AI uses modern technologies such as natural language processing (NLP), chatbots, and machine learning (ML) to personalize and simplify client interactions, resulting in faster and more effective service delivery (Mehrotra, 2019). AI systems can learn about clients' interests, movements, and purchasing patterns by analyzing data. This information enables firms to offer more tailored interactions, recommendations, and promotions, resulting in a more engaging and focused consumer experience (Dwivedi et al., 2021).

AI-equipped chatbots and virtual assistants offer 24/7 assistance for consumer inquiries, assuring invariant accessibility. This availability enhances customer satisfaction and morale by letting customers pursue assistance whenever required. Furthermore, AI systems respond to familiar challenges and often requested queries, alleviating human mechanisms from routine tasks and authorizing them to concentrate on more complicated problems. This guides to quicker problem resolution, shorter wait times, and higher customer satisfaction (Zerilli et al., 2019).  

With its predictive processes, artificial intelligence boosts organizations to anticipate and fulfill customer requirements. AI can assist firms in proactively handling complications, offering suitable products or services, and altering their approach to fulfill customer expectations satisfactorily (Rygielski et al., 2002). Automating redundant operations with AI-powered customer service lowers operating costs and increases efficiency, allowing businesses to allocate resources generously and subsidize in improving customer experiences (Al-Mekhlal et al., 2023). Progressively, AI-powered customer service improves client loyalty by providing individualized experiences, increasing service efficiency, and fostering a deeper understanding of consumer needs. Businesses that include AI in customer service are more likely to build reliable, long-term, and productive client connections (Trawnih et al., 2022).

The link between perceived efficiency and customer happiness in AI-powered customer service is strong. Customers' satisfaction is evaluated by how soon the AI system solves problems and provides relevant assistance (Yeo et al., 2022). Positive interactions with AI systems boost people's opinions of their effectiveness. Perceived efficiency and customer satisfaction are vital to customer loyalty because they foster confidence, meet expectations, and considerably boost brand reputation, an important part of corporate management (Floridi et al., 2021).  A well-designed and optimized AI-powered customer support system may drastically boost customer loyalty and establish long-term relationships with businesses (Gupta, 2021; Fernando et al., 2023). This study investigates how perceived efficiency and customer satisfaction affect the relationship between AI-powered customer service and customer loyalty. Businesses can integrate AI systems with client desires and preferences by specifying a customer-centric design, which leads to higher engagement and long-term loyalty.

In contemporary times, there is an increasing trend of superseding human customer service representatives with conversational software agents, occasionally known as chatbots. These agents are developed to interact with consumers in regular language, frequently using artificial intelligence (AI) approaches (Adam et al., 2021). Chatbots are beneficial to businesses, particularly when they improve customer experience. The capacity to resolve client issues and the relevance of responses are two significant aspects impacting chatbot performance in customer support. These elements frequently produce beneficial results, such as increased customer happiness, chatbot usage, and referrals, and importantly, they can significantly boost product purchases (Nicolescu & Tudorache, 2022).

Customers favor AI-based services because they believe AI has superior problem-solving capabilities to human agents for low-complexity jobs such as answering frequently asked questions or processing simple transactions. On the other hand, customers believe that human customer service is more effective than AI solutions for high-complexity activities like complex problem-solving or emotional support. Despite its efficacy, a lack of consumer trust frequently hinders the mainstream adoption of AI chatbots. Systematic design knowledge is still lacking, which promotes user confidence in chatbots (Yang et al., 2023; Zierau et al., 2020).

AI, with its potential to significantly enhance and improve competitiveness, also brings new challenges. However, the prospect of AI-driven convenience, customization, and human-like interactions improving the value-generation process is a hopeful sign of its potential impact. Nonetheless, disadvantages such as privacy concerns, loss of control, and decreased human connection can lessen perceived value and lead to value co-destruction. Enterprises must carefully integrate AI-powered value co-creation methods to create a customer-centric service ecosystem while mitigating the difficulties (Chaturvedi & Verma, 2023). Al-Adwan and Al-Horani (2019) underscored the critical role of trust in affecting consumer happiness, a factor that significantly influences their intention to repurchase.

Businesses are at a critical point when implementing AI. These technologies can reckon data, yield new insights, automate processes, and navigate strategic objectives. Companies that desire to remain competitive and increase consumer loyalty must urgently embrace these technologies (Patel & Trivedi, 2020). Service quality, both AI-based and human-delivered, has a significant impact on entire service ratings, customer pleasure, and loyalty. On the other hand, consumers often have a poorer perception of AI-based customer service than human referrals. This is mainly due to perceived limitations in AI's problem-solving abilities, response times, and lack of human interaction. Positive feedback is more likely when AI systems incorporate natural voice elements like tone modulation, conversational language, and a pleasant service attitude (Zhao et al., 2022). 

The service retrieval quality and conversational capabilities of AI chatbots have a noteworthy impact on user satisfaction. Furthermore, the quality and effectiveness of AI chatbot services can significantly raise user loyalty (Hsu & Lin, 2023). High-quality chatbot services can significantly boost user pleasure and confidence in AI's future in customer service by enhancing cognitive trust, sensitive trust, and perceived value (Qian et al., 2023). Al-Adwan et al. (2022) discovered that service quality features are important in building consumer trust in Jordanian online environments. Similarly, Al-Adwan et al. (2020) investigated the aspects impacting online trust, happiness, and loyalty in Jordan, providing valuable insights that could potentially transform the strategies of local and international e-commerce companies. 

A comprehensive review of the existing scholastic literature on the intervening roles of customer satisfaction and perceived efficiency in the relationship between AI-powered customer service and customer loyalty finds a considerable research gap. Proactive communication refers to an AI system's capacity to foresee consumer desires and deliver assistance before the client requests it. Apprehending the effects of proactive communication in AI-powered customer service is critical since it can provide valuable understanding for improving client interactions and enhancing customer loyalty. 

Sampling method

Participants in this cross-sectional study were selected using a purposive sampling strategy. Individuals from diverse industries who had recently interacted with AI-powered customer service within the previous four months were incorporated into the target group.

Sample size

In its initial phase, the study comprised 480 participants. However, 102 persons who had not operated AI-powered customer service in the last four months have been dropped from the analysis. As a result, the total sample size was 378 respondents.

Nature of sampling

Purposive sampling was used to contrast users and non-users of AI apps and services in several disciplines, including chatbot services and other customized offerings, to ensure the dataset was diverse. This approach was a perfect fit for the study because it was expressly designed to formulate a complete list of persons who use AI services to aggregate their sentiments and experiences.

Duration of the study

This study accumulated data from August to November 2024 using an online survey.

Variables

We examined several significant characteristics to understand better the relationship between AI-powered customer service and customer loyalty. The independent variable (IV), AI-powered customer service, was assessed based on respondents' encounters and views of AI-powered customer support platforms. Mediating variables (MVs) such as perceived efficiency and customer satisfaction were also included in the study. Perceived Efficiency was calculated by asking respondents how they perceived the effectiveness of AI-powered services in answering questions and completing tasks. Simultaneously, Customer Satisfaction was calculated by their overall satisfaction with the AI interactions. Finally, the dependent variable (DV), Customer Loyalty, was thoroughly studied by looking at respondents' intentions, behaviors, and attitudes toward organizations or brands that deploy AI-powered customer service solutions.

 Data collection

To accumulate data, an online questionnaire consisted of 5-point Likert scale items was distributed to participants, via the Google Forms platform with consideration given to the sensitive nature of the information they provided. This method ensured the anonymity and privacy of the participants' responses by ethical standards.

Hypotheses

Following were the proposed hypotheses for the study:

(H1) AI-powered customer service has no significant result on customer satisfaction.

(H2) AI-powered customer service has no significant result on perceived efficiency.

(H3) Customer satisfaction does not significantly affect customer loyalty.

(H4) Perceived efficiency does not significantly affect customer loyalty.

(H5) The association amongst AI-powered customer service, customer satisfaction, and loyalty holds no significant effect.

(H6) The association amongst AI-powered customer service, perceived efficiency, and customer loyalty holds no significant.

Data analysis                                                     

Data analysis was done using the PLS-SEM technique, facilitating the evaluation of direct and indirect effects in the proposed mediation model. The primary focus of the study was to find the relations amongst AI-powered customer service, perceived efficiency, customer satisfaction, and customer loyalty. Notably, mediation analysis sought to determine whether IVs mediate the link among AI-powered customer service and DV.

Respondent summary

Table 1 displays an overview of respondents categorized by gender and age group, ensuring a representative sample for analysis. The table below illustrates the distribution of responses across the following age groups: 'Under 21,' '21-30,' '31-40,' and '40 years and above,' with gender further separated into 'Female' and 'Male.' Among the female respondents, there are 60 individuals under 21, 34 in the 21–30 age range, 44 in the 31–40 range, and 10 aged 40 or older. For male respondents, the breakdown shows 86 individuals under 21, 60 in the 21–30 age group, 65 in the 31–40 group, and 19 aged 40 and above.

Table-1: Respondent Summary

Structural equation modeling

Table 2 presents an outline of the outer loadings within the structural equation model, which reviews the relationships among latent variables, including Cross-Cultural Adaptation (CCA), Cultural Intelligence (CI), Key Competencies (KC), and Managerial Success (MS). The outer loadings, which reflect the assertiveness and tendency of the associations between each indicator (item) and its corresponding latent variable, are an essential part of structural equation modeling. Nunnally (1994) proposed a benchmark value of 0.6 to test the adequacy of these loadings. Indicators with loadings greater than 0.6 are regarded as robust measurements of their respective latent variables, successfully understanding the underlying components.

CCA items CCA1 through CCA6 have decisive outer loadings ranging from 0.75 to 0.91, denoting their reliability in calculating CCA. Similarly, the CI indicators (CI1–CI5) show significant outer loadings ranging from 0.64 to 0.882, validating their strength as CI measures. The KC indicators (KC1–KC4) also have high outside loadings ranging from 0.672 to 0.869, exhibiting their possibility of apprehending KC as a latent variable. MS markers (MS1–MS5) have substantial outer loadings ranging from 0.598 to 0.85, confirming their standing as dependable indices of MS.

Table 2 provides valuable facts about the potency of the links between the indicators and their latent variables in the structural equation model. Higher outer loadings, more significant than the 0.6 thresholds, indicate the suitability of these indicators in effectively estimating and contributing to the acquaintance of the underlying components.

Table-2: Outer loadings of latent variables

Source: SMART-PLS computation

Assessment of the measurement model

Table 3 depicts significant metrics such as Cronbach's Alpha, Composite Reliability, and Average Variance Extracted (AVE) linked to the reliability and validity of the constructs. These indicators provide insights into constructs' internal consistency and convergent validity. The table indicates that Cronbach's Alpha values for all constructs are above the required level of 0.7. Thus, it denotes considerable internal consistency, implying that the items used to calculate these constructs are dependable and consistent. Furthermore, the Composite Reliability values for individual constructs are consistently high, illustrating the constructs' reliability. All AVE values extend the urged threshold of 0.5 (Fornell & Larcker, 1981), indicating that the underlying constructs account for the significance of the discovered variables. These findings demonstrate that the items effectively assess their individual constructs, furnishing evidence of convergent validity.

Table-3: Descriptive statistics, reliability and validity assessment

Source: SMART-PLS computation

Discriminant validity determines whether constructs are distinct and measurable. It affirms that each construct is unique and has no correlation with others. Based on the suggestions of Henseler et al. (2015), this study examined discriminant validity using heterotrait-monotrait (HTMT) ratios, with a threshold value of 0.85, as displayed in Table 4.

Table-4: Heterotrait-monotrait ratio

Structural model and hypothesis testing

The structural model was comprehensively assessed by exploring standardized path coefficients (β values), t-statistics, and corresponding p-values. This means that standardized regression coefficients provide functional insights into the significant relationships between independent and dependent variables (DVs) (Hair et al., 2021). Table 5 shows the results, which strongly support the validity of all hypotheses.

The path from AI service to CS has a substantial path coefficient of 0.92, an exceptionally high t-statistic of 59.13, and a p-value of 0.00, signifying they are associated. Likewise, the path from AI-powered customer service to Perceived Efficiency exhibits a strong path coefficient of 0.94, supported by a strikingly high t-statistic of 85.60 and a p-value of 0.00, indicating that AI service positively affects PE.

The linkage between CS and CL is also significant, with a path coefficient of 1.07, a t-statistic of 9.45, and a p-value of 0.00, indicating that CS has a strong positive impact on CL. Finally, the path from PE to CL has a path coefficient of 0.23, supported by a t-statistic of 3.00 and a p-value of 0.00, implying a statistically influential effect of Perceived Efficiency on Customer Loyalty. As a result, the null hypotheses H1, H2, H3, and H4 are rejected, as illustrated in Figure 2.

                       Figure-2: Structured model                                          Source: SMART-PLS computation

Mediation indirect effect

The mediation effect from AI service to CS to CL is particularly noteworthy. It is sustained by a remarkably high t-statistic of 9.45 and a p-value of 0.00, confirming the presence of a significant mediation effect. This result indicates that the impact of AI service on CL is mediated, at least partially, by CS. In simpler terms, AI service substantially indirectly affects Customer Loyalty by enhancing Customer Satisfaction. Similarly, the mediation effect for the path from AI service to PE to CL also shows statistical significance, with a t-statistic of 3.00 and a p-value of 0.00. These findings have significant practical implications, proving that AI service influences CL through its influence on PE. In essence, AI service positively affects Customer Loyalty indirectly by improving Perceived Efficiency, a result that can guide strategic decisions in customer service management.

The results suggest that both proposed mediation effects are statistically significant within the structural model. The high t-statistics and associated p-values validate the existence of these mediating pathways, highlighting the key roles of Customer Satisfaction and Perceived Efficiency in transmitting the influence of AI-powered customer Service to Customer Loyalty. Hence, the null hypotheses H5 and H6 are rejected, as shown in Table 6.

Table-6: Mediation effect hypothesis testing

Coefficient of determination

The R² values provided for the dependent variables (DVs) in the analysis offer valuable insights into the explanatory power of the structural model. The R² value of 0.855 for Customer Loyalty advises that the independent variables (IVs) in the model report for 85.5% of its variability. The model's high R² value reveals its proficiency in predicting and explaining variances in customer loyalty, emphasizing the implication of AI service, CS, and PE in driving this essential outcome. The R² value of 0.847 for CS points to the model's explanatory variables, depicting 84.7% of the variance, verifying its ability to represent the elemental drivers of Customer Satisfaction. The model explains roughly 87.8% of the variability in Perceived Efficiency, as indicated by an R² value of 0.878.

The R² values confirm the structural model's capacity to depict the elaborate relationships and factors affecting customer loyalty, satisfaction, and perceived efficiency, as shown in Table 7.

                                                           Table-7: Coefficient of determination

This study endeavored to examine the impact of AI services on CL, especially underlining the mediating roles of CS and PE. The findings shed light on the intricate exchanges between these pivotal variables. The first two hypotheses predicted a strong positive association between AI service and CS and PE. The outcomes firmly support both hypotheses, with path coefficients of 0.92 and 0.94 and unusually high t-statistics of 59.13 and 85.61, respectively. These findings indicate that artificial intelligence-powered customer service has an influential mark on customer satisfaction and perceived efficiency. As a result, firms financing AI-powered customer service should anticipate notable advancements in satisfaction and efficiency.

The third and fourth hypotheses anticipate a positive affinity between CS and CL, as well as PE and CL. The results firmly support both deductions. A path coefficient of 1.07 for the relationship between CS and CL, backed by a t-statistic of 9.45, illustrates the considerable influence of customer satisfaction on customer loyalty. Similarly, the path coefficient of 0.23 for the linkage between PE and CL, concurrently with a t-statistic of 3.00, indicates the implication of perceived efficiency in fostering customer loyalty. These findings show that satisfied and well-served customers are more likely to exhibit brand loyalty.

In addition, our study found strong mediation effects. The mediation paths from AI service to CS to CL and AI service to PE to CL revealed statistically significant results. These data demonstrate that AI-powered customer service has both a direct and indirect impact on CS and PE. This emphasizes AI's diverse role in promoting CL.

The DVs had significant R² values of 0.855 for Customer Loyalty, 0.847 for Customer Satisfaction, and 0.878 for Perceived Efficiency. These results imply that the model's independent variables account for a significant part of the variance in these variables. This highlights the model's effectiveness in exemplifying and forecasting customer loyalty, satisfaction and perceived efficiency in AI-powered customer service.

This study focuses on the integral mediating functions that PE and CS conform to in the relationship between AI service and CL. An exhaustive computation of these interconnected variables reveals that AI-powered customer service can significantly impact consumer perceptions and behaviors. Perceived efficiency, the most vital driver, affects customers' perceptions of the AI system's capacity to meet their needs instantly and precisely. The system's evident speed, efficiency, and availability contribute to the overall insight of its efficacy. This perception of efficiency has an unanticipated effect on customer satisfaction, impacting customers' perceptions of the service, brand, and overall experience. Customer satisfaction is firmly bonded to loyalty, which is vital for developing long-term customer relationships. Customers are more likely to form strong emotional ties with a brand if they are pleased with the interactions, services, and personalized experiences provided by the AI. These emotional connections lead to enhanced loyalty, increased possibility of recurrence patronage, positive word-of-mouth, and a higher propensity to promote the brand to others.

Understanding the mediating features of PE and CS is imperative for firms seeking to improve their AI-powered customer service strategy. Organizations may create loyalty and achieve long-term success and growth in today's competitive economy by enhancing perceived efficiency and providing incredible client experiences. As AI advances and becomes more incorporated into customer service policies, further research is mandated to investigate the underlying mechanisms. Such research will provide vital facts that firms can utilize to upgrade their strategy and capitalize on AI's ability to foster long-term customer loyalty.

Limitations and future scope

This study delivered meaningful insights into the mediating roles of PE and CS in the relationship between AI service and CL. Nonetheless, certain limitations that may influence these findings' interpretation and general applicability must be admitted. First, the sample size may not correctly apprehend the different demographic and market segments, restricting the generalizability of the results. The study's emphasis on a single background, industry, or geographical region limits its pertinence to other sectors or regions with considerable variations in customer behavior and sentiments. Likewise, the study depends on subjective metrics like self-reported surveys to assess perceived efficiency and customer satisfaction, which are inclined to response inclination and personal interpretations.

Another area for development is the oversight to justify the dissimilarities in AI maturity and integration across various organizations. The effectiveness and impact of AI-powered customer service differ greatly depending on how progressive the AI technologies are within each organization. Furthermore, external factors such as economic conditions or industry competition were not considered, despite their potential impact on consumer loyalty.

Regardless of such drawbacks, this study provides an adequate basis for future research on customer service, AI technology, and consumer behavior. Prospective studies could include longitudinal research to examine the long-term relationship between AI service, CS, PE, and CL. Such research would provide further insight into the temporal dynamics of these factors. Furthermore, the study might look at the consequence of AI on customer loyalty across industries, as customer expectations and needs vary greatly. Comparative research could look into how the perceived effectiveness of AI services affects CS and loyalty in businesses with diverse levels of AI use. Another probable approach is to investigate how consumer education regarding AI's capabilities and limitations influences satisfaction and loyalty, which could provide valuable insights for developing customer education initiatives.

Future research could also look into the function of AI in emerging domains such as the Internet of Things (IoT), virtual reality (VR), and augmented reality (AR), as well as how these technologies impact customer experiences and loyalty. By exploring these potential approaches, academics can better understand the dynamics driving AI-powered customer service and recommend that firms optimize their AI strategies to elevate long-term customer connections and attain sustainable success.

Strategic Implications

The strategic implications of this study enclose critical advice for organizations and governments on how to use AI technologies adequately to increase consumer satisfaction and loyalty. Firms should have a comprehensive plan for integrating AI-powered customer service. This practice must ensure that AI systems reach properly corporate goals and operational workflows. Designating industry standards for reckoning the perceived effectiveness of AI-powered customer service is critical. Firms can estimate the success of their AI systems by developing uniform standards and Key Performance Indicators (KPIs), assuring constant performance and higher customer service quality. Firms can launch customer education drives to help customers comprehend the benefits and drawbacks of AI in customer service. Educating customers about AI-powered interactions can aid them in setting appropriate expectations, lowering misconceptions, and raising satisfaction. Reinforcing legislation and governing data privacy and security are critical to securing client data in AI-powered customer service. In strengthening customer trust, businesses must follow privacy regulations and levy substantial consequences for violators. Governments might provide incentives, subsidies, or tax breaks for companies that concentrate on evolving AI solutions to improve consumer satisfaction and loyalty. These endeavors can foster organizations to invest in customer-centric AI technologies. Encourage research and innovation in AI technology to improve customer service capabilities. Funding in AI research can lead to the creation of more effective and optimized customer interactions. Companies should integrate systems for gathering consumer feedback into AI-powered customer support platforms. Using consumer insights enhances system performance, customization, and overall customer pleasure.

Enterprises and governments can maximize the benefits of AI services by implementing these policy implications, resulting in greater operational effectiveness, higher CS, and long-term CL. Balancing technology improvements with ethical considerations and maintaining a strong customer-focused strategy will be required to achieve success.

Acknowledgements

Not applicable

Author contributions

A: Written the original draft, Ideation of Methodology & literature review, B: Conceptualisation, Supervision. C: Design the questionnaire and revisions. D. Data analysis, E. Data Collection, F. Visualization & Mapping.

All the authors read and approved the final manuscript.

Funding

Not applicable

Availability of data and materials

The data that support the results during the current study are of primary nature obtained by online survey using structured questionnaire. A total of 285 working professionals participated in the survey. Due to its privacy and confidentiality, it cannot be shared in the public domain.

Declarations

This paper is the authors’ original work, which has not been previously published elsewhere. The paper is not currently being considered for publication elsewhere.

Consent for publication

 Not applicable

Competing interests

The authors declare that they have no competing interests.

1. Adam, M., Wessel, M., & Benlian, A. (2021). AI-based chatbots in customer service and their effects on user compli- ance. Electronic Markets, 31(2), 1–15. https://doi.org/10.1007/s12525-020-00414-7
2. Al-Adwan, A. S., Al-Debei, M. M., & Dwivedi, Y. K. (2022). E-commerce in high uncertainty avoidance cultures: The driving forces of repurchase and word-of-mouth intentions. Technology in Society, 71, 102083. https://doi. org/10.1016/j.techsoc.2022.102083
3. Al-Adwan, A. S., & Al-Horani, M. A. (2019). Boosting customer e-loyalty: An extended scale of online service quality.
4. Information, 10(12), 380. https://doi.org/10.3390/info10120380
5. Al-Adwan, A. S., Kokash, H., Adwan, A. A., Alhorani, A., & Yaseen, H. (2020). Building customer loyalty in online shop- ping: The role of online trust, online satisfaction and electronic word of mouth. International Journal of Electronic Marketing and Retailing, 11(3), 278–306. https://doi.org/10.1504/IJEMR.2020.108132
6. Al-Mekhlal, M., Al-Buraik, M., & Al-Lubli, M. (2023 Digital transformation: AI-powered bot solutions and automation for customer services [Paper presentation]. 2023 International Conference on Digital Applications, Transformation & Economy (ICDATE), Malaysia (pp. 1–7). IEEE. https://doi.org/10.1109/ICDATE58146.2023.10248458
7. Chaturvedi, R., & Verma, S. (2023). Opportunities and Challenges of AI-Driven Customer Service. In J. N. Sheth, V. Jain,
8. Mogaji, & A. Ambika. (Eds.), Artificial Intelligence in Customer Service. Palgrave Macmillan. https://doi.org/10.1007/ 978-3-031-33898-4_3
9. Dwivedi, Y. K., Hughes, L., Ismagilova, E., Aarts, G., Coombs, C., Crick, T., Duan, Y., Dwivedi, R., Edwards, J., Eirug, A., Galanos, V., Ilavarasan, P. V., Janssen, M., Jones, P., Kar, A. K., Kizgin, H., Kronemann, B., Lal, B., Lucini, B., … Williams,
10. D. (2021). Artificial Intelligence (AI): Multidisciplinary perspectives on emerging challenges, opportunities, and agenda for research, practice and policy. International Journal of Information Management, 57, 101994. https://doi. org/10.1016/j.ijinfomgt.2019.08.002
11. Fernando, E., Sutomo, R., Prabowo, Y. D., Gatc, J., & Winanti, W. (2023). Exploring customer relationship management: Trends, challenges, and innovations. Journal of Information Systems and Informatics, 5(3), 984–1001. https://doi. org/10.51519/journalisi.v5i3.541
12. Floridi, L., Cowls, J., Beltrametti, M., Chatila, R., Chazerand, P., Dignum, V., Luetge, C., Madelin, R., Pagallo, U., Rossi, F., Schafer, B., Valcke, P., & Vayena, E. (2021). An ethical framework for a good AI society: Opportunities, risks, princi- ples, and recommendations. Philosophical studies series (pp. 19–39). Springer International Publishing.
13. Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measure- ment error. Journal of Marketing Research, 18(1), 39–50. https://doi.org/10.2307/3151312
14. George, A. S., & George, A. H. (2023). A review of ChatGPT AI’s impact on several business sectors. Partners Universal International Innovation Journal, 1(1), 9–23. https://doi.org/10.5281/zenodo.7644359
15. Gupta, S. (2021). Impact of artificial intelligence on financial decision making: A qualitative study. Journal of Cardiovascular Disease Research, 12(6), 2130–2137.
16. Hair, J., Hair, J. F., Hult, G. T. M., Ringle, C. M., & Sarstedt, M. (2021). A primer on partial least squares structural equation modeling (PLS-SEM). Sage publications.
17. Henseler, J., Ringle, C. M., & Sarstedt, M. (2015). A new criterion for assessing discriminant validity in variance-based structural equation modeling. Journal of the Academy of Marketing Science, 43(1), 115–135. https://doi.org/10.1007/s11747-014-0403-8
18. Hsu, C. L., & Lin, J. C. C. (2023). Understanding the user satisfaction and loyalty of customer service chatbots. Journal of Retailing and Consumer Services, 71, 103211. https://doi.org/10.1016/j.jretconser.2022.103211
19. Mehrotra, A. (2019 Artificial intelligence in financial services-need to blend automation with human touch [Paper presen- tation]. 2019 International Conference on Automation, Computational and Technology Management (ICACTM), (pp. 342–347). IEEE. https://doi.org/10.1109/ICACTM.2019.8776741
20. Nicolescu, L., & Tudorache, M. T. (2022). Human-computer interaction in customer service: The experience with AI chatbots—A systematic literature review. Electronics, 11(10), 1579. https://doi.org/10.3390/electronics11101579
21. Nunnally, J. (1994). Psychometric theory.
22. Patel, N., & Trivedi, S. (2020). Leveraging predictive modeling, machine learning personalization and NLP customer support and chatbots to increase customer loyalty. EQM ResearchBerg, 3(3), 1–24.
23. Prentice, C., Dominique Lopes, S., & Wang, X. (2020). The impact of artificial intelligence and employee service qual- ity on customer satisfaction and loyalty. Journal of Hospitality Marketing & Management, 29(7), 739–756. https://doi. org/10.1080/19368623.2020.1722304
24. Qian, C., Lu, Y., Gong, Y., & Xiong, J. (2023). Can AI chatbots help retain customers? Impact of AI service quality on customer loyalty. Internet Research, 33. https://doi.org/10.1108/INTR-09-2021-0686
25. Rygielski, C., Wang, J. C., & Yen, D. C. (2002). Data mining techniques for customer relationship management.
26. Technology in Society, 24(4), 483–502. https://doi.org/10.1016/s0160-791x(02)00038-6
27. Song, M., Xing, X., Duan, Y., Cohen, J., & Mou, J. (2022). Will artificial intelligence replace human customer service? The impact of communication quality and privacy risks on adoption intention. Journal of Retailing and Consumer Services, 66, 102900. https://doi.org/10.1016/j.jretconser.2021.102900
28. Trawnih, A., Al-Masaeed, S., Alsoud, M., & Alkufahy, A. (2022). Understanding artificial intelligence experience: A cus- tomer perspective. International Journal of Data and Network Science, 6(4), 1471–1484. https://doi.org/10.5267/j. ijdns.2022.5.004
29. Xu, Y., Shieh, C. H., van Esch, P., & Ling, I. L. (2020). AI customer service: Task complexity, problem-solving ability, and usage intention. Australasian Marketing Journal, 28(4), 189–199. https://doi.org/10.1016/j.ausmj.2020.03.005
30. Yang, B., Sun, Y., & Shen, X. L. (2023). Understanding AI-based customer service resistance: A perspective of defective AI features and tri-dimensional distrusting beliefs. Information Processing & Management, 60(3), 103257. https://doi. org/10.1016/j.ipm.2022.103257
31. Yeo, S. F., Tan, C. L., Kumar, A., Tan, K. H., & Wong, J. K. (2022). Investigating the impact of AI-powered technologies on Instagrammers’ purchase decisions in digitalization era–A study of the fashion and apparel industry. Technological Forecasting and Social Change, 177, 121551. https://doi.org/10.1016/j.techfore.2022.121551
32. Zerilli, J., Knott, A., Maclaurin, J., & Gavaghan, C. (2019). Algorithmic decision-making and the control problem. Minds and Machines, 29(4), 555–578. https://doi.org/10.1007/s11023-019-09513-7
33. Zhao, T., Cui, J., Hu, J., Dai, Y., & Zhou, Y. (2022). Is artificial intelligence customer service satisfactory? Insights based on microblog data and user interviews. Cyberpsychology, Behavior and Social Networking, 25(2), 110–117. https:// doi.org/10.1089/cyber.2021.0155
34. Zierau, N., Hausch, M., Bruhin, O., & Söllner, M. (2020 Towards developing trust-dupporting design features for AI-based vhatbots in vustomer service [Paper presentation]. Forty-First International Conference on Information Systems, India.