, useful, and intelligent tool for travel personalized recommendations and therefore they trust it when revealing travel recommendations. Furthermore, it is found that the trust of travellers in ChatGPT recommendations is affecting their behavioural intention of usage continuity in travel planning. However, the interesting result is that privacy and data security of ChatGPT is negatively moderating the relationship between travellers’ trust and their behavioural intention of using ChatGPT for travel services.

The findings revealed a positive effect of four factors of ChatGPT usage on travellers’ trust in travel recommendations, namely: relevance, credibility, usefulness, and intelligence. As for relevance, travellers have now different tastes and desires and when they feel that personalized travel recommendations are made specifically for them and find it appropriate for their preferences, then they think that these recommendations are of value, and relate to their requirements. Having valuable and useful personalized travel recommendations from ChatGPT leads to travellers’ trust in the specific travel recommendations of ChatGPT. This finding is in line with previous investigations by Wang (2024) and Ali et al. (2023) who mentioned that providing refresh and appropriate ideas leads to travellers’ confidence and trust in ChatGPT recommendations.

In addition, findings revealed that travellers perceive ChatGPT recommendations as credible, accurate, and convincing. When travellers find the ChatGPT tailored and exclusive recommendations are relevant and accurate, this increases their trust in it and improves their travel experience. This will definitely increase their dependency on the ChatGPT trustworthy travel recommendations and increase their satisfaction and loyalty. This finding is consistent with Wang (2024) who claim that having accurate and credible recommendations from AI generative technologies increases travellers’ trust in these recommendations and increases the probable continuity of usage in the future. Ali et al. (2023) also revealed that credibility of travel recommendations is a main determinant of ChatGPT traveller’s trust.

Results have also revealed that the usefulness of ChatGPT travel recommendations is another key determinant of travellers’ trust in it. It is found that travellers believe that ChatGPT’s personalized recommendations are useful for their travel planning, improving planning efficiency and performance, resulting in overall valuable travel planning. Perceiving ChatGPT as useful tool for travel recommendations and trust it increases their probable continuity to depend on these suggestions and feel happy and satisfied. This result is concurrent with Kim et al. (2024a) who believes that receiving recommendations as fit and useful for individual interests feel more satisfied and trusting and increases their reliability and confidence in the AI tool (Camilleri, 2024).

Furthermore, respondent travellers described ChatGPT’s personalized recommendations as competent, knowledgeable, responsible, and sensible, which can be summarized in one concept, intelligence. The appropriateness of generated travel recommendations, its credibility, and usefulness leads travellers to describe ChatGPT as intelligent or smart. This intelligence increases traveller’s' trust in the ChatGPT and its personalized suggestions and improves their satisfaction and loyalty. This finding is in line with Chi et al. (2022) and Kannan (2024) who think that intelligence fosters travellers’ trust, and this trust increases when recommendations fit travellers’ preferences and desires.

Following ChatGPT's trip planning advice helps travellers maintain interacting with ChatGPT to get customized recommendations; they also communicate good things about ChatGPT's recommendations to others. Ma et al. (2024) reported a link between trust in AI's recommendations on booking decisions and. In addition, trusting ChatGPT recommendations will lead travellers to be committed to use these suggestions and encourage others to use ChatGPT as perceived by Camilleri (2024). This result also explains that trust is a crucial moderation between travellers’ adoption of ChatGPT and its future behavioural intention of usage. Trust is a factor that maximizes travellers’ intention to be committed to ChatGPT recommendations.

As for the moderating effect of privacy and data security, findings revealed a negative moderating effect on the relationship between travellers’ trust and their behavioural intention to continue use ChatGPT for travel planning and recommendations. Privacy and data security imply the perception of travellers on security concerns, safety of personal data and authorized access to personal data. Despite the credibility and usefulness of travel recommendations generated by ChatGPT, travellers become sensitive when it comes to the use and access of their personal data. Any misuse, loss, or unauthorized access to their personal data may result in termination of using ChatGPT recommendations and increases their anxiety. This finding is consistent with Parasuraman and Colby (2015) and Ali et al. (2023) who think that privacy and data protection play a crucial role in using AI tools and that travellers feel anxious with any misuse of their personal data.

Conclusion

This research examines the effect of ChatGPT adoption factors (i.e., relevance, credibility, usefulness, and intelligence) on travellers’ trust in ChatGPT travel recommendations and their behavioural intention for future use. It also investigates the moderating effect of privacy and data security on the relationship between traveller’s trust and behavioural intention. Based on data collected form travellers using ChatGPT for travel planning in the UAE and Oman and using PLS-SEM, findings revealed that the adoption factors of ChatGPT are positively affecting travellers’ trust in the personalized recommendations for travel generated by ChatGPT. In addition, travellers’ trust is mediating the relationship between adoption factors and travellers’ behavioural intention of ChatGPT future use for travel planning. However, privacy and data security were found to have a negative moderating effect on the relationship between traveller’s trust and their behavioural intention of future usage of ChatGPT for travel planning.

Theoretical implications

This research contributes to theory and adds to the extant knowledge of using ChatGPT for travel planning in the Arab country’s context. It compares the perceptions of travellers in the United Arab Emirates and Oman, two countries that emphasize the use of AI technologies in travel and tourism industry to improve travel experiences. In addition, this study tests two different types of effect, a mediation and a moderation effect in one robust model. It reveals the mediation effect of traveller’s trust in ChatGPT travel recommendations and behavioural intentions of travellers to use it in the future and the negative moderation effect of privacy and data security on the relationship between traveller’s trust and behavioural intention of use. Developing the research model of Ali et al. (2023), it examines the moderating effect of privacy and data security which despite the trust of travellers in ChatGPT travel recommendations and their behavioural intention to use it in the future, they express their worries and anxiety of misuse, loss, or unauthorized access to their personal data while using ChatGPT. This study responds to Ali et al. (2023) for investigating the effect of privacy and data security on traveller’s trust. Therefore, integrating ad credibility model with affordance-actualization theory provides a two-view point for analyzing ChatGPT acceptance. Although the affordance-actualization hypothesis stresses the functional and interactive aspects of artificial intelligence (e.g., relevance, intelligence), the ad credibility model underlines the crucial part of trustworthiness in affecting user behavior. These models taken together support our hypothesis by offering complete knowledge of the processes behind travellers’ confidence and behavioral intentions.

Practical implications

Contributing to practice, the findings of this research concluded that relevance in a key predictor of traveller’s trust in ChatGPT recommendations. This highlights the importance of meeting travellers’ needs and preferences of their travel plans. Service providers who offer ChatGPT to generate travel recommendations should provide relevant and appropriate travel options and recommendations and should be based on travel history of travellers and their travel preferences. Furthermore, another predictor of traveller’s trust revealed in findings is credibility. This suggests that when travellers get valuable and appropriate trave recommendation from ChatGPT, they will trust it and will be committed to using it in the future. This is another burden for travel providers who should ensure credible and trustworthy provided recommendations. The third factor is usefulness. Having valuable and useful recommendations of travel by ChatGPT will increase the likelihood of traveller’s trust in its recommendations. This again can be ensured by fetching travel history, experiences, and preferences of travellers. The fourth factor is intelligence. Travel service providers should ensure competent, knowledgeable, responsible, and sensible recommendations generated about their travel services in order to increase the level of trust in travel planning and the commitment of travellers to use it in the future.

One critical aspect highlighted by findings is the privacy and data protection of travellers. This critical determinant of travellers’ trust and intention to use ChatGPT recommendations must be cautiously considered by ChatGPT, and a clear policy of personal data use and access should be provided in order to decrease the traveller’s anxiety and worries of using such AI tools in travel planning. Overall, these results highlight the need to build ChatGPT interfaces and algorithms maximizing user affordances, that is, suggestions based on context, clarity, and personalization. Travel companies might focus on enhancing user interactions with ChatGPT by best matching its characteristics to travel needs using the affordance-actualization theory. Moreover, ad credibility model would help to ensure the dependability and openness of AI-generated ideas, which are fundamental in creating and sustaining customer confidence.

Limitations

Further research is required to test the credibility and usefulness of travel recommendations by travellers from different cultures. Demographic data including gender, age, education, and technology usage experience level should be incorporated in the research model to examine its effect on the traveller’s trust and behavioural intentions. Qualitative research is encouraged to examine the appropriateness and relevance of generated travel recommendations and its role in increasing the trust level among ChatGPT users.

Supplemental Material

Supplemental Material - Reimagining AI-powered travel: A cross-country investigation of the Privacy and security impact on ChatGPT’s trust in personalized tourism experiences

Supplemental Material for Trusting ChatGPT Usage in Personalized Travel Planning: The Moderating Role of Privacy and Data Security by Mohamed Abou-Shouk, Ahmed Mohamed Elbaz, Saleh Muhammad Zeki Mahmood Al-Leheabi, EmadEddin AbuElEnain and Maha Misbah Shabana in Tourism and Hospitality Research.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Mohamed Abou-Shouk

EmadEddin AbuElEnain

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Mohamed Abou-Shouk holds a PhD in Tourism and Hospitality from Plymouth University, UK. He is currently an Associate Professor at the College of Arts, Sciences, and Information Technology, University of Al Dhaid, UAE. His research interests are technology adoption behaviour in tourism and hospitality SMEs, e-tourism, e-commerce and e-marketing in tourism and hospitality, entrepreneurship in tourism and hospitality, and structural equation modelling-based quantitative research in tourism. His work has been published in a range of reputable journals including Tourism Management, International Journal of Contemporary Hospitality Management, Current Issues in Tourism, International Journal of Tourism Research, Journal of Travel and Tourism Marketing, and Journal of Hospitality and Tourism Technology and other journals.

Ahmed Mohamed Elbaz’s research interests include tourism actor networks, tourism leadership, knowledge transfer, environmental protection, poverty reduction, sustainable tourism development, governance in developing countries, PLS-SEM analysis, and fsQCA. I have published multiple papers in reputed journals indexed in WoS and Scopus (e.g., Tourism Management; Annals of Tourism Research; Journal of Knowledge Management; International Journal of Tourism Research; Tourism Review, etc.).

Saleh Muhammad Zeki Mahmood Al-Leheabi holds a PhD in History and Islamic Civilization. He is currently an associate professor at the University of Sharjah, College of Arts, Humanities and Social Sciences. His research interests include historical manuscripts, heritage and museum studies.

EmadEddin AbuElEnain (Tourism Management and Economics Associate Professor, UKF), used to be the Dean & Head of Dept. - College of Tourism Sciences, UMS, Dubai, UAE, before joining the UoS and UKF. During his Senior academic appointments in the UAE, Dr. AbuElEnain was the chair and member of various universities committees and chaired The Modern Sciences Institution for Development and Training. Dr. EmadEddin has supported and coordinated various research and innovations projects in the fields of sustainable development, innovation and creativity in tourism. https://orcid.org/0000-0002-3346-136X

Maha Misbah Shabana is associate professor at Department of Business Administration, Faculty of Commerce, Damietta University, Egypt. She holds a PhD of business management from University of Naples Federico II, Italy. Her research interests include e-marketing; consumer behaviors; e-booking in hotels; entrepreneurship, and new trends in green tourism SMEs, e-commerce; blended learning adoption in high education enterprises and structural equation modeling analysis.

References

Abou-Shouk

Elbaz

Maher

(2021) Breaking the silence of travel agency employees: the moderating role of gender. Tourism and Hospitality Research 21(4): 487–500. DOI: 10.1177/1467358420988048.

Abou-Shouk

Zouair

Abdelhakim

, et al. (2024) The effect of immersive technologies on tourist satisfaction and loyalty: the mediating role of customer engagement and customer perceived value. International Journal of Contemporary Hospitality Management 36: 3587–3606. DOI: 10.1108/IJCHM-09-2023-1496.

Ali

Yasar

Ali

, et al. (2023) Antecedents and consequences of travelers' trust towards personalized travel recommendations offered by ChatGPT. International Journal of Hospitality Management 114: 103588.

Alkathiri

Elbaz

Shah

, et al. (2021) Predictors and outcomes of successful localization in the aviation industry: the case of Oman. Advances in Hospitality and Tourism Research 9(2): 418–443.

Alsahafi

Alzahrani

Mehmood

(2023) Smarter sustainable tourism: data-driven multi-perspective parameter discovery for autonomous design and operations. Sustainability 15(5): 4166.

Camilleri

(2024) Factors affecting performance expectancy and intentions to use ChatGPT: using SmartPLS to advance an information technology acceptance framework. Technological Forecasting and Social Change 201: 123247.

Chi

Gursoy

Chi

(2022) Tourists’ attitudes toward the use of artificially intelligent (AI) devices in tourism service delivery: moderating role of service value seeking. Journal of Travel Research 61(1): 170–185.

Choi

Rifon

(2002) Antecedents and consequences of web advertising credibility: a study of consumer response to banner ads. Journal of Interactive Advertising 3(1): 12–24.

Elbaz

Mathew

Maher

, et al. (2021) Motherhood desire as a stimulant for fertility tourism intention. African Journal of Hospitality, Tourism and Leisure 10(1): 69–84.

10.

Elbaz

Salem

Alkathiri

, et al. (2023) Personality traits as a factor for entrepreneurial success: gender differences in tourism in Oman. In: Gender and Entrepreneurship in Tourism. Cheltenham, UK: Edward Elgar Publishing, 144–166.

11.

Elbaz

Salem

Darwish

, et al. (2024) Getting to know ChatGPT: how business students feel, what they think about personal morality, and how their academic outcomes affect Oman's higher education. Computers and Education: Artificial Intelligence 7(2024): 1–10.

12.

Filieri

Acikgoz

Ndou

, et al. (2021) Is TripAdvisor still relevant? The influence of review credibility, review usefulness, and ease of use on consumers’ continuance intention. International Journal of Contemporary Hospitality Management 33(1): 199–223.

13.

Følstad

Taylor

(2021) Investigating the user experience of customer service chatbot interaction: a framework for qualitative analysis of chatbot dialogues. Quality and User Experience 6(1): 6. DOI: 10.1007/s41233-021-00046-5.

14.

Hamid

Albahri

Alwan

, et al. (2021) How smart is e-tourism? A systematic review of smart tourism recommendation system applying data management. Computer Science Review 39: 100337.

15.

Hassani

Silva

(2023) The role of ChatGPT in data science: how AI-assisted conversational interfaces are revolutionizing the field. Big data and cognitive computing 7(2): 62. DOI: 10.3390/bdcc7020062.

16.

Ioannou

Tussyadiah

Miller

(2021) That’s private! Understanding travelers’ privacy concerns and online data disclosure. Journal of Travel Research 60(7): 1510–1526.

17.

Jeong

Shin

(2020) Tourists’ experiences with smart tourism technology at smart destinations and their behavior intentions. Journal of Travel Research 59(8): 1464–1477.

18.

Kakhki

Oguz

Gendron

(2024) Exploring the affordances of chatbots in higher education: a framework for understanding and utilizing ChatGPT. Journal of Information Systems Education 35(3): 284–302.

19.

Kannan

(2024) Revolutionizing the tourism industry through artificial intelligence: a comprehensive review of AI integration, impact on customer experience, operational efficiency, and future trends. International Journal for Multidimensional Research Perspectives 2(2): 01–14.

20.

Kim

Giroux

Lee

(2021) When do you trust AI? The effect of number presentation detail on consumer trust and acceptance of AI recommendations. Psychology and Marketing 38(7): 1140–1155.

21.

Kim

Promsivapallop

(2024b) Investigating the influence of generative AI’s credibility and utility on travel consumer behaviour and recommendations through the lens of personal innovativeness. Current Issues in Tourism. 1–5. DOI: 10.1080/13683500.2024.2364764.

22.

Kim

Shin

, et al. (2024a) Artificial intelligence in hospitality and tourism: insights from industry practices, research literature, and expert opinions. Journal of Hospitality & Tourism Research. doi: 10.1177/10963480241229235.

23.

Kumar

Ashraf

Nadeem

(2024) AI-powered marketing: what, where, and how? International Journal of Information Management 77: 102783.

24.

Lee

(2024) Navigating the generative AI travel landscape: the influence of ChatGPT on the evolution from new users to loyal adopters. International Journal of Contemporary Hospitality Management. doi: 10.1108/IJCHM-11-2023-1767.

25.

Wang

Lei

, et al. (2024) “Are you really sure?” Understanding the effects of human self-confidence calibration in AI-assisted decision making. In: Proceedings of the CHI conference on human factors in computing systems, Honolulu, HI, USA, 11 - 16 May 2024, 1–20. DOI: 10.1145/3613904.36426.

26.

Nazir

Khadim

Ali Asadullah

, et al. (2023) Exploring the influence of artificial intelligence technology on consumer repurchase intention: the mediation and moderation approach. Technology in Society 72: 102190.

27.

Parasuraman

Colby

(2015) An updated and streamlined technology readiness index: tri 2.0. Journal of Service Research 18(1): 59–74.

28.

Ramanathan

Meyyappan

(2019) Twitter text mining for sentiment analysis on people’s feedback about Oman tourism. In: 2019 4th MEC international conference on big data and smart city (ICBDSC), Muscat, Oman, 15-16 January 2019, IEEE, 1–5.

29.

Salem

Elbaz

Al-Alawi

, et al. (2023) Is eco-label hotel engagement the pathway to sustainability practices via entrepreneurial resilience and orientation in Oman? Findings from PLS-SEM and fsQCA. International Journal of Contemporary Hospitality Management 35(2): 717–742.

30.

Shi

Gong

Gursoy

(2021) Antecedents of trust and adoption intention toward artificially intelligent recommendation systems in travel planning: a heuristic–systematic model. Journal of Travel Research 60(8): 1714–1734.

31.

Shin

Kang

(2023) Bridging the gap of bibliometric analysis: the evolution, current state, and future directions of tourism research using ChatGPT. Journal of Hospitality and Tourism Management 57: 40–47.

32.

Wang

(2024) Personalizing guest experience with generative AI in the hotel industry: there’s more to it than meets a Kiwi’s eye. Current Issues in Tourism 1–18. doi: 10.1080/13683500.2023.2300030IF:5.7Q1.

33.

Wang

Tang

(2018) A review of application of affordance theory in information systems. Journal of Service Science and Management 11(1): 56–70.

34.

Law

Lovett

, et al. (2024) Tourist acceptance of ChatGPT in travel services: the mediating role of parasocial interaction. Journal of Travel & Tourism Marketing 41(7): 955–972.

35.

Yhee

Goo

Koo

, et al. (2024) Meme-affordance tourism: the power of imitation and self-presentation. Decision Support Systems 179: 114177.

36.

Zlatanov

Popesku

(2019) Current applications of artificial intelligence in tourism and hospitality. In: Paper presented at Sinteza 2019 - international scientific conference on information technology and data related research, 20 April 2019, Singidunum University, Serbia. DOI: 10.15308/Sinteza-2019-84-90.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.24 MB